Ulysses J. Balis

Viable tumour-derived epithelial cells (circulating tumour cells or CTCs) have been identified in peripheral blood from cancer patients and are probably the origin of intractable metastatic disease. Although extremely rare, CTCs represent a potential alternative to invasive biopsies as a source of tumour tissue for the detection, characterization and monitoring of non-haematologic cancers. The ability to identify, isolate, propagate and molecularly characterize CTC subpopulations could further the discovery of cancer stem cell biomarkers and expand the understanding of the biology of metastasis. Current strategies for isolating CTCs are limited to complex analytic approaches that generate very low yield and purity. Here we describe the development of a unique microfluidic platform (the 'CTC-chip') capable of efficient and selective separation of viable CTCs from peripheral whole blood samples, mediated by the interaction of target CTCs with antibody (EpCAM)-coated microposts under precisely controlled laminar flow conditions, and without requisite pre-labelling or processing of samples. The CTC-chip successfully identified CTCs in the peripheral blood of patients with metastatic lung, prostate, pancreatic, breast and colon cancer in 115 of 116 (99%) samples, with a range of 5-1,281 CTCs per ml and approximately 50% purity. In addition, CTCs were isolated in 7/7 patients with early-stage prostate cancer. Given the high sensitivity and specificity of the CTC-chip, we tested its potential utility in monitoring response to anti-cancer therapy. In a small cohort of patients with metastatic cancer undergoing systemic treatment, temporal changes in CTC numbers correlated reasonably well with the clinical course of disease as measured by standard radiographic methods. Thus, the CTC-chip provides a new and effective tool for accurate identification and measurement of CTCs in patients with cancer. It has broad implications in advancing both cancer biology research and clinical cancer management, including the detection, diagnosis and monitoring of cancer.

Tamoxifen significantly reduces tumor recurrence in certain patients with early-stage estrogen receptor-positive breast cancer, but markers predictive of treatment failure have not been identified. Here, we generated gene expression profiles of hormone receptor-positive primary breast cancers in a set of 60 patients treated with adjuvant tamoxifen monotherapy. An expression signature predictive of disease-free survival was reduced to a two-gene ratio, HOXB13 versus IL17BR, which outperformed existing biomarkers. Ectopic expression of HOXB13 in MCF10A breast epithelial cells enhances motility and invasion in vitro, and its expression is increased in both preinvasive and invasive primary breast cancer. The HOXB13:IL17BR expression ratio may be useful for identifying patients appropriate for alternative therapeutic regimens in early-stage breast cancer.

Heterotypic cell interaction between parenchymal cells and nonparenchymal neighbors has been reported to modulate cell growth, migration, and/or differentiation. In both the developing and adult liver, cell-cell interactions are imperative for coordinated organ function. In vitro, cocultivation of hepatocytes and nonparenchymal cells has been used to preserve and modulate the hepatocyte phenotype. We summarize previous studies in this area as well as recent advances in microfabrication that have allowed for more precise control over cell-cell interactions through 'cellular patterning' or 'micropatterning'. Although the precise mechanisms by which nonparenchymal cells modulate the hepatocyte phenotype remain unelucidated, some new insights on the modes of cell signaling, the extent of cell-cell interaction, and the ratio of cell populations are noted. Proposed clinical applications of hepatocyte cocultures, typically extracorporeal bioartificial liver support systems, are reviewed in the context of these new findings. Continued advances in microfabrication and cell culture will allow further study of the role of cell communication in physiological and pathophysiological processes as well as in the development of functional tissue constructs for medical applications.

The classification of appendiceal mucinous tumors is controversial and terminology used for them inconsistent, particularly when they lack overtly malignant features but are associated with extra-appendiceal spread. We reviewed 107 appendiceal mucinous neoplasms and classified them as low-grade appendiceal mucinous neoplasm (LAMN) (n = 88), mucinous adenocarcinomas (MACAs) (n = 16), or discordant (n = 3) based on architectural and cytologic features. LAMNs were characterized by a villous or flat proliferation of mucinous epithelium with low-grade atypia. Thirty-nine tumors were confined to the appendix, but 49 had extra-appendiceal tumor spread, including 39 with peritoneal tumor characterized by mucin pools harboring low-grade mucinous epithelium, usually dissecting in a hyalinized stroma. Eight of the 16 MACAs lacked destructive invasion of the appendiceal wall and eight showed an infiltrative pattern of invasion. Extra-appendiceal tumor spread was present in 12 MACAs (four peritoneum, seven peritoneum and ovaries; one ovaries only). In MACAs with an infiltrative pattern, peritoneal tumor consisted of glands and single cells in a desmoplastic stroma. The peritoneal tumor in the remaining cases consisted of mucin pools that contained mucinous epithelium with high-grade atypia and, in some cases, increased cellularity compared with that seen in peritoneal spread in cases of LAMN. Three cases were classified as discordant because the appendiceal tumors were LAMNs but the peritoneal tumors were high-grade. Follow-up was available for 49 LAMNs, 15 MACAs, and 2 discordant cases. None of the patients with LAMNs confined to the appendix experienced recurrence (median follow-up 6 years). LAMNs with extra-appendiceal spread were associated with 3-, 5-, and 10-year survival rates of 100%, 86%, and 45%, respectively. Patients with MACA had 3- and 5-year survival rates of 90% and 44%, respectively (p = 0.04). The bulk of peritoneal disease correlated with prognosis among patients with MACA (p = 0.04) and, to a lesser degree, among patients with LAMNs (p = 0.07). We conclude that: 1) appendiceal mucinous neoplasms can be classified as either low-grade mucinous neoplasms or mucinous adenocarcinoma based on architectural and cytologic features; 2) tumors that can be confidently placed in the low-grade group (which requires rigorous pathologic evaluation of the appendix) and are confined to the appendix are clinically benign in our experience to date; 3) low-grade tumors confined to the appendix are morphologically identical to those with extra-appendiceal spread (except for the usual identification of breach of the wall in the latter cases) and the same designation is appropriate for the appendiceal neoplasia in each situation; 4) the long-term outlook for patients with low-grade tumors and peritoneal spread is guarded with just over half dying of disease after 10 years; 5) appendiceal mucinous tumors with destructive invasion of the appendiceal wall, complex epithelial proliferations, or high-grade nuclear atypia generally pursue an aggressive clinical course and should be classified as mucinous adenocarcinomas; 6) peritoneal tumor can be classified as involvement by LAMN or MACA, and this distinction is of prognostic significance; 7) bulky peritoneal tumor worsens prognosis; and 8) LAMNs associated with high-grade peritoneal tumor behave as adenocarcinoma.

Cell−cell interactions are important in embryogenesis, in adult physiology and pathophysiology of many disease processes. Co-cultivation of parenchymal and mesenchymal cells has been widely utilized as a paradigm for the study of cell−cell interactions in vitro. In addition, co-cultures of two cell types provide highly functional tissue constructs for use in therapeutic or investigational applications. The inherent complexity of such co-cultures creates difficulty in characterization of cell−cell interactions and their effects on function. In the present study, we utilize conventional “randomly distributed” co-cultures of primary rat hepatocytes and murine 3T3-J2 fibroblasts to investigate the role of increasing fibroblast density on hepatic function. In addition, we utilize microfabrication techniques to localize both cell populations in patterned configurations on rigid substrates. This technique allowed the isolation of fibroblast number as an independent variable in hepatic function. Notably, homotypic hepatocyte interactions were held constant by utilization of similar hepatocyte patterns in all conditions, and the heterotypic interface (region of contact between cell populations) was also held constant. Co-cultures were probed for synthetic and metabolic markers of liver-specific function. The data suggest that fibroblast number plays a role in modulation of hepatocellular response through homotypic fibroblast interactions. The response to changes in fibroblast number are distinct from those attributed to increased contact between hepatocytes and fibroblasts. This approach will allow further elucidation of the complex interplay between two cell types as they form a functional model tissue in vitro or as they interact in vivo to form a functional organ.

Despite growing information on the clinical behavior of hepatocellular carcinoma, the histologic features associated with survival are not well characterized. Clinical and pathologic data on 425 patients who underwent complete resection for hepatocellular carcinoma were reviewed. Six microscopic features, namely, microvascular invasion, nuclear pleomorphism, mitosis, tumor architecture, growth interface, and tumor necrosis, were examined. Independent predictors of survival were identified and combined into a simple prognostic index. By univariate analysis, microvascular invasion, seen in 51.3% of patients (p <0.001), nuclear grade 3, present in 42% of the cases (p <0.001), and mitosis (p <0.008) were significant predictors of poor survival. Hepatocellular carcinoma with a compact growth pattern had a better prognosis as compared with macrotrabecular (p = 0.014) and acinar (p = 0.051) patterns. By multiple regression analysis, only microvascular invasion (p <0.001) and nuclear grade 3 (p = 0.008) were independent predictors of poor survival. The predictive values of microvascular invasion and nuclear grade allowed the construction of a hepatocellular prognostic index (HPI) whereby HPI = (microvascular invasion status × 0.459) + (nuclear grade × 0.287), with microvascular invasion either absent (0) or present (1) and nuclear grade scored as 1, 2, or 3. Using a cut-off of 0.746 (corresponding to at least nuclear grade 2 with microvascular invasion), two groups could be segregated: fair prognosis (HPI ≤0.746), with a 50% survival of 5.06 years, and poor prognosis (HPI >0.746) with a 50% survival of 2.71 years (p <0.001). HPI was more discriminating than Edmondson grade, with Edmondson II hepatocellular carcinomas dispersed in both fair and poor prognosis groups. Microvascular invasion and nuclear grade 3 emerge as strong prognostic indicators, and their combination provides adequate prognostic stratification. Practically, hepatocellular carcinoma can be stratified in two groups with regard to prognosis: 1) fair prognosis group (nuclear grade 1 with or without microvascular invasion and nuclear grade 2 without microvascular invasion), and 2) poor prognosis (nuclear grade 2 with microvascular invasion and nuclear grade 3 with or without microvascular invasion). The combination of these histologic parameters provides adequate prognostic stratification.

Replacement of liver function using extracorporeal bioartificial systems has been attempted with limited success. The instability of the hepatocyte phenotype in vitro has restricted the useful lifetime of these devices. Co-cultivation of hepatocytes with mesenchymal cells is one method that has been widely utilized to stabilize the liver-specific function of isolated cells; however, co-culture has yet to be successfully incorporated in a bioreactor setting. In this study, we probed heterotypic cell interactions in co-cultures of hepatocytes and 3T3 in order to better understand the cellular microenvironment necessary to induce and stabilize liver-specific functions. Using microfabrication and conventional techniques to control the heterotypic interface, the effects of varying degrees of heterotypic interaction on tissue function (albumin and urea synthesis) were examined. Our data indicated maximal induction of liver-specific functions in cultures with maximal initial heterotypic interaction, and that induction of hepatic functions in hepatocytes was increased in the vicinity of fibroblasts as compared to hepatocytes far from the heterotypic interface. Furthermore, our data suggested that heterotypic cell contact is necessary for induction of these functions. These studies will aid in the formation of design criteria for a co-culture based bioartificial liver, as well as provide a useful tool to study the role of heterotypic and homotypic interactions in liver physiology and pathophysiology.

Oxygen uptake rate (OUR) of hepatocytes is an important parameter for the design of bioartificial liver assist (BAL) devices. Porcine hepatocytes were cultured in a specially constructed measurement chamber with an incorporated mixing system and a Clark polarographic oxygen electrode. Signal noise associated with conventional Clark electrode implementations was circumvented by the combination of real time digital numerical averaging and subsequent finite impulse response (FIR) spectral filtering. Additional software allowed for the automated generation of cellular oxygen consumption coefficients, namely, Vmax and K0.5, adding a high degree of objectivity to parameter determination. Optimization of the above numerical techniques identified a 0.1 Hz/200 data point sample size and a 0.004 Hz cutoff frequency as ideal parameters. Vmax values obtained for porcine hepatocytes during the first two weeks of culture showed a maximal consumption of 0.9 nmole/sec/10(6) cells occurring on Day 4 post seeding, and a gradual decrease to 0.31 nmole/sec/10(6) cells by Day 15. K0.5 values increased from 2 mm Hg on Day 2 to 8 mm Hg by Day 8, with gradual subsequent decrease to 4 mm Hg by Day 15. The Vmax and K0.5 values measured for porcine cells were higher than maximal values for rat hepatocytes (Vmax: 0.43 nmole/sec/10(6) cells, K0.5: 5.6 mmHg) and thus may necessitate significantly altered BAL device design conditions to ensure no oxygen limitations. Finally, these results highlight the need for species specific characterization of cellular function for optimal BAL device implementations.

Electronic medical records, including pathology reports, are often used for research purposes. Currently, there are few programs freely available to remove identifiers while leaving the remainder of the pathology report text intact. Our goal was to produce an open source, Health Insurance Portability and Accountability Act (HIPAA) compliant, deidentification tool tailored for pathology reports. We designed a three-step process for removing potential identifiers. The first step is to look for identifiers known to be associated with the patient, such as name, medical record number, pathology accession number, etc. Next, a series of pattern matches look for predictable patterns likely to represent identifying data; such as dates, accession numbers and addresses as well as patient, institution and physician names. Finally, individual words are compared with a database of proper names and geographic locations. Pathology reports from three institutions were used to design and test the algorithms. The software was improved iteratively on training sets until it exhibited good performance. 1800 new pathology reports were then processed. Each report was reviewed manually before and after deidentification to catalog all identifiers and note those that were not removed.1254 (69.7 %) of 1800 pathology reports contained identifiers in the body of the report. 3439 (98.3%) of 3499 unique identifiers in the test set were removed. Only 19 HIPAA-specified identifiers (mainly consult accession numbers and misspelled names) were missed. Of 41 non-HIPAA identifiers missed, the majority were partial institutional addresses and ages. Outside consultation case reports typically contain numerous identifiers and were the most challenging to deidentify comprehensively. There was variation in performance among reports from the three institutions, highlighting the need for site-specific customization, which is easily accomplished with our tool.We have demonstrated that it is possible to create an open-source deidentification program which performs well on free-text pathology reports.

Data protection and security are critical components of routine pathology practice because laboratories are legally required to securely store and transmit electronic patient data. With increasing connectivity of information systems, laboratory work-stations, and instruments themselves to the Internet, the demand to continuously protect and secure laboratory information can become a daunting task. This review addresses informatics security issues in the pathology laboratory related to passwords, biometric devices, data encryption, internet security, virtual private networks, firewalls, anti-viral software, and emergency security situations, as well as the potential impact that newer technologies such as mobile devices have on the privacy and security of electronic protected health information (ePHI). In the United States, the Health Insurance Portability and Accountability Act (HIPAA) govern the privacy and protection of medical information and health records. The HIPAA security standards final rule mandate administrative, physical, and technical safeguards to ensure the confidentiality, integrity, and security of ePHI. Importantly, security failures often lead to privacy breaches, invoking the HIPAA privacy rule as well. Therefore, this review also highlights key aspects of HIPAA and its impact on the pathology laboratory in the United States.

Considering the large yield of viable cells comparable to human liver, primary porcine hepatocytes offer a valuable resource for constructing a bioartificial liver device. In this study, the ability of cultured primary porcine hepatocytes to detoxify xenobiotics has been examined using various known substrates of cytochrome P450 isoenzymes and UDP-glucuronosyltransferases. Present investigation demonstrated the stability of the isoenzymes responsible for the metabolism of diazepam in native state and stabilization of other isoenzymes, as judged by ethoxycoumarin o-dealkylase (ECOD), ethoxyresorufin o-dealkylase (EROD), benzyloxyresorufin o-dealkylase (BROD), and pentoxyresorufin o-dealkylase (PROD) activities following induction in culture environment, for a period of 8 days. Resorufin O-dealkylase activities were found to be the most unstable and deteriorated within first 5 days in culture. These activities were restored following induction with 3-methylcholanthrene (3-MC) or sodium phenobarbital (PB) to 20-fold of 1 activity for EROD, and 60 and 174% of day 1 activity for PROD and BROD on day 8, respectively. Metabolism of methoxyresorufin was most strikingly increased following induction with 3-MC to approximately 60-fold of day 1 activity, on day 8. UDP-glucuronosyltransferase-dependent glucuronidation of phenol red, however, stayed intact during the course of our study without induction. Our study indicated that porcine hepatocytes in vitro maintain many important liver-specific functions including detoxification (steady state and inducibility).

The incorporation of monolayers of cultured hepatocytes into an extracorporeal perfusion system has become a promising approach for the development of a temporary bioartificial liver (BAL) support system. In this paper we present a numerical investigation of the oxygen tension, shear stress, and pressure drop in a bioreactor for a BAL composed of plasma-perfused chambers containing monolayers of porcine hepatocytes. The chambers consist of microfabricated parallel disks with center-to-edge radial flow. The oxygen uptake rate (OUR), measured in vitro for porcine hepatocytes, was curve-fitted using Michaelis-Menten kinetics for simulation of the oxygen concentration profile. The effect of different parameters that may influence the oxygen transport inside the chambers, such as the plasma flow rate, the chamber height, the initial oxygen tension in the perfused plasma, the OUR, and K(m) was investigated. We found that both the plasma flow rate and the initial oxygen tension may have an important effect upon oxygen transport. Increasing the flow rate and/or the inlet oxygen tension resulted in improved oxygen transport to cells in the radial-flow microchannels, and allowed significantly greater diameter reactor without oxygen limitation to the hepatocytes. In the range investigated in this paper (10 microns < H < 100 microns), and for a constant plasma flow rate, the chamber height, H, had a negligible effect on the oxygen transport to hepatocytes. On the contrary, it strongly affected the mechanical stress on the cells that is also crucial for the successful design of the BAL reactors. A twofold decrease in chamber height from 50 to 25 microns produced approximately a fivefold increase in maximal shear stress at the inlet of the reactor from 2 to 10 dyn/cm2. Further decrease in chamber height resulted in shear stress values that are physiologically unrealistic. Therefore, the channel height needs to be carefully chosen in a BAL design to avoid deleterious hydrodynamic effects on hepatocytes.

In this study, we quantified the use of uncertainty expressions, referred to as 'hedge' phrases, among a corpus of 100,000 clinical documents retrieved from our institution's electronic health record system. The frequency of each hedge phrase appearing in the corpus was characterized across document types and clinical departments. We also used a natural language processing tool to identify clinical concepts that were spatially, and potentially semantically, associated with the hedge phrases identified. The objective was to delineate the prevalence of hedge phrase usage in clinical documentation which may have a profound impact on patient care and provider-patient communication, and may become a source of unintended consequences when such documents are made directly accessible to patients via patient portals.

HISTORICALLY, EFFECTIVE CLINICAL UTILIZATION OF IMAGE ANALYSIS AND PATTERN RECOGNITION ALGORITHMS IN PATHOLOGY HAS BEEN HAMPERED BY TWO CRITICAL LIMITATIONS: 1) the availability of digital whole slide imagery data sets and 2) a relative domain knowledge deficit in terms of application of such algorithms, on the part of practicing pathologists. With the advent of the recent and rapid adoption of whole slide imaging solutions, the former limitation has been largely resolved. However, with the expectation that it is unlikely for the general cohort of contemporary pathologists to gain advanced image analysis skills in the short term, the latter problem remains, thus underscoring the need for a class of algorithm that has the concurrent properties of image domain (or organ system) independence and extreme ease of use, without the need for specialized training or expertise.In this report, we present a novel, general case pattern recognition algorithm, Spatially Invariant Vector Quantization (SIVQ), that overcomes the aforementioned knowledge deficit. Fundamentally based on conventional Vector Quantization (VQ) pattern recognition approaches, SIVQ gains its superior performance and essentially zero-training workflow model from its use of ring vectors, which exhibit continuous symmetry, as opposed to square or rectangular vectors, which do not. By use of the stochastic matching properties inherent in continuous symmetry, a single ring vector can exhibit as much as a millionfold improvement in matching possibilities, as opposed to conventional VQ vectors. SIVQ was utilized to demonstrate rapid and highly precise pattern recognition capability in a broad range of gross and microscopic use-case settings.With the performance of SIVQ observed thus far, we find evidence that indeed there exist classes of image analysis/pattern recognition algorithms suitable for deployment in settings where pathologists alone can effectively incorporate their use into clinical workflow, as a turnkey solution. We anticipate that SIVQ, and other related class-independent pattern recognition algorithms, will become part of the overall armamentarium of digital image analysis approaches that are immediately available to practicing pathologists, without the need for the immediate availability of an image analysis expert.

This report presents an overview for pathologists of the development and potential applications of a novel Web enabled system allowing indexing and retrieval of pathology specimens across multiple institutions. The system was developed through the National Cancer Institute's Shared Pathology Informatics Network program with the goal of creating a prototype system to find existing pathology specimens derived from routine surgical and autopsy procedures (“paraffin blocks”) that may be relevant to cancer research. To reach this goal, a number of challenges needed to be met. A central aspect was the development of an informatics system that supported Web-based searching while retaining local control of data. Additional aspects included the development of an eXtensible Markup Language schema, representation of tissue specimen annotation, methods for deidentifying pathology reports, tools for autocoding critical data from these reports using the Unified Medical Language System, and hierarchies of confidentiality and consent that met or exceeded federal requirements. The prototype system supported Web-based querying of millions of pathology reports from 6 participating institutions across the country in a matter of seconds to minutes and the ability of bona fide researchers to identify and potentially to request specific paraffin blocks from the participating institutions. With the addition of associated clinical and outcome information, this system could vastly expand the pool of annotated tissues available for cancer research as well as other diseases.

There is great interest about the possible contribution of ER stress to the apoptosis of pancreatic beta cells in the diabetic state and with islet transplantation.Expression of genes involved in ER stress were examined in beta cell enriched tissue obtained with laser capture microdissection (LCM) from frozen sections of pancreases obtained from non-diabetic subjects at surgery and from human islets transplanted into ICR-SCID mice for 4 wk. Because mice have higher glucose levels than humans, the transplanted beta cells were exposed to mild hyperglycemia and the abnormal environment of the transplant site. RNA was extracted from the LCM specimens, amplified and then subjected to microarray analysis. The transplanted beta cells showed an unfolded protein response (UPR). There was activation of many genes of the IRE-1 pathway that provide protection against the deleterious effects of ER stress, increased expression of ER chaperones and ERAD (ER-associated protein degradation) proteins. The other two arms of ER stress, PERK and ATF-6, had many down regulated genes. Downregulation of EIF2A could protect by inhibiting protein synthesis. Two genes known to contribute to apoptosis, CHOP and JNK, were downregulated.Human beta cells in a transplant site had UPR changes in gene expression that protect against the proapoptotic effects of unfolded proteins.

Fatty liver is a common predisposing risk factor for postoperative liver failure and accounts for most discarded livers during triage of donors. We investigated the effect of heat shock preconditioning (HPc) on recipient survival in a rat fatty liver transplantation model. Fatty liver donor rats were exposed to brief whole-body hyperthermia (10 minutes at 42.5 degrees C) and allowed to recover. HPc induced heat shock proteins (HSPs) (HSP72, HSP90, and heme oxygenase [HO]-1) in donor livers, with levels peaking 12 to 48 hours after HPc. Subsequently, donor livers were harvested 24 hours after HPc, placed in cold storage for 10 hours, and transplanted into normal rats. At 3 hours posttransplantation, HPc reduced serum liver enzymes in the recipients and almost completely suppressed the release of tumor necrosis factor (TNF)-alpha and interleukin (IL)-10. Histologic evaluation 3 and 24 hours after transplantation showed that HPc significantly reduced hepatic inflammation and hepatocellular necrosis without affecting the steatotic appearance of hepatocytes. One week after transplantation, control non-heat-shocked and heat-shocked fatty liver recipients exhibited survival rates of less than 10% and more than 80%, respectively. The evaluation of the survival of recipients receiving fatty livers at different times after HPc showed that the protective effect of HPc was significant when donor livers were transplanted 3 to 48 hours after HPc, with the maximum effect seen 6 to 48 hours after HPc. In conclusion, HPc is a promising avenue to salvage rejected donor fatty livers and enhance the survival rate of fatty liver recipients. We estimate that this technique could increase the annual donor pool by 600 livers.

Hipp, Jason D. MD, PhD*; Lucas, David R. MD*; Emmert-Buck, Michael R. MD, PhD†; Compton, Carolyn C. MD, PhD‡; Balis, Ulysses J. MD* Author Information

Color nonstandardness — the propensity for similar objects to exhibit different color properties across images — poses a significant problem in the computerized analysis of histopathology. Though many papers propose means for improving color constancy, the vast majority assume image formation via reflective light instead of light transmission as in microscopy, and thus are inappropriate for histological analysis. Previously, we presented a novel Bayesian color segmentation algorithm for histological images that is highly robust to color nonstandardness; this algorithm employed the expectation maximization (EM) algorithm to dynamically estimate — for each individual image — the probability density functions that describe the colors of salient objects. However, our approach, like most EM-based algorithms, ignored important spatial constraints, such as those modeled by Markov random field (MRFs). Addressing this deficiency, we now present spatially-constrained EM (SCEM), a novel approach for incorporating Markov priors into the EM framework. With respect to our segmentation system, we replace EM with SCEM and then assess its improved ability to segment nuclei in H&E stained histopathology. Segmentation performance is evaluated over seven (nearly) identical sections of gastrointestinal tissue stained using different protocols (simulating severe color nonstandardness). Over this dataset, our system identifies nuclear regions with an area under the receiver operator characteristic curve (AUC) of 0.838. If we disregard spatial constraints, the AUC drops to 0.748.

Laser capture microdissection (LCM) facilitates procurement of defined cell populations for study in the context of histopathology. The morphologic assessment step in the LCM procedure is time consuming and tedious, thus restricting the utility of the technology for large applications.Here, we describe the use of Spatially Invariant Vector Quantization (SIVQ) for histological analysis and LCM. Using SIVQ, we selected vectors as morphologic predicates that were representative of normal epithelial or cancer cells and then searched for phenotypically similar cells across entire tissue sections. The selected cells were subsequently auto-microdissected and the recovered RNA was analyzed by expression microarray. Gene expression profiles from SIVQ-LCM and standard LCM-derived samples demonstrated highly congruous signatures, confirming the equivalence of the differing microdissection methods.SIVQ-LCM improves the work-flow of microdissection in two significant ways. First, the process is transformative in that it shifts the pathologist's role from technical execution of the entire microdissection to a limited-contact supervisory role, enabling large-scale extraction of tissue by expediting subsequent semi-autonomous identification of target cell populations. Second, this work-flow model provides an opportunity to systematically identify highly constrained cell populations and morphologically consistent regions within tissue sections. Integrating SIVQ with LCM in a single environment provides advanced capabilities for efficient and high-throughput histological-based molecular studies.

In healthcare, patient risk stratification models are often learned using time-series data extracted from electronic health records. When extracting data for a clinical prediction task, several formulations exist, depending on how one chooses the time of prediction and the prediction horizon. In this paper, we show how the formulation can greatly impact both model performance and clinical utility. Leveraging a publicly available ICU dataset, we consider two clinical prediction tasks: in-hospital mortality, and hypokalemia. Through these case studies, we demonstrate the necessity of evaluating models using an outcome-independent reference point, since choosing the time of prediction relative to the event can result in unrealistic performance. Further, an outcome-independent scheme outperforms an outcome-dependent scheme on both tasks (In-Hospital Mortality AUROC .882 vs. .831; Serum Potassium: AUROC .829 vs. .740) when evaluated on test sets that mimic real-world use.

In the past decade molecular diagnostics has changed the clinical management of lung adenocarcinoma patients. Molecular diagnostics, however, is largely dependent on the quantity and quality of the tumor DNA that is retrieved from the tissue or cytology samples. Frequently, patients are diagnosed on cytology specimens where the tumor cells are scattered within the cell block, making selecting for tumor enrichment difficult. In the past we have used laser capture microdissection (LCM) to select for pure populations of tumor cells to increase the sensitivity of molecular assays. This study explores several methods for semiautomated computer-guided LCM.Hematoxylin and eosin- or TTF-1-immunostained slides from a pleural effusion cell block with metastatic lung adenocarcinoma were used for LCM with either AutoScan or a recently described pattern-matching algorithm, spatially invariant vector quantization (SIVQ), to define morphologic predicates (vectors) to select cells of interest.We retrieved pure populations of tumor cells using both algorithm-guided LCM approaches with slight variations in cellular retrievals. Both methods were semiautomated, requiring minimum technical supervision.In this study we demonstrate the first semiautomated, computer-guided LCM of a cytology specimen using SIVQ and AutoScan, a first step towards the long-term goal of integrating LCM into the clinical cytology-molecular workflow.

Pathologists and informaticians are becoming increasingly interested in electronic clinical decision support for pathology, laboratory medicine and clinical diagnosis. Improved decision support may optimize laboratory test selection, improve test result interpretation and permit the extraction of enhanced diagnostic information from existing laboratory data. Nonetheless, the field of pathology decision support is still developing. To facilitate the exchange of ideas and preliminary studies, we convened a symposium entitled: Pathology data integration and clinical decision support.The symposium was held at the Massachusetts General Hospital, on May 10, 2013. Participants were selected to represent diverse backgrounds and interests and were from nine different institutions in eight different states.The day included 16 plenary talks and three panel discussions, together covering four broad areas. Summaries of each presentation are included in this manuscript.A number of recurrent themes emerged from the symposium. Among the most pervasive was the dichotomy between diagnostic data and diagnostic information, including the opportunities that laboratories may have to use electronic systems and algorithms to convert the data they generate into more useful information. Differences between human talents and computer abilities were described; well-designed symbioses between humans and computers may ultimately optimize diagnosis. Another key theme related to the unique needs and challenges in providing decision support for genomics and other emerging diagnostic modalities. Finally, many talks relayed how the barriers to bringing decision support toward reality are primarily personnel, political, infrastructural and administrative challenges rather than technological limitations.

Shared Pathology Informatics Network (SPIN) is a tissue resource initiative that utilizes clinical reports of the vast amount of paraffin-embedded tissues routinely stored by medical centers. SPIN has an informatics component (sending tissue-related queries to multiple institutions via the internet) and a service component (providing histopathologically annotated tissue specimens for medical research). This paper examines if tissue blocks, identified by localized computer searches at participating institutions, can be retrieved in adequate quantity and quality to support medical researchers.Four centers evaluated pathology reports (1990-2005) for common and rare tumors to determine the percentage of cases where suitable tissue blocks with tumor were available. Each site generated a list of 100 common tumor cases (25 cases each of breast adenocarcinoma, colonic adenocarcinoma, lung squamous carcinoma, and prostate adenocarcinoma) and 100 rare tumor cases (25 cases each of adrenal cortical carcinoma, gastro-intestinal stromal tumor [GIST], adenoid cystic carcinoma, and mycosis fungoides) using a combination of Tumor Registry, laboratory information system (LIS) and/or SPIN-related tools. Pathologists identified the slides/blocks with tumor and noted first 3 slides with largest tumor and availability of the corresponding block.Common tumors cases (n = 400), the institutional retrieval rates (all blocks) were 83% (A), 95% (B), 80% (C), and 98% (D). Retrieval rate (tumor blocks) from all centers for common tumors was 73% with mean largest tumor size of 1.49 cm; retrieval (tumor blocks) was highest-lung (84%) and lowest-prostate (54%). Rare tumors cases (n = 400), each institution's retrieval rates (all blocks) were 78% (A), 73% (B), 67% (C), and 84% (D). Retrieval rate (tumor blocks) from all centers for rare tumors was 66% with mean largest tumor size of 1.56 cm; retrieval (tumor blocks) was highest for GIST (72%) and lowest for adenoid cystic carcinoma (58%).Assessment shows availability and quality of archival tissue blocks that are retrievable and associated electronic data that can be of value for researchers. This study serves to compliment the data from which uniform use of the SPIN query tools by all four centers will be measured to assure and highlight the usefulness of archival material for obtaining tumor tissues for research.

In our recently published Editorial in the American Journal of Surgical Pathology entitled “Digital Slide Repositories for Publications – Lessons Learned from the Microarray Community”, we describe a number of potential benefits of a Whole Slide Imaging (WSI) repository for the pathology and biomedical communities, including: 1) an increased level of rigor for peer review of scientific studies submitted for publication, 2) open confirmation and validation of the entire collection of slides from published studies, rather than the contemporary reality where there is only the availability of limited fields of view as selected by authors, 3) the potential that multiple perspectives generated from the review of primary WSI data will identify alternate interpretations and/or conclusions, 4) improved educational opportunities, and 5) availability of reference data sets for inter-observer concordance studies and image analysis algorithm development.[1] While such repositories of scientific data are usually designed by information and computer scientists, we will discuss an alternative solution whereby pathology informaticists, in combination with computer scientists, would design a repository based on the needs and abilities of the end-user in question – in this case, the pathologist. Furthermore, the recent announcement of the closure of the proteomics repository and the Sequencing Read Archive (SRA)[2] highlights the need for the pathology community to sponsor/support such a repository such that it is immune to fluctuations in the federal budget and other external factors associated with the oversight being stewarded by non-pathologists. The purpose of this communication is to alert the pathology community to the need for sponsorship of their own WSI repository and to alert the pathology informatics community of the need for their help in the design and implementation of such a repository.

In this study, we comparatively examined the linguistic properties of narrative clinician notes created through voice dictation versus those directly entered by clinicians via a computer keyboard. Intuitively, the nature of voice-dictated notes would resemble that of natural language, while typed-in notes may demonstrate distinctive language features for reasons such as intensive usage of acronyms. The study analyses were based on an empirical dataset retrieved from our institutional electronic health records system. The dataset contains 30,000 voice-dictated notes and 30,000 notes that were entered manually; both were encounter notes generated in ambulatory care settings. The results suggest that between the narrative clinician notes created via these two different methods, there exists a considerable amount of lexical and distributional differences. Such differences could have a significant impact on the performance of natural language processing tools, necessitating these two different types of documents being differentially treated.

For personalization of medicine, increasingly clinical and demographic data are integrated into nomograms for prognostic use, while molecular biomarkers are being developed to add independent diagnostic, prognostic, or management information. In a number of cases in surgical pathology, morphometric quantitation is already performed manually or semi-quantitatively, with this effort contributing to diagnostic workup. Digital whole slide imaging, coupled with emerging image analysis algorithms, offers great promise as an adjunctive tool for the surgical pathologist in areas of screening, quality assurance, consistency, and quantitation. We have recently reported such an algorithm, SIVQ (Spatially Invariant Vector Quantization), which avails itself of the geometric advantages of ring vectors for pattern matching, and have proposed a number of potential applications. One key test, however, remains the need for demonstration and optimization of SIVQ for discrimination between foreground (neoplasm- malignant epithelium) and background (normal parenchyma, stroma, vessels, inflammatory cells). Especially important is the determination of relative contributions of each key SIVQ matching parameter with respect to the algorithm’s overall detection performance. Herein, by combinatorial testing of SIVQ ring size, sub-ring number, and inter-ring wobble parameters, in the setting of a morphologically complex bladder cancer use case, we ascertain the relative contributions of each of these parameters towards overall detection optimization using urothelial carcinoma as a use case, providing an exemplar by which this algorithm and future histology-oriented pattern matching tools may be validated and subsequently, implemented broadly in other appropriate microscopic classification settings.

Introduction : The advent of digital slides offers new opportunities within the practice of pathology such as the use of image analysis techniques to facilitate computer aided diagnosis (CAD) solutions. Use of CAD holds promise to enable new levels of decision support and allow for additional layers of quality assurance and consistency in rendered diagnoses. However, the development and testing of prostate cancer CAD solutions requires a ground truth map of the cancer to enable the generation of receiver operator characteristic (ROC) curves. This requires a pathologist to annotate, or paint, each of the malignant glands in prostate cancer with an image editor software - a time consuming and exhaustive process. Recently, two CAD algorithms have been described: probabilistic pairwise Markov models (PPMM) and spatially-invariant vector quantization (SIVQ). Briefly, SIVQ operates as a highly sensitive and specific pattern matching algorithm, making it optimal for the identification of any epithelial morphology, whereas PPMM operates as a highly sensitive detector of malignant perturbations in glandular lumenal architecture. Methods : By recapitulating algorithmically how a pathologist reviews prostate tissue sections, we created an algorithmic cascade of PPMM and SIVQ algorithms as previously described by Doyle el al. [1] where PPMM identifies the glands with abnormal lumenal architecture, and this area is then screened by SIVQ to identify the epithelium. Results : The performance of this algorithm cascade was assessed qualitatively (with the use of heatmaps) and quantitatively (with the use of ROC curves) and demonstrates greater performance in the identification of malignant prostatic epithelium. Conclusion : This ability to semi-autonomously paint nearly all the malignant epithelium of prostate cancer has immediate applications to future prostate cancer CAD development as a validated ground truth generator. In addition, such an approach has potential applications as a pre-screening/quality assurance tool.

The development and application of new molecular diagnostic assays based on next-generation sequencing and proteomics require improved methodologies for procurement of target cells from histological sections. Laser microdissection can successfully isolate distinct cells from tissue specimens based on visual selection for many research and clinical applications. However, this can be a daunting task when a large number of cells are required for molecular analysis or when a sizeable number of specimens need to be evaluated.To improve the efficiency of the cellular identification process, we describe a microdissection workflow that leverages recently developed and open source image analysis algorithms referred to as computer-aided laser dissection (CALD). CALD permits a computer algorithm to identify the cells of interest and drive the dissection process.We describe several "use cases" that demonstrate the integration of image analytic tools probabilistic pairwise Markov model, ImageJ, spatially invariant vector quantization (SIVQ), and eSeg onto the ThermoFisher Scientific ArcturusXT and Leica LMD7000 microdissection platforms.The CALD methodology demonstrates the integration of image analysis tools with the microdissection workflow and shows the potential impact to clinical and life science applications.

Background Apparent diffusion coefficient is not specifically sensitive to tumor microstructure and therapy‐induced cellular changes. Purpose To investigate time‐dependent diffusion imaging with the short‐time‐limit random walk with barriers model (STL‐RWBM) for quantifying microstructure parameters and early cancer cellular response to therapy. Study Type Prospective. Population Twenty‐seven patients (median age of 58 years and 7.4% of females) with p16+/p16− oropharyngeal/oral cavity squamous cell carcinomas (OPSCC/OCSCC) underwent MRI scans before therapy, of which 16 patients had second scans at 2 weeks of the 7‐weeks chemoradiation therapy (CRT). Field Strength/Sequence 3‐T, diffusion sequence with oscillating gradient spine echo (OGSE) and pulse gradient spin echo (PGSE). Assessment Diffusion weighted images were acquired using OGSE and PGSE. Effective diffusion times were derived for the STL‐RWBM to estimate free diffusion coefficient D 0 , volume‐to‐surface area ratio of cellular membranes V / S , and cell membrane permeability κ . Mean values of these parameters were calculated in tumor volumes. Statistical Tests Tumor microstructure parameters were compared with clinical stages of p16+ I‐II OPSCC, p16+ III OPSCC, and p16− IV OCSCC by Spearman's rank correlation and with digital pathological analysis of a resected tissue sample. Tumor microstructure parameter responses during CRT in the 16 patients were assessed by paired t ‐tests. A P ‐value of &lt;0.05 was considered statistically significant. Results The derived effective diffusion times affected estimated values of V / S and κ by 40%. The tumor V / S values were significantly correlated with clinical stages ( r = 0.47) as an increase from low to high clinical stages. The in vivo estimated cell size agreed with one from pathological analysis of a tissue sample. Early tumor cellular responses showed a significant increase in D 0 (14%, P = 0.03) and non‐significant increases in κ (56%, P = 0.6) and V / S (10%, P = 0.1). Data Conclusion Effective diffusion time estimation might impact microstructure parameter estimation. The tumor V / S was correlated with OPSCC/OCSCC clinical stages. Level of Evidence 1 Technical Efficacy Stage 1

Spatially invariant vector quantization (SIVQ) is a texture and color-based image matching algorithm that queries the image space through the use of ring vectors. In prior studies, the selection of one or more optimal vectors for a particular feature of interest required a manual process, with the user initially stochastically selecting candidate vectors and subsequently testing them upon other regions of the image to verify the vector's sensitivity and specificity properties (typically by reviewing a resultant heat map). In carrying out the prior efforts, the SIVQ algorithm was noted to exhibit highly scalable computational properties, where each region of analysis can take place independently of others, making a compelling case for the exploration of its deployment on high-throughput computing platforms, with the hypothesis that such an exercise will result in performance gains that scale linearly with increasing processor count.An automated process was developed for the selection of optimal ring vectors to serve as the predicate matching operator in defining histopathological features of interest. Briefly, candidate vectors were generated from every possible coordinate origin within a user-defined vector selection area (VSA) and subsequently compared against user-identified positive and negative "ground truth" regions on the same image. Each vector from the VSA was assessed for its goodness-of-fit to both the positive and negative areas via the use of the receiver operating characteristic (ROC) transfer function, with each assessment resulting in an associated area-under-the-curve (AUC) figure of merit.Use of the above-mentioned automated vector selection process was demonstrated in two cases of use: First, to identify malignant colonic epithelium, and second, to identify soft tissue sarcoma. For both examples, a very satisfactory optimized vector was identified, as defined by the AUC metric. Finally, as an additional effort directed towards attaining high-throughput capability for the SIVQ algorithm, we demonstrated the successful incorporation of it with the MATrix LABoratory (MATLAB™) application interface.The SIVQ algorithm is suitable for automated vector selection settings and high throughput computation.

Introduction: The increasing availability of whole slide imaging (WSI) data sets (digital slides) from glass slides offers new opportunities for the development of computer-aided diagnostic (CAD) algorithms. With the all-digital pathology workflow that these data sets will enable in the near future, literally millions of digital slides will be generated and stored. Consequently, the field in general and pathologists, specifically, will need tools to help extract actionable information from this new and vast collective repository. Methods: To address this limitation, we designed and implemented a tool (dCORE) to enable the systematic capture of image tiles with constrained size and resolution that contain desired histopathologic features. Results: In this communication, we describe a user-friendly tool that will enable pathologists to mine digital slides archives to create image microarrays (IMAs). IMAs are to digital slides as tissue microarrays (TMAs) are to cell blocks. Thus, a single digital slide could be transformed into an array of hundreds to thousands of high quality digital images, with each containing key diagnostic morphologies and appropriate controls. Current manual digital image cut-and-paste methods that allow for the creation of a grid of images (such as an IMA) of matching resolutions are tedious. Conclusion: The ability to create IMAs representing hundreds to thousands of vetted morphologic features has numerous applications in education, proficiency testing, consensus case review, and research. Lastly, in a manner analogous to the way conventional TMA technology has significantly accelerated in situ studies of tissue specimens use of IMAs has similar potential to significantly accelerate CAD algorithm development.

For personalization of medicine, increasingly clinical and demographic data are integrated into nomograms for prognostic use, while molecular biomarkers are being developed to add independent diagnostic, prognostic, or management information. In a number of cases in surgical pathology, morphometric quantitation is already performed manually or semi-quantitatively, with this effort contributing to diagnostic workup. Digital whole slide imaging, coupled with emerging image analysis algorithms, offers great promise as an adjunctive tool for the surgical pathologist in areas of screening, quality assurance, consistency, and quantitation. We have recently reported such an algorithm, SIVQ (Spatially Invariant Vector Quantization), which avails itself of the geometric advantages of ring vectors for pattern matching, and have proposed a number of potential applications. One key test, however, remains the need for demonstration and optimization of SIVQ for discrimination between foreground (neoplasm- malignant epithelium) and background (normal parenchyma, stroma, vessels, inflammatory cells). Especially important is the determination of relative contributions of each key SIVQ matching parameter with respect to the algorithm's overall detection performance. Herein, by combinatorial testing of SIVQ ring size, sub-ring number, and inter-ring wobble parameters, in the setting of a morphologically complex bladder cancer use case, we ascertain the relative contributions of each of these parameters towards overall detection optimization using urothelial carcinoma as a use case, providing an exemplar by which this algorithm and future histology-oriented pattern matching tools may be validated and subsequently, implemented broadly in other appropriate microscopic classification settings.

Isolation of well-preserved pure cell populations is a prerequisite for sound studies of the molecular basis of pancreatic malignancy and other biological phenomena. This chapter reviews current methods for obtaining anatomically specific signals from molecules isolated from tissues, a basic requirement for productive linking of phenotype and genotype. The quality of samples isolated from tissue and used for molecular analysis is often glossed-over or omitted from publications, making interpretation and replication of data difficult or impossible. Fortunately, recently developed techniques allow life scientists to better document and control the quality of samples used for a given assay, creating a foundation for improvement in this area. Tissue processing for molecular studies usually involves some or all of the following steps: tissue collection, gross dissection/identification, fixation, processing/embedding, storage/archiving, sectioning, staining, microdissection/annotation, and pure analyte labeling/identification. High-quality tissue microdissection does not necessarily mean high-quality samples to analyze. The quality of biomaterials obtained for analysis is highly dependent on steps upstream and downstream from tissue microdissection. We provide protocols for each of these steps, and encourage you to improve upon these. It is worth the effort of every laboratory to optimize and document its technique at each stage of the process, and we provide a starting point for those willing to spend the time to optimize. In our view, poor documentation of tissue and cell type of origin and the use of nonoptimized protocols is a source of inefficiency in current life science research. Even incremental improvement in this area will increase productivity significantly.

Background: Last year, our pathology informatics fellowship added informatics-based interactive case studies to its existing educational platform of operational and research rotations, clinical conferences, a common core curriculum with an accompanying didactic course, and national meetings. Methods: The structure of the informatics case studies was based on the traditional business school case study format. Three different formats were used, varying in length from short, 15-minute scenarios to more formal multiple hour-long case studies. Case studies were presented over the course of three retreats (Fall 2011, Winter 2012, and Spring 2012) and involved both local and visiting faculty and fellows. Results: Both faculty and fellows found the case studies and the retreats educational, valuable, and enjoyable. From this positive feedback, we plan to incorporate the retreats in future academic years as an educational component of our fellowship program. Conclusions: Interactive case studies appear to be valuable in teaching several aspects of pathology informatics that are difficult to teach in more traditional venues (rotations and didactic class sessions). Case studies have become an important component of our fellowship's educational platform.

Recently, with the advent of the 7th edition of the AJCC Cancer Staging manual, at least one set of criteria (e.g. breast) were modified to now require the measurement of maximal depth of stromal invasion. With the current manual interpretive morphological approaches typically employed by surgical pathologists to assess tumor extent, the specialty now potentially has stumbled upon a crossroads of practice, where the diagnostic criteria have exceeded the capabilities of our commonly available tools. While whole slide imaging (WSI) technology holds the potential to offer many improvements in clinical workflow over conventional slide microscopy including unambiguous utility for facilitating quantitative diagnostic tasks with one important example being the determination of both linear dimension and surface area. However, the availability of histology data in digital form is of little utility if time-consuming and cumbersome manual workflow steps are necessarily imposed upon the pathologist in order to generate such measurements, especially as encountered with the complex and ill-defined shapes inherent to infiltrative tumors. In this communication, we demonstrate the utility of the recently described SIVQ algorithm to serve as the basis of a highly accurate, precise and semi-automated tool for direct surface area measurement of tumor infiltration from WSI data sets. By anticipating the current trend in cancer staging that emphasizes increasingly precise feature characterization, as witnessed by the recent publication of AJCC's 7th edition of the Cancer Staging Manual, this tool holds promise to will be of value to pathologists for clinical utility.

Recently, with the advent of the 7th edition of the AJCC Cancer Staging manual, at least one set of criteria (e.g. breast) were modified to now require the measurement of maximal depth of stromal invasion. With the current manual interpretive morphological approaches typically employed by surgical pathologists to assess tumor extent, the specialty now potentially has stumbled upon a crossroads of practice, where the diagnostic criteria have exceeded the capabilities of our commonly available tools. While whole slide imaging (WSI) technology holds the potential to offer many improvements in clinical workflow over conventional slide microscopy including unambiguous utility for facilitating quantitative diagnostic tasks with one important example being the determination of both linear dimension and surface area. However, the availability of histology data in digital form is of little utility if time-consuming and cumbersome manual workflow steps are necessarily imposed upon the pathologist in order to generate such measurements, especially as encountered with the complex and ill-defined shapes inherent to infiltrative tumors. In this communication, we demonstrate the utility of the recently described SIVQ algorithm to serve as the basis of a highly accurate, precise and semi-automated tool for direct surface area measurement of tumor infiltration from WSI data sets. By anticipating the current trend in cancer staging that emphasizes increasingly precise feature characterization, as witnessed by the recent publication of AJCC's 7th edition of the Cancer Staging Manual, this tool holds promise to will be of value to pathologists for clinical utility.

The quantitation of B lymphocytes in paraffin-decalcified bone marrow sections is often used as supportive evidence of primary or residual B-cell lymphoproliferative disorders; however, well-defined normal ranges for B lymphocytes are not available. Thirty-four B5-fixed decalcified bone marrow sections were analyzed from a group of patients having an essentially normal bone marrow morphologic examination and lacking any evidence immunologic or hematopoietic disease. All samples were stained with CD20, CD45RA (4KB5), and DBA.44 for B cells, and CD45RO (UCHL-1) for T cells. Image analysis was performed on all cases. CD20-positive cells ranged from 0% to 5.97%. CD45RA positivity ranged from 0% to 5.48%. DBA.44 was positive in 0% to 2.07% of the cells. CD45RO positivity varied from 0% to 6.7%. These results provide useful immunohistochemical B-lymphocyte reference ranges in B5-fixed paraffin bone marrow sections.

In June 2012, the Association of Pathology Informatics (API) convened a 1 day. Strategic Summit conference in Pittsburgh, Pennsylvania, USA, to discuss the future role of laboratory information systems (LISs) as Electronic Health Records (EHRs) are being deployed on an increasingly broad basis across the U.S., some of which include an integrated LIS. Additionally, some of these EHR deployments involve the pursuit of an enterprise-wide solution (EWS) with all of the software components of the EWS provided by a single vendor in an ostensibly unified single software application framework. Such an EWS may suppress consideration of other LIS solutions available in the market, some of which are considered best-of-breed (BoB). BoB, in the context of this communication, is defined as any stand-alone LIS with optimum functionality as compared with the field of equivalent competing systems. An EWS strategy, from the perspective of lab professionals, is only optimal if and when its integrated LIS functionality performs at or beyond the level of performance of established BoB solutions. This selection strategy is necessary because choosing a BoB LIS lowers the overall cost of laboratory tests and attracts higher quality personnel, who relish working in the most modern laboratory environment that is achievable. The Strategic Summit was organized to better understand the effects that the growing popularity of EHRs and the pursuit of the EWS by hospital and healthcare executive will have on the LIS industry, laboratories served by LISs, and laboratory professionals in general. In order to frame this discussion, a series of panelists and faculty were invited to the Summit on a competitive basis to address a predetermined set of questions to an invitation-only audience of pathologists, industry representatives, and laboratory executives. One of the nagging questions that begged addressing in the minds of the faculty and audience was how to define and measure LIS functionality in order to obtain a clearer understanding of the value proposition offered by the BoB LISs available in the market. A clear take-away lesson from the Strategic Summit was that laboratory professionals needed a tool to assess and quantify the functionality of their current LISs in order to call attention to the need for an upgrade of a current LIS or drive its replacement. In addition, such a tool could also be used to assess the functionality of an LIS integrated into the software suite “offered” to a hospital by an EHR vendor. In this latter case, lab professionals might be facing a significant loss of functionality over their currently installed LIS, in the name of system integration at the enterprise level. We put the word offered in quotes here because there is often intense, combined financial and political pressure from the EHR vendor and hospital executive for lab professionals in the hospital to accept an enterprise-wide “primary vendor strategy,” where the LIS is a component of the overall EWS software suite. For the Chief Information Officer (CIO), such a strategy avoids the need for the integration of a “foreign” LIS that then needs to be integrated with the EWS software suite. However, such promises are illusory, because the primary vendor solution may still be based upon disparate architectures that are themselves integrated by the vendor into a more monolithic construct. Experience shows that such integrative exercises carried out by single vendor solutions can be far inferior to locally executed integration. Hence, there is, in fact, no guarantee that a primary vendor will carry out integration in a manner that exceeds the level of integration that is possible with thoughtfully implemented BoB solutions. To achieve a greater understanding of LIS functionality, the API appointed a working group to develop a “toolkit” to help laboratories assess their current LIS functionality as well as serve as a guide to functionality that would be expected by an LIS replacement, sometimes on the basis of an EWS strategy. Further and by corollary, the toolkit could also be used to highlight functionality that would be lost by LIS replacement and requiring the purchase of additional software at additional cost. After nearly a year of effort, the availability of the “toolkit” developed by the API working group was announced at the American Society for Clinical Pathology (ASCP) Annual Meeting in September, 2013, in Chicago. The announcement was made as part of a plenary session presented by the API as part of the API's inaugural participation in the ASCP annual meeting. Whether used to reject a proposal of an inadequate LIS integrated with an EHR by hospital executives or as a justification for the replacement or upgrade of a current LIS, the announcement of the availability of the toolkit (available at no cost on the API web site) has had a significant impact on the practice of pathology informatics in the USA (http://pathologyinformatics.org/toolkit). The LIS Functionality Assessment Toolkit (LIS-FAT) consists of four sections: (1) A narrative overview of the rationale for the development of the toolkit and how to use it; (2) Approximately 850 LIS functionality statements covering most general and specific laboratory units; (3) A set of LIS scenarios that can be used to guide on-site LIS demonstrations; and finally; (4) A work sheet that can help laboratory's assess the total cost of ownership (TCO) of an LIS. TCO is an important calculation in relation to a new LIS because of the additional software purchases that will be required when an immature or inadequate LIS is deployed. Such additional purchases are necessary to achieve BoB functionality and, obviously, will add to the total cost of the system. Not only is it license fees to provide the additional functionality lost, but also the cost for hardware and LIS staff support moving forward. Taken as a whole, the four components of the toolkit can assist lab professionals to select an optimal system but can also assist them in defending against poor system selection by those who may not have an adequate and detailed understanding of laboratory workflow and operations. We believe the LIS-FAT will help laboratories develop a meaningful dialogue with their hospital organizational leadership about the need for fully functional LIS as opposed to those that are a merely” good enough” system. How can a “good enough’ LIS suffice if it lacks a transfusion management module! How can laboratories perform adequately without support for the microbiology laboratory? How will you integrate and manage data across all of the various hospital laboratories when your LIS lacks a common database platform and in tandem requires the use of multiple third party integration engine data transformation nodes? These are problems that lead us back to the earliest days of the LIS development, most of which were already thoroughly addressed by the workflow solutions and software architecture intrinsic in modern BoB LISs. To date, there have been over 1500 visits to LIS-FAT web site. While that may sound low in an Internet world of billions of hits and downloads, in the relatively small world of clinical and anatomic pathology that is LIS, we believe this is a significant amount of activity. Moreover, the availability of LIS-FAT is only now being understood and acted upon. Advertising has been mainly through web sites, e-mail, and word of mouth. Thus, we expect “hits” to continue to grow as more and more laboratorians are faced with the problem of LIS replacement, particularly with enterprise solutions that may significantly compromise laboratory efficiency. Such a latter outcome is totally unacceptable in this era of healthcare reform and pressure to reduce costs in all hospital units. Just as Bruce Schnier, an internationally recognized IT security expert, stated in his recent publication concerning the Snowden-informed vulnerabilities in the very fabric of the Internet, we too as laboratorians entrusted with ensuring the highest standards of clinical laboratory medicine, need to “take back” the LIS from those who threaten the 40 or more years of progress we have experienced and leading to the current state of the art. Inaction is not an option.

The advent of digital slides offers new opportunities within the practice of pathology such as the use of image analysis techniques to facilitate computer aided diagnosis (CAD) solutions. Use of CAD holds promise to enable new levels of decision support and allow for additional layers of quality assurance and consistency in rendered diagnoses. However, the development and testing of prostate cancer CAD solutions requires a ground truth map of the cancer to enable the generation of receiver operator characteristic (ROC) curves. This requires a pathologist to annotate, or paint, each of the malignant glands in prostate cancer with an image editor software - a time consuming and exhaustive process. Recently, two CAD algorithms have been described: probabilistic pairwise Markov models (PPMM) and spatially-invariant vector quantization (SIVQ). Briefly, SIVQ operates as a highly sensitive and specific pattern matching algorithm, making it optimal for the identification of any epithelial morphology, whereas PPMM operates as a highly sensitive detector of malignant perturbations in glandular lumenal architecture.By recapitulating algorithmically how a pathologist reviews prostate tissue sections, we created an algorithmic cascade of PPMM and SIVQ algorithms as previously described by Doyle el al. [1] where PPMM identifies the glands with abnormal lumenal architecture, and this area is then screened by SIVQ to identify the epithelium.The performance of this algorithm cascade was assessed qualitatively (with the use of heatmaps) and quantitatively (with the use of ROC curves) and demonstrates greater performance in the identification of malignant prostatic epithelium.This ability to semi-autonomously paint nearly all the malignant epithelium of prostate cancer has immediate applications to future prostate cancer CAD development as a validated ground truth generator. In addition, such an approach has potential applications as a pre-screening/quality assurance tool.

SIVQ-LCM is a new methodology that automates and streamlines the more traditional, user-dependent laser dissection process. It aims to create an advanced, rapidly customizable laser dissection platform technology. In this report, we describe the integration of the image analysis software Spatially Invariant Vector Quantization (SIVQ) onto the ArcturusXT instrument. The ArcturusXT system contains both an infrared (IR) and ultraviolet (UV) laser, allowing for specific cell or large area dissections. The principal goal is to improve the speed, accuracy, and reproducibility of the laser dissection to increase sample throughput. This novel approach facilitates microdissection of both animal and human tissues in research and clinical workflows.

Cytoskeletal and kinetic epithelial differences between NSAID gastropathy and Helicobacter pylori gastritis: an immunohistochemical determination Aims : Distinguishing histological features between non‐steroidal anti‐inflammatory drug (NSAID) gastropathy and Helicobacter pylori gastritis have been accepted. However, the molecular basis explaining these dissimilar histologies has not been elucidated. In an attempt to clarify this question we investigated the differences in the structural cytoskeleton and proliferative activity of these two gastropathies. Methods and results : We assessed the distribution of five cytokeratins (CK) (CK7, 8, 18, 19 and 20) and Ki67 for the ability to distinguish NSAID from H. pylori gastropathies. In H. pylori gastritis, CK7, 8, 18 and 19 were expressed comparably to normal mucosa from the deep foveolae up to the tips of the glands. The detection of CK20, normally expressed in the upper foveolar region and surface, was decreased with only an epithelial surface reaction. In NSAID gastropathy, CK expression was increased in intensity, with normal distribution for CK8, 18 and 19. Modification of localization was noted for CK7 and 20, with labelling extending toward the deep foveolar region. Unlike H. pylori gastritis, no surface epithelial labelling with Ki67 was noted with NSAID gastropathy but downward elongation of the proliferative zone occurred instead. Conclusions : Contrasting cytostructural alterations and distinct proliferative patterns distinguish NSAID gastropathy from H. pylori gastritis, possibly reflecting different injury pathways.

Standards development in medical imaging originated in the field of radiology and has rapidly spread to numerous other medical specialties. Pathology is well represented in this effort, and it benefits from Interoperability made possible by the utilization of the DICOM 3.0 standard in the fabrication of pathology workstations. Because there currently exists a strong trend for both standards convergence and international harmonization, the formation for a world-wide electronic medical image standard is likely only several years away.

Presentation of CaseA 14-year-old boy with a history of ulcerative colitis was admitted to the hospital because of vomiting and abdominal pain.Ulcerative colitis had been diagnosed at the age of three years by colonoscopic examination after a barium-enema examination had revealed no abnormalities. In subsequent years, progression of the disease was well controlled with the administration of mesalamine and short courses of an adrenocorticosteroid medication for flares; hospital admission was not required. The patient lived with his parents, two siblings, and a healthy dog. There was no family history of gastrointestinal or autoimmune diseases.First AdmissionFourteen months . . .

Telepathology is entering a golden era in which both the technology and the rationale for utilizing this technology are evolving. The following decade will see the introduction of national and international telepathology networks, founded on the principle of improving patient care and increasing cost-effectiveness. As telecommunications and imaging technology concurrently advance, newer generations of telepathology platforms will ultimately deliver performance that will be essentially indistinguishable from that of microscope-derived diagnoses.

Monitoring and control of a bioartificial liver (BAL) support system have the potential to allow for maximization of device bioactivity and protection of both patient and device from untoward consequences of a complex hemoperfused fluidic process, such as coagulation, leakage, or decreased metabolic output. In this work, an integrated embedded systems controller and associated experimental platform were developed to allow for simultaneous monitoring and control of the physical environment of the BAL support system to ensure optimal and sustained hepatic metabolic function and to allow for simplified recording of experimental data. The user interface and core embedded system kernel were developed with rapid prototyping software tools and allowed for operation of easily modified user interface panels. BAL environment monitoring consisted of real-time recording of ambient and reactor temperatures, reactor inlet pressure, the presence of bubbles in the prereactor inlet tubing, and aqueous oxygen tension. Environmental parameters under direct real-time control included reactor inlet flow rate, ambient temperature, and adaptive control of flow rate in response to changes in either inlet pressure or outlet oxygen tension. Use of embedded system integration techniques will facilitate subsequent BAL studies that are dependent on scale-up of reactor size and number, fluidic complexity, and the degree of parallelism such as large animal studies and, ultimately, human clinical studies. In addition, further studies of the effects of flow rate, shear, oxygenation and metabolic substrate on real-time cellular respiration can be pursued with the use of real-time ruthenium oxymetry, as described in this article.

The analysis of digital images using computer algorithms to mine digital data directly is an important and increasingly utilized tool in many different fields of medicine. However, despite the clear value of image analysis to forensics, there is little research and no current practical applications of this technology to forensic pathology. This paper demonstrates the technique of Spatially Invariant Vector Quantization (SIVQ) in the analysis of gross forensic images. SIVQ is a pattern recognition algorithm that can run on a laptop and be utilized by a forensic pathologist with no programming experience. The user merely points and clicks to the feature of interest and SIVQ searches the entire image for the desired feature. The results can be displayed in a variety of formats, both qualitative and quantitative. Gross forensic images from gunshot wounds, burns and patterned contusions were analyzed by a forensic pathologist using SIVQ. Based on features selected by the user, the algorithm was able to detect both the gunshot wound and its marginal abrasion, calculate the surface area of a gunshot wound defect, distinguish between a thermal burn and the surrounding areas of healing, analyze a patterned contusion, and both identify and calculate the density of gunpowder stippling. Even though the work is preliminary, it is obvious from these few examples that SIVQ is easily able to identify features of interest in gross forensic images. Additional work needs to be done to fully explore the potential for this application of technology to the practice of forensic pathology.

The decline of urethral function with advancing age plays a major role in urinary incontinence in women, impairing quality of life and economically burdening the health care system. However, none of the current urinary incontinence treatments address the declining urethral function with aging, and the mechanisms by which aging impacts urethra physiology remain little known or explored. Here, we have compared functional, morphometric, and global gene expression of urethral tissues between young and old female mice. Bladder leak point pressure (LPP) measurement showed that the aged female mice had 26.55% lower LPP compared to younger mice. Vectorized Scale-Invariant Pattern Recognition (VIPR) analysis of the relative abundance of different tissue components revealed that the mid-urethra of old female mice contains less striated muscle, more extracellular matrix/fibrosis, and diminished elastin fibers ratio compared to young mice. Gene expression profiling analysis (bulk RNA-seq of the whole urethra) showed more down-regulated genes in aged than young mice. Immune response and muscle-related (striated and smooth) pathways were predominantly enriched. In contrast, keratinization, skin development, and cell differentiation pathways were significantly downregulated in aged urethral tissues compared to those from young female mice. These results suggest that molecular pathways (i.e., ACVR1/FST signaling and CTGF/TGF-β signaling) leading to a decreased striated muscle mass and an increase in fibrous extracellular matrix in the process of aging deserve further investigation for their roles in the declined urethral function.

Imaging input technology is at its greatest rate of development ever, with CCD technology improving both in terms of image quality and image economy. The current diversity of imager modalities and application specializations allows for elegant matching of device capability to image subject photographic requirements. Both color and low-light imaging technologies continue to evolve, with ever-increasing resolution and dynamic range. The "order of magnitude" cost decreases that so aptly characterized the computer industry of the 1970s and 1980s is now upon the digital imager industry. Although analog cameras, based on video formats, still prevail, the direct digital camera will emerge as the preferred input device within a decade.

SIVQ-LCM is a new methodology that automates and streamlines the more traditional, user-dependent laser dissection process. It aims to create an advanced, rapidly customizable laser dissection platform technology. In this report, we describe the integration of the image analysis software Spatially Invariant Vector Quantization (SIVQ) onto the ArcturusXT instrument. The ArcturusXT system contains both an infrared (IR) and ultraviolet (UV) laser, allowing for specific cell or large area dissections. The principal goal is to improve the speed, accuracy, and reproducibility of the laser dissection to increase sample throughput. This novel approach facilitates microdissection of both animal and human tissues in research and clinical workflows.

Once an image has been captured and stored, whether it is in analog or digital format, for it to be useful in the future there must exist some way of re-rendering it. It is hoped this will be in a manner that is imperceptibly different from the original. Because it is usually impossible to render an image in identical form to the original, numerous decisions need to be made concerning which attributes are the most important, in terms of image archival quality and diagnostic accuracy.

The topic of optics and its effect in ultimate rendered image quality has largely been underemphasized for medical imaging systems. In actuality, adequate attention to optical issues is essential in ensuring that image quality is optimal. Generally, optical issues can be divided into two broad categories: optical interface optimization and lighting optimization. Although these issues deal with disparate aspects of optical physics, they are covered within a single article because they both apply to the common goal of generating superior medical images.

The expanding and diverse opportunities offered by the information technology and biotechnology fields provides for many obvious and perhaps more not-so-obvious compelling opportunities for pathologists and clinical laboratorians. It can be reasonably expected that individuals who choose to pursue such careers outside the conventional opportunities sought out by most pathologists will be rewarded with the satisfaction that comes with the application of established science in new and challenging ways. The realization of the power of the multidisciplinary approach, both within academia and industry, will only serve to further foster opportunities and success stories for those who choose to sail off the edge of the known pathology world.

Background: Image analysis algorithms, coupled with maturing digital whole slide imaging (WSI) technology, holds promise to provide tools for morphometric quantitation in surgical pathology.However, implementation of such strategies will require development and optimization of pattern recognition algorithms adaptable to diseases showing complex architectural features and cytologic atypias.One such example is urothelial carcinoma (UC), of which the aggressive micropapillary variant (MPUC), an aggressive variant of UC which is frequently under-recognized causing diagnostic diffi culties.Herein, we demonstrate the potential of a recently described pattern recognition algorithm and its application to this challenging use case.Methods: We have recently reported SIVQ (Spatially Invariant Vector Quantization), an algorithm that uses ring vector predicates for pattern recognition (Hipp and Cheng, 2010).However, the relative contributions of key SIVQ ring parameters have not been fully characterized.Consequently, we systematically tested SIVQ ring parameters for detection of micropapillary nests in fi elds of a classic case of MPUC by comparing pattern match quality scores at pixels inside and outside of a pathologist-determined "ground truth", using the receiver operating characteristic (ROC) curve.Results: To standardize ring parameter optimization, we tested various ring diameters, number of subrings, and inter-ring rotational "wobble" angles.First,

In healthcare, patient risk stratification models are often learned using time-series data extracted from electronic health records. When extracting data for a clinical prediction task, several formulations exist, depending on how one chooses the time of prediction and the prediction horizon. In this paper, we show how the formulation can greatly impact both model performance and clinical utility. Leveraging a publicly available ICU dataset, we consider two clinical prediction tasks: in-hospital mortality, and hypokalemia. Through these case studies, we demonstrate the necessity of evaluating models using an outcome-independent reference point, since choosing the time of prediction relative to the event can result in unrealistic performance. Further, an outcome-independent scheme outperforms an outcome-dependent scheme on both tasks (In-Hospital Mortality AUROC .882 vs. .831; Serum Potassium: AUROC .829 vs. .740) when evaluated on test sets that mimic real-world use.

DIFFUSE ALVEOLAR

Section I. Non-Neoplastic Disease Airway Disease Alveolar Disease Interstitial Lung Disease Histological Patterns Consistent With Pulmonary Infectious Disease Granulomatosis Pulmonary Vascular Disease Miscellaneous Pulmonary Disease Non-Neoplastic Pleural Disease Section II. Neoplastic Pulmonary and Pleural Disease Lung Tumors Pleural Tumors Index

Abstract Hepatocellular carcinoma (HCC) is heterogeneous in both its gross and microscopic features. Importantly, an increasing number of its characteristics are being demonstrably associated with tumor progression and outcome. Thus, a broad-based understanding of the pathology of HCC is of particular importance to the clinician charged with the management of patients afflicted with this disease.

Abstract In this paper we investigate the possibility of creating an animal model for hypertrophic scar formation by grafting healthy split thickness human skin onto the backs of athymic nude mice and then burning the grafts. We were able to show that the human tissue grafts do remain stable over extended periods and prove that the human dermis does not get infiltrated by mouse tissue. By burning the grafts for ten seconds at 70°C we found that we could not reproduce results previously reported; but by changing the burn parameters we were able to produce regions of dermal disorder that may indicate the formation of hypertrophic scar. These results represent the first replication of positive results using this animal model since it was first proposed 10 years ago.