Clinical impact and frequency of anatomic pathology errors in cancer diagnoses

Evaluation of Turnaround Times of Diagnostic Biopsies: A Metric of Quality in Surgical Pathology

Introduction. Timely and accurate diagnosis of diseases is crucial for effective patient care. Turnaround time (TAT) in surgical pathology, defined as the time between accessioning the sample and reporting results, is a key performance indicator reflecting quality and efficiency. This study explores factors affecting TAT for diagnostic biopsies in a tertiary oncology hospital. Methods. A 1-month pilot study was conducted, focusing on 695 in-house diagnostic biopsies. Biopsies were categorized as routine (requiring only hematoxylin and eosin (H&E) staining) or complex cases (requiring additional tests). TAT was defined as the time between sample accessioning and report availability in the electronic medical record, with delays defined as exceeding 3 days for routine cases and 4 days for complex cases. Survival analysis using Kaplan-Meier plots was utilized to analyze TAT. Results. The overall mean TAT was 3.7 ± 2 days, with routine cases at 3.1 ± 2 days and complex cases at 4.8 ± 2 days (P < 0.001). Survival analysis revealed prolonged TAT for complex cases. Organ-specific analysis highlighted variations in TAT, with brain biopsies presenting the highest complexity and longest TAT. Surprisingly, malignant cases demonstrated slightly shorter TATs compared to benign cases (P = 0.026). Delays were observed in 34% of all cases. Conclusions. Laboratory TAT is crucial and is frequently used as a performance benchmark. We analyzed the various causes of delayed TAT in our hospital's histopathology department, with an emphasis on variables in the analytical phase. The results of this study demonstrate that cases involving ancillary techniques had significantly longer TATs compared to routine H&E cases.

Understanding the errors made by artificial intelligence algorithms in histopathology in terms of patient impact

An increasing number of artificial intelligence (AI) tools are moving towards the clinical realm in histopathology and across medicine. The introduction of such tools will bring several benefits to diagnostic specialities, namely increased diagnostic accuracy and efficiency, however, as no AI tool is infallible, their use will inevitably introduce novel errors. These errors made by AI tools are, most fundamentally, misclassifications made by a computational algorithm. Understanding of how these translate into clinical impact on patients is often lacking, meaning true reporting of AI tool safety is incomplete. In this Perspective we consider AI diagnostic tools in histopathology, which are predominantly assessed in terms of technical performance metrics such as sensitivity, specificity and area under the receiver operating characteristic curve. Although these metrics are essential and allow tool comparison, they alone give an incomplete picture of how an AI tool's errors could impact a patient's diagnosis, management and prognosis. We instead suggest assessing and reporting AI tool errors from a pathological and clinical stance, demonstrating how this is done in studies on human pathologist errors, and giving examples where available from pathology and radiology. Although this seems a significant task, we discuss ways to move towards this approach in terms of study design, guidelines and regulation. This Perspective seeks to initiate broader consideration of the assessment of AI tool errors in histopathology and across diagnostic specialities, in an attempt to keep patient safety at the forefront of AI tool development and facilitate safe clinical deployment.

Grading of Gliomas by Contrast-Enhanced CT Radiomics Features

Background

Gliomas, as Central Nervous System (CNS) tumors, are greatly common with 80% of malignancy. Treatment methods for gliomas, such as surgery, radiation therapy, and chemotherapy depend on the grade, size, location, and the patient's age.

Objective

This study aimed to quantify glioma based on the radiomics analysis and classify its grade into High-grade Glioma (HGG) or Low-grade Glioma (LGG) by various machine-learning methods using contrast-enhanced brain Computerized Tomography (CT) scans.

Material and Methods

This retrospective study involved acquiring and segmenting data, selecting and extracting features, classifying, analyzing, and evaluating classifiers. The study included a total of 62 patients (31 with LGG and 31 with HGG). The tumors were segmented by an experienced CT-scan technologist with 3D slicer software. A total of 14 shape features, 18 histogram-based features, and 75 texture-based features were computed. The Area Under the Curve (AUC) and Receiver Operating Characteristic Curve (ROC) were used to evaluate and compare classification models.

Results

A total of 13 out of 107 features were selected to differentiate between LGGs and HGGs and to perform various classifier algorithms with different cross-validations. The best classifier algorithm was linear-discriminant with 93.5% accuracy, 96.77% sensitivity, 90.3% specificity, and 0.98% AUC in the differentiation of LGGs and HGGs.

Conclusion

The proposed method can identify LGG and HGG with 93.5% accuracy, 96.77% sensitivity, 90.3% specificity, and 0.98% AUC, leading to the best treatment for glioma patients by using CT scans based on radiomics analysis.

An Experimental Platform for Tomographic Reconstruction of Tissue Images in Brightfield Microscopy

(1) Background: Reviewing biological material under the microscope is a demanding and time-consuming process, prone to diagnostic pitfalls. In this study, a methodology for tomographic imaging of tissue sections is presented, relying on the idea that each tissue sample has a finite thickness and, therefore, it is possible to create images at different levels within the sample, revealing details that would probably not be seen otherwise. (2) Methods: Optical slicing was possible by developing a custom-made microscopy stage controlled by an ARDUINO. The custom-made stage, besides the normal sample movements that it should provide along the x-, y-, and z- axes, may additionally rotate the sample around the horizontal axis of the microscope slide. This rotation allows the conversion of the optical microscope into a CT geometry, enabling optical slicing of the sample using projection-based tomographic reconstruction algorithms. (3) Results: The resulting images were of satisfactory quality, but they exhibited some artifacts, which are particularly evident in the axial plane images. (4) Conclusions: Using classical tomographic reconstruction algorithms at limited angles, it is possible to investigate the sample at any desired optical plane, revealing information that would be difficult to identify when focusing only on the conventional 2D images.

Use of nongynecologic cytologic-histologic correlation to identify patterns of error: An institutional experience

BACKGROUND

Quality management practices empower cytology laboratories to deliver consistent, high-quality patient care. Monitoring of key performance indicators is one way by which laboratories can identify patterns of error and focus their improvement activities. Cytologic-histologic correlation (CHC) identifies error by retrospectively reviewing cytology cases when discordant surgical pathology diagnoses are reported. Analysis of CHC data can elucidate patterns of error and direct quality improvement initiatives.

METHODS

CHC data of nongynecologic cytology specimens were reviewed over a 3-year period (2018-2021). Errors were separated by anatomic site and classified as either sampling or interpretive errors.

RESULTS

A total of 364 discordant cases were identified out of 4422 cytologic-histologic pairs (a discordant rate of 8%). The majority (272; 75%) were sampling errors, with fewer interpretive errors (92; 25%). Sampling errors were found to occur most commonly in lower urinary tract and lung. Interpretive errors were most commonly found in lower urinary tract and thyroid.

CONCLUSIONS

Nongynecologic CHC data can be a valuable resource for cytology laboratories. By studying the types of errors, quality improvement activities can be targeted toward problem areas.

Development and validation of a deep learning algorithm for pattern-based classification system of cervical cancer from pathological sections

Background

Multi-center research has demonstrated that adopting Silva's pattern-based classification system (SPBC) enhances the clinical prognosis and facilitates hierarchical management of patients with endocervical adenocarcinomas (EAC). However, inconsistencies in SPBC can arise due to variations in pathologists' experience levels. Thus, the implementation of standardized decision-making tools becomes crucial to enhance the practicality of SPBC in clinical diagnosis and treatment.

Methods

We enrolled a total of 90 patients with EAC in this study, of which 63 were assigned to the training group, and the remaining 27 were allocated to the validation group. To create and validate the prediction models for SPBC, we utilized a deep learning system (DLS) and calculated the area under the receiver operating characteristic curve (AUC).

Results

In Silva pattern classification, ResNet50 achieved an average accuracy of 74.36% (63.64% for pattern A, 55.56% for pattern B, and 89.47% for pattern C respectively). Moreover, in test set, ResNet50 achieved an AUC of 0.69 for pattern A, 0.58 for pattern B, and 0.91 for pattern C.

Conclusions

We successfully established a DLS for SPBC, which holds the potential to aid pathologists in accurately classifying patients with EAC.

The Role of Cue-Based Strategies in Skilled Diagnosis Among Pathologists

OBJECTIVE

This research was designed to test whether behavioral indicators of pathology-related cue utilization were associated with performance on a diagnostic task.

BACKGROUND

Across many domains, including pathology, successful diagnosis depends on pattern recognition that is supported by associations in memory in the form of cues. Previous studies have focused on the specific information or knowledge on which medical image expertise relies. The target in this study is the more general ability to identify and interpret relevant information.

METHOD

Data were collected from 54 histopathologists in both conference and online settings. The participants completed a pathology edition of the Expert Intensive Skills Evaluation 2.0 (EXPERTise 2.0) to establish behavioral indicators of context-related cue utilization. They also completed a separate diagnostic task designed to examine related diagnostic skills.

RESULTS

Behavioral indicators of higher or lower cue utilization were based on the participants' performance across five tasks. Accounting for the number of cases reported per year, higher cue utilization was associated with greater accuracy on the diagnostic task. A post hoc analysis suggested that higher cue utilization may be associated with a greater capacity to recognize low prevalence cases.

CONCLUSION

This study provides support for the role of cue utilization in the development and maintenance of skilled diagnosis amongst pathologists.

APPLICATION

Pathologist training needs to be structured to ensure that learners have the opportunity to form cue-based strategies and associations in memory, especially for less commonly seen diseases.

Three-dimensional tissue volume generation in conventional brightfield microscopy

The purpose of the study is to develop and automate a series of steps for enabling digital 3D tissue volume generation in conventional Brightfield microscopy for histopathology applications. Tissue samples were retrieved from the General Hospital of Athens "Hippocration", Greece. Samples were placed on a microtome that produced consecutive 2 μm sections. Each section was stained using Hematoxylin and Eosin and placed on microscope slides. A histopathologist specified the region of interest (ROI) on each slide. A 2D image was created from each ROI using a LEICA DM2500 microscope with a LEICA DFC 420C camera. Τhe 3D volume was created by stacking consecutive 2D images using a deep learning image interpolation method. The reconstructed 3D tissue volumes were evaluated by an expert histopathologist. Results showed that the 3D volumes might reveal information that is not clearly visible or even undetectable in the conventional 2D Brightfield images. In contrast to other 3D tissue imaging technologies, the proposed method (a) does not depend on the distance of the sample from the objectives producing 3D tissue volumes at any desired magnification, (b) does not require a special instrument, it may be implemented with any conventional Brightfield microscope, and (c) can be used for any given routine application, not only for some specialized clinical studies. The proposed study provides the basis for a feasible, cost-less and time-less upgrade of any standard 2D microscope into a 3D imaging instrument that may enhance the quality of diagnostic assessments in histopathology.

Multi-Class Classification of Breast Cancer Using 6B-Net with Deep Feature Fusion and Selection Method

Breast cancer has now overtaken lung cancer as the world's most commonly diagnosed cancer, with thousands of new cases per year. Early detection and classification of breast cancer are necessary to overcome the death rate. Recently, many deep learning-based studies have been proposed for automatic diagnosis and classification of this deadly disease, using histopathology images. This study proposed a novel solution for multi-class breast cancer classification from histopathology images using deep learning. For this purpose, a novel 6B-Net deep CNN model, with feature fusion and selection mechanism, was developed for multi-class breast cancer classification. For the evaluation of the proposed method, two large, publicly available datasets, namely, BreaKHis, with eight classes containing 7909 images, and a breast cancer histopathology dataset, containing 3771 images of four classes, were used. The proposed method achieves a multi-class average accuracy of 94.20%, with a classification training time of 226 s in four classes of breast cancer, and a multi-class average accuracy of 90.10%, with a classification training time of 147 s in eight classes of breast cancer. The experimental outcomes show that the proposed method achieves the highest multi-class average accuracy for breast cancer classification, and hence, the proposed method can effectively be applied for early detection and classification of breast cancer to assist the pathologists in early and accurate diagnosis of breast cancer.

A deep learning model for molecular label transfer that enables cancer cell identification from histopathology images

Deep-learning classification systems have the potential to improve cancer diagnosis. However, development of these computational approaches so far depends on prior pathological annotations and large training datasets. The manual annotation is low-resolution, time-consuming, highly variable and subject to observer variance. To address this issue, we developed a method, H&E Molecular neural network (HEMnet). HEMnet utilizes immunohistochemistry as an initial molecular label for cancer cells on a H&E image and trains a cancer classifier on the overlapping clinical histopathological images. Using this molecular transfer method, HEMnet successfully generated and labeled 21,939 tumor and 8782 normal tiles from ten whole-slide images for model training. After building the model, HEMnet accurately identified colorectal cancer regions, which achieved 0.84 and 0.73 of ROC AUC values compared to p53 staining and pathological annotations, respectively. Our validation study using histopathology images from TCGA samples accurately estimated tumor purity, which showed a significant correlation (regression coefficient of 0.8) with the estimation based on genomic sequencing data. Thus, HEMnet contributes to addressing two main challenges in cancer deep-learning analysis, namely the need to have a large number of images for training and the dependence on manual labeling by a pathologist. HEMnet also predicts cancer cells at a much higher resolution compared to manual histopathologic evaluation. Overall, our method provides a path towards a fully automated delineation of any type of tumor so long as there is a cancer-oriented molecular stain available for subsequent learning. Software, tutorials and interactive tools are available at: https://github.com/BiomedicalMachineLearning/HEMnet.

Cue utilisation reduces the impact of response bias in histopathology

Histopathologists make diagnostic decisions that are thought to be based on pattern recognition, likely informed by cue-based associations formed in memory, a process known as cue utilisation. Typically, the cases presented to the histopathologist have already been classified as 'abnormal' by clinical examination and/or other diagnostic tests. This results in a high disease prevalence, the potential for 'abnormality priming', and a response bias leading to false positives on normal cases. This study investigated whether higher cue utilisation is associated with a reduction in positive response bias in the diagnostic decisions of histopathologists. Data were collected from eighty-two histopathologists who completed a series of demographic and experience-related questions and the histopathology edition of the Expert Intensive Skills Evaluation 2.0 (EXPERTise 2.0) to establish behavioural indicators of context-related cue utilisation. They also completed a separate, diagnostic task comprising breast histopathology images where the frequency of abnormality was manipulated to create a high disease prevalence context for diagnostic decisions relating to normal tissue. Participants were assigned to higher or lower cue utilisation groups based on their performance on EXPERTise 2.0. When the effects of experience were controlled, higher cue utilisation was specifically associated with a greater accuracy classifying normal images, recording a lower positive response bias. This study suggests that cue utilisation may play a protective role against response biases in histopathology settings.

Analytical and Post Analytical Phase of an ISO 15189:2012 Certified Cytopathology Laboratory-A Five Year Institutional Experience

Objective

Analytical and post analytical phase are integral part of total quality management system and include steps from submission of slides till reports are dispatched. The present study was conducted to analyze the analytical and post analytical phase of the ISO15189:2012 certified cytopathology laboratory. It was also intended to study the various errors which were encountered and steps taken to reduce these discrepancies.

Methods

The study included all documents of quality program from 1 November 2014 till 31 Oct 2019 in medical institute situated in north Himalayan region of India. All the data was recorded and analyzed for analytical and post analytical phase.

Results

The number of samples received in cytopathology lab was 21,566 with total quality errors of 5.19%. Out of these pre-analytical errors were 55%, analytical 10.5% and post analytical errors constituted 34.4%. The maximum errors detected were due to typographical errors followed by delayed turnaround time. Cyto-histopathological discordance was 10.5% in non-gynecological cases and 2.2% in gynecological cases.

Conclusion

Analytical and post analytical phase analysis is essential to minimize the errors and improve the quality of cytopathology lab. Cyto-histopathological correlation is valuable for continuous data tracking in the cytopathology with analytical errors analysis. Maintenance of external, internal quality program, turnaround time with documentation, continuous training and communication with clinician is fundamental for quality improvement in any cytopathology lab. Acknowledgement of nonconformance with root cause analysis and sincere efforts to minimize them is the basic key for successful quality management.

Advancing Research on Medical Image Perception by Strengthening Multidisciplinary Collaboration

Medical image interpretation is central to detecting, diagnosing, and staging cancer and many other disorders. At a time when medical imaging is being transformed by digital technologies and artificial intelligence, understanding the basic perceptual and cognitive processes underlying medical image interpretation is vital for increasing diagnosticians’ accuracy and performance, improving patient outcomes, and reducing diagnostician burn-out. Medical image perception remains substantially understudied. In September of 2019, the National Cancer Institute convened a multidisciplinary panel of radiologists and pathologists together with researchers working in medical image perception and adjacent fields of cognition and perception for the “Cognition and Medical Image Perception Think Tank.” The Think Tank’s key objectives were: to identify critical unsolved problems related to visual perception in pathology and radiology from the perspective of diagnosticians; to discuss how these clinically relevant questions could be addressed through cognitive and perception research; to identify barriers and solutions for transdisciplinary collaborations; to define ways to elevate the profile of cognition and perception research within the medical image community; to determine the greatest needs to advance medical image perception; and to outline future goals and strategies to evaluate progress. The Think Tank emphasized diagnosticians’ perspectives as the crucial starting point for medical image perception research, with diagnosticians describing their interpretation process and identifying perceptual and cognitive problems that arise. This paper reports the deliberations of the Think Tank participants to address these objectives and highlight opportunities to expand research on medical image perception.

In vivo optical imaging-guided targeted sampling for precise diagnosis and molecular pathology

Conventional tissue sampling can lead to misdiagnoses and repeated biopsies. Additionally, tissue processed for histopathology suffers from poor nucleic acid quality and/or quantity for downstream molecular profiling. Targeted micro-sampling of tissue can ensure accurate diagnosis and molecular profiling in the presence of spatial heterogeneity, especially in tumors, and facilitate acquisition of fresh tissue for molecular analysis. In this study, we explored the feasibility of performing 1–2 mm precision biopsies guided by high-resolution reflectance confocal microscopy (RCM) and optical coherence tomography (OCT), and reflective metallic grids for accurate spatial targeting. Accurate sampling was confirmed with either histopathology or molecular profiling through next generation sequencing (NGS) in 9 skin cancers in 7 patients. Imaging-guided 1–2 mm biopsies enabled spatial targeting for in vivo diagnosis, feature correlation and depth assessment, which were confirmed with histopathology. In vivo 1-mm targeted biopsies achieved adequate quantity and high quality of DNA for next-generation sequencing. Subsequent mutational profiling was confirmed on 1 melanoma in situ and 2 invasive melanomas, using a 505-gene mutational panel called Memorial Sloan Kettering-Integrated mutational profiling of actionable cancer targets (MSK-IMPACT). Differential mutational landscapes, in terms of number and types of mutations, were found between invasive and in situ melanomas in a single patient. Our findings demonstrate feasibility of accurate sampling of regions of interest for downstream histopathological diagnoses and molecular pathology in both in vivo and ex vivo settings with broad diagnostic, therapeutic and research potential in cutaneous diseases accessible by RCM-OCT imaging.

Patient-initiated second medical consultations-patient characteristics and motivating factors, impact on care and satisfaction: a systematic review

OBJECTIVES

To review the characteristics and motivations of patients seeking second opinions, and the impact of such opinions on patient management, satisfaction and cost effectiveness.

DATA SOURCES

Embase, Medline, PsycINFO and Health Management Information Consortium (HMIC) databases.

STUDY DESIGN

A systematic literature search was performed for terms related to second opinion and patient characteristics. Study quality was assessed using the National Institutes of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies.

DATA COLLECTION/EXTRACTION METHODS

We included articles focused on patient-initiated second opinions, which provided quantitative data on their impact on diagnosis, treatment, prognosis or patient satisfaction, described the characteristics or motivating factors of patients who initiated a second opinion, or the cost-effectiveness of patient-initiated second opinions.

PRINCIPAL FINDINGS

Thirty-three articles were included in the review. 29 studies considered patient characteristics, 19 patient motivating factors, 10 patient satisfaction and 17 clinical agreement between the first and second opinion. Seeking a second opinion was more common in women, middle-age patients, more educated patients; and in people having a chronic condition, with higher income or socioeconomic status or living in central urban areas. Patients seeking a second opinion sought to gain more information or reassurance about their diagnosis or treatment. While many second opinions confirm the original diagnosis or treatment, discrepancies in opinions had a potential major impact on patient outcomes in up to 58% of cases. No studies reported on the cost effectiveness of patient initiated second opinions.

CONCLUSIONS

This review identified several demographic factors associated with seeking a second opinion, including age, gender, health status, and socioeconomic status. Differences in opinion received, and in the impact of change in opinion, varies significantly between medical specialties. More research is needed to understand the cost effectiveness of second opinions and identify patient groups most likely to benefit from second opinions.

Deep Learning-Based Multi-Class Classification of Breast Digital Pathology Images

Introduction

Breast cancer, one of the most common health threats to females worldwide, has always been a crucial topic in the medical field. With the rapid development of digital pathology, many scholars have used AI-based systems to classify breast cancer pathological images. However, most existing studies only stayed on the binary classification of breast lesions (normal vs tumor or benign vs malignant), far from meeting the clinical demand. Therefore, we established a multi-class classification system of breast digital pathology images based on AI, which is more clinically practical than the binary classification system.

Methods

In this paper, we adopted a two-stage architecture based on deep learning method and machine learning method for the multi-class classification (normal tissue, benign lesion, ductal carcinoma in situ, and invasive carcinoma) of breast digital pathological images.

Results

The proposed approach achieved an overall accuracy of 86.67% at patch-level. At WSI-level, the overall accuracies of our classification system were 88.16% on validation data and 90.43% on test data. Additionally, we used two public datasets, the BreakHis and BACH, for independent verification. The accuracies our model obtained on these two datasets were comparable to related publications. Furthermore, our model could achieve accuracies of 85.19% on multi-classification and 96.30% on binary classification (non-malignant vs malignant) using pathology images of frozen sections, which was proven to have good generalizability. Then, we used t-SNE for visualization of patch classification efficiency. Finally, we analyzed morphological characteristics of patches learned by the model.

Conclusion

The proposed two-stage model could be effectively applied to the multi-class classification task of breast pathology images and could be a very useful tool for assisting pathologists in diagnosing breast cancer.

Prediction of Target-Drug Therapy by Identifying Gene Mutations in Lung Cancer With Histopathological Stained Image and Deep Learning Techniques

Lung cancer is a kind of cancer with high morbidity and mortality which is associated with various gene mutations. Individualized targeted-drug therapy has become the optimized treatment of lung cancer, especially benefit for patients who are not qualified for lung lobectomy. It is crucial to accurately identify mutant genes within tumor region from stained pathological slice. Therefore, we mainly focus on identifying mutant gene of lung cancer by analyzing the pathological images. In this study, we have proposed a method by identifying gene mutations in lung cancer with histopathological stained image and deep learning to predict target-drug therapy, referred to as DeepIMLH. The DeepIMLH algorithm first downloaded 180 hematoxylin-eosin staining (H&E) images of lung cancer from the Cancer Gene Atlas (TCGA). Then deep convolution Gaussian mixture model (DCGMM) was used to perform color normalization. Convolutional neural network (CNN) and residual network (Res-Net) were used to identifying mutated gene from H&E stained imaging and achieved good accuracy. It demonstrated that our method can be used to choose targeted-drug therapy which might be applied to clinical practice. More studies should be conducted though.

Deep Learning Can Differentiate IDH-Mutant from IDH-Wild GBM

Isocitrate dehydrogenase (IDH) mutant and wildtype glioblastoma multiforme (GBM) often show overlapping features on magnetic resonance imaging (MRI), representing a diagnostic challenge. Deep learning showed promising results for IDH identification in mixed low/high grade glioma populations; however, a GBM-specific model is still lacking in the literature. Our aim was to develop a GBM-tailored deep-learning model for IDH prediction by applying convoluted neural networks (CNN) on multiparametric MRI. We selected 100 adult patients with pathologically demonstrated WHO grade IV gliomas and IDH testing. MRI sequences included: MPRAGE, T1, T2, FLAIR, rCBV and ADC. The model consisted of a 4-block 2D CNN, applied to each MRI sequence. Probability of IDH mutation was obtained from the last dense layer of a softmax activation function. Model performance was evaluated in the test cohort considering categorical cross-entropy loss (CCEL) and accuracy. Calculated performance was: rCBV (accuracy 83%, CCEL 0.64), T1 (accuracy 77%, CCEL 1.4), FLAIR (accuracy 77%, CCEL 1.98), T2 (accuracy 67%, CCEL 2.41), MPRAGE (accuracy 66%, CCEL 2.55). Lower performance was achieved on ADC maps. We present a GBM-specific deep-learning model for IDH mutation prediction, with a maximal accuracy of 83% on rCBV maps. Highest predictivity achieved on perfusion images possibly reflects the known link between IDH and neoangiogenesis through the hypoxia inducible factor.

Use of radiomics based on 18F-FDG PET/CT and machine learning methods to aid clinical decision-making in the classification of solitary pulmonary lesions: an innovative approach

PURPOSE

This study was designed and performed to assess the ability of ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET) and computed tomography (CT) radiomics features combined with machine learning methods to differentiate between primary and metastatic lung lesions and to classify histological subtypes. Moreover, we identified the optimal machine learning method.

METHODS

A total of 769 patients pathologically diagnosed with primary or metastatic lung cancers were enrolled. We used the LIFEx package to extract radiological features from semiautomatically segmented PET and CT images within the same volume of interest. Patients were randomly distributed in training and validation sets. Through the evaluation of five feature selection methods and nine classification methods, discriminant models were established. The robustness of the procedure was controlled by tenfold cross-validation. The model's performance was evaluated using the area under the receiver operating characteristic curve (AUC).

RESULTS

Based on the radiomics features extracted from PET and CT images, forty-five discriminative models were established. Combined with appropriate feature selection methods, most classifiers showed excellent discriminative ability with AUCs greater than 0.75. In the differentiation between primary and metastatic lung lesions, the feature selection method gradient boosting decision tree (GBDT) combined with the classifier GBDT achieved the highest classification AUC of 0.983 in the PET dataset. In contrast, the feature selection method eXtreme gradient boosting combined with the classifier random forest (RF) achieved the highest AUC of 0.828 in the CT dataset. In the discrimination between squamous cell carcinoma and adenocarcinoma, the combination of GBDT feature selection method with GBDT classification had the highest AUC of 0.897 in the PET dataset. In contrast, the combination of the GBDT feature selection method with the RF classification had the highest AUC of 0.839 in the CT dataset. Most of the decision tree (DT)-based models were overfitted, suggesting that the classification method was not appropriate for practical application.

CONCLUSION

¹⁸F-FDG PET/CT radiomics features combined with machine learning methods can distinguish between primary and metastatic lung lesions and identify histological subtypes in lung cancer. GBDT and RF were considered optimal classification methods for the PET and CT datasets, respectively, and GBDT was considered the optimal feature selection method in our analysis.

Frozen section in thyroid gland follicular neoplasms: It's high time to abandon it!

Thyroid nodules are a very common clinical condition. The 2015 American Thyroid Association (ATA) guidelines recommend surgical excision for Bethesda IV nodules. The use of intraoperative frozen section (FS) has been recommended as a strategy to tailor the extent of the initial surgery. We critically evaluated the literature that discusses the utility and cost-effectiveness of FS to make an intraoperative decision in patients with thyroid nodules classified as follicular neoplasm. FS should not be recommended as a routine intraoperative test to assess for malignancy in thyroid follicular patterned lesions due to its low performance; the high number of deferred results; the inability to adequately assess histologically defining features; the improvements in risk stratification guiding total thyroidectomy; and the low cost-effectiveness of FS.

Artificial Intelligence-Based Multiclass Classification of Benign or Malignant Mucosal Lesions of the Stomach

Gastric cancer (GC) is one of the leading causes of cancer-related death worldwide. It takes some time from chronic gastritis to develop in GC. Early detection of GC will help patients obtain timely treatment. Understanding disease evolution is crucial for the prevention and treatment of GC. Here, we present a convolutional neural network (CNN)-based system to detect abnormalities in the gastric mucosa. We identified normal mucosa, chronic gastritis, and intestinal-type GC: this is the most common route of gastric carcinogenesis. We integrated digitalizing histopathology of whole-slide images (WSIs), stain normalization, a deep CNN, and a random forest classifier. The staining variability of WSIs was reduced significantly through stain normalization, and saved the cost and time of preparing new slides. Stain normalization improved the effect of the CNN model. The accuracy rate at the patch-level reached 98.4%, and 94.5% for discriminating normal → chronic gastritis → GC. The accuracy rate at the WSIs-level for discriminating normal tissue and cancerous tissue reached 96.0%, which is a state-of-the-art result. Survival analyses indicated that the features extracted from the CNN exerted a significant impact on predicting the survival of cancer patients. Our CNN model disclosed significant potential for adjuvant diagnosis of gastric diseases, especially GC, and usefulness for predicting the prognosis.

Digital pathology for second opinion consultation and donor assessment during organ procurement: Review of the literature and guidance for deployment in transplant practice

Telepathology has been an important application for second opinion consultation ever since the introduction of digital pathology. However, little is known regarding teleconsultation for second opinion in transplantation. There is also limited literature on telepathology during organ donor procurement, typically utilized when general pathologists on-call request back-up to help assess donor biopsies for organ suitability or to diagnose newly discovered tumors with urgent time constraints. In this review, we searched Pubmed/Embase and websites of transplant organizations to collect and analyze published evidence on teleconsultation for donor evaluation and organ procurement. Of 2725 records retrieved using the key terms 'telepathology', 'second opinion' and 'transplantation', 26 suitable studies were included. Most records were from North America and included validation studies of telepathology being used for remote frozen section interpretation of donor biopsies with whole slide imaging. The data from these published studies supports the transition towards digital teleconsultation in transplant settings where consultations among pathologists are still handled by pathologists being called on site, via telephone and/or email.

Raman spectroscopy for rapid intra-operative margin analysis of surgically excised tumour specimens

Raman spectroscopy, a form of vibrational spectroscopy, has the ability to provide sensitive and specific biochemical analysis of tissue. This review article provides an in-depth analysis of the suitability of different Raman spectroscopy techniques in providing intra-operative margin analysis in a range of solid tumour pathologies. Surgical excision remains the primary treatment of a number of solid organ cancers. Incomplete excision of a tumour and positive margins on histopathological analysis is associated with a worse prognosis, the need for adjuvant therapies with significant side effects and a resulting financial burden. The provision of intra-operative margin analysis of surgically excised tumour specimens would be beneficial for a number of pathologies, as there are no widely adopted and accurate methods of margin analysis, beyond histopathology. The limitations of Raman spectroscopic studies to date are discussed and future work necessary to enable translation to clinical use is identified. We conclude that, although there remain a number of challenges in translating current techniques into a clinically effective tool, studies so far demonstrate that Raman Spectroscopy has the attributes to successfully perform highly accurate intra-operative margin analysis in a clinically relevant environment.

The Value Proposition for Pathologists: A Population Health Approach

The transition to a value-based payment system offers pathologists the opportunity to play an increased role in population health by improving outcomes and safety as well as reducing costs. Although laboratory testing itself accounts for a small portion of health-care spending, laboratory data have significant downstream effects in patient management as well as diagnosis. Pathologists currently are heavily engaged in precision medicine, use of laboratory and pathology test results (including autopsy data) to reduce diagnostic errors, and play leading roles in diagnostic management teams. Additionally, pathologists can use aggregate laboratory data to monitor the health of populations and improve health-care outcomes for both individual patients and populations. For the profession to thrive, pathologists will need to focus on extending their roles outside the laboratory beyond the traditional role in the analytic phase of testing. This should include leadership in ensuring correct ordering and interpretation of laboratory testing and leadership in population health programs. Pathologists in training will need to learn key concepts in informatics and data analytics, health-care economics, public health, implementation science, and health systems science. While these changes may reduce reimbursement for the traditional activities of pathologists, new opportunities arise for value creation and new compensation models. This report reviews these opportunities for pathologist leadership in utilization management, precision medicine, reducing diagnostic errors, and improving health-care outcomes.

Patch-level Tumor Classification in Digital Histopathology Images with Domain Adapted Deep Learning

Tumor histopathology is a crucial step in cancer diagnosis which involves visual inspection of imaging data to detect the presence of tumor cells among healthy tissues. This manual process can be time-consuming, error-prone, and influenced by the expertise of the pathologist. Recent deep learning methods for image classification and detection using convolutional neural networks (CNNs) have demonstrated marked improvements in the accuracy of a variety of medical imaging analysis tasks. However, most well-established deep learning methods require large annotated training datasets that are specific to the particular problem domain; such datasets are difficult to acquire for histopathology data where visual characteristics differ between different tissue types, in addition to the need for precise annotations. In this study, we overcome the lack of annotated training dataset in histopathology images of a particular domain by adapting annotated histopathology images from different domains (tissue types). The data from other tissue types are used to pre-train CNNs into a shared histopathology domain (e.g., stains, cellular structures) such that it can be further tuned/optimized for a specific tissue type. We evaluated our classification method on publically available datasets of histopathology images; the accuracy and area under the receiver operating characteristic curve (AUC) of our method was higher than CNNs trained from scratch on limited data (accuracy: 84.3% vs. 78.3%; AUC: 0.918 vs. 0.867), suggesting that domain adaptation can be a valuable approach to histopathological images classification.

Payment Innovations To Improve Diagnostic Accuracy And Reduce Diagnostic Error

Diagnostic accuracy is essential for treatment decisions but is largely unaccounted for by payers, including in fee-for-service Medicare and proposed Alternative Payment Models (APMs). We discuss three payment-related approaches to reducing diagnostic error. First, coding changes in the Medicare Physician Fee Schedule could facilitate the more effective use of teamwork and information technology in the diagnostic process and better support the cognitive work and time commitment that physicians make in the quest for diagnostic accuracy, especially in difficult or uncertain cases. Second, new APMs could be developed to focus on improving diagnostic accuracy in challenging cases and make available support resources for diagnosis, including condition-specific centers of diagnostic expertise or general diagnostic centers of excellence that provide second (or even third) opinions. Performing quality improvement activities that promote safer diagnosis should be a part of the accountability of APM recipients. Third, the accuracy of diagnoses that trigger APM payments and establish payment amounts should be confirmed by APM recipients. Implementation of these multipronged approaches can make current payment models more accountable for addressing diagnostic error and position diagnostic performance as a critical component of quality-based payment.

Curation and analysis of clinical pathology parameters and histopathologic findings from eTOXsys, a large database project (eTOX) for toxicologic studies

Large data sharing projects amongst the pharmaceutical industry have the potential to generate new insights using data on a scale that has not been previously available. A retrospective analysis of the preclinical toxicology data collected as part of the eTOX project was conducted with the aim to provide background rates and treatment-related value analysis on both clinical pathology and histopathology datasets. Incorporated into this analysis was an extensive data consolidation task to standardise all data. Reference intervals for common clinical pathology parameters in rat and dog were generated, alongside background histopathology incidence rates in the liver, heart and kidney. Systematically applied decision thresholds allowed consistent relabelling of data points considered anomalous, and maximum fold change estimates. Relabelling of anomalous data points was conducted for the histopathology data using a Bayesian model to identify dose-dependent increases in pathologies. The results of this study allow: newly generated data to be analysed using the same methodology, rates and distributions to be used when building predictive dose-response models, and the possibility to correlate clinical pathology findings with concurrent histopathology findings. In the first half of this paper we discuss data curation, in the second half we report on the analytical methods and results.

Barriers and facilitators impacting reliability of the electronic health record-facilitated total testing process

OBJECTIVE

Despite ongoing efforts to improve reliability of the total testing process (TTP), breakdowns continue to occur resulting in diagnostic delays and suboptimal patient outcomes. We performed an exploratory study to identify factors that impact TTP reliability in electronic health record (EHR)-enabled care.

MATERIALS AND METHODS

We interviewed experts at three large EHR-enabled health care organizations and identified all TTP steps performed from clinician test ordering to result communication to patients. Findings from all sites were combined to develop a detailed process map of known TTP activities. We additionally asked experts about factors that positively or negatively impacted TTP reliability at each step. We describe the specific TTP steps identified and associated barriers and facilitators to TTP reliability.

RESULTS

We interviewed 39 experts involved in or overseeing the TTP. Most TTP activities identified were similar across sites, but we found significant differences with test order transmission to diagnostic services and relay of results back to clinicians and patients. Twenty-five unique barriers were identified related to technology and EHR usability issues, time and resource constraints, suboptimal clinic workflows, patient-related factors, information access limitations, and insufficient clinician training. Twenty-four unique facilitators were identified related to personnel training, workflow optimization and standardization, helpful EHR features, and improved electronic communication between clinics and diagnostic services.

DISCUSSION

Barriers related to EHR usability and with communication between clinicians and diagnostic services increase TTP vulnerability and should be targeted by future efforts to improve process reliability. Several facilitators identified in the study could inform future strategies and solutions to improve TTP reliability.

A software tool for the quantification of metastatic colony growth dynamics and size distributions in vitro and in vivo

The majority of cancer-related deaths are due to metastasis, hence improved methods to biologically and computationally model metastasis are required. Computational models rely on robust data that is machine-readable. The current methods used to model metastasis in mice involve generating primary tumors by injecting human cells into immune-compromised mice, or by examining genetically engineered mice that are pre-disposed to tumor development and that eventually metastasize. The degree of metastasis can be measured using flow cytometry, bioluminescence imaging, quantitative PCR, and/or by manually counting individual lesions from metastatic tissue sections. The aforementioned methods are time-consuming and do not provide information on size distribution or spatial localization of individual metastatic lesions. In this work, we describe and provide a MATLAB script for an image-processing based method designed to obtain quantitative data from tissue sections comprised of multiple subpopulations of disseminated cells localized at metastatic sites in vivo. We further show that this method can be easily adapted for high throughput imaging of live or fixed cells in vitro under a multitude of conditions in order to assess clonal fitness and evolution. The inherent variation in mouse studies, increasing complexity in experimental design which incorporate fate-mapping of individual cells, result in the need for a large cohort of mice to generate a robust dataset. High-throughput imaging techniques such as the one that we describe will enhance the data that can be used as input for the development of computational models aimed at modeling the metastatic process.

Diagnostic accuracy of MRI texture analysis for grading gliomas

PURPOSE

Texture analysis (TA) can quantify variations in surface intensity or patterns, including some that are imperceptible to the human visual system. The purpose of this study was to determine the diagnostic accuracy of radiomic based filtration-histogram TA to differentiate high-grade from low-grade gliomas by assessing tumor heterogeneity.

METHODS

Patients with a histopathological diagnosis of glioma and preoperative 3T MRI imaging were included in this retrospective study. A region of interest was manually delineated on post-contrast T1 images. TA was performed using commercially available research software. The histogram parameters including mean, standard deviation, entropy, mean of the positive pixels, skewness, and kurtosis were analyzed at spatial scaling factors ranging from 0 to 6 mm. The parameters were correlated with WHO glioma grade using Spearman correlation. Areas under the curve (AUC) were calculated using ROC curve analysis to distinguish tumor grades.

RESULTS

Of a total of 94 patients, 14 had low-grade gliomas and 80 had high-grade gliomas. Mean, SD, MPP, entropy and kurtosis each showed significant differences between glioma grades for different spatial scaling filters. Low and high-grade gliomas were best-discriminated using mean of 2 mm fine texture scale, with a sensitivity and specificity of 93% and 86% (AUC of 0.90).

CONCLUSIONS

Quantitative measurement of heterogeneity using TA can discriminate high versus low-grade gliomas. Radiomic data of texture features can provide complementary diagnostic information for gliomas.

Review of diagnostic error in anatomical pathology and the role and value of second opinions in error prevention

Aims

Diagnostic/interpretative accuracy can be challenging in anatomical pathology due to the subjective element of the diagnostic process. This can lead to false-negative or false-positive diagnoses of malignancy, variations in grading and diagnostic misclassification of a condition.

It is imperative that an accurate diagnosis is achieved so that an appropriate and timely treatment is administered to the patient, for example, the success of targeted molecular therapeutic options for treatment of cancer is dependent on accurate anatomical pathology diagnoses being issued.

Methods

A literature review of diagnostic accuracy in selected specimen categories was undertaken and was compared with data on metropolitan and regional pathologist diagnostic proficiency performance in an external quality assurance programme from surveys provided 2015–2017. For each specimen category, cases having attracted a diagnostic inaccuracy (ie, major discordance) of ≥20% and cases attracting a combined error rate (ie, major and minor discordance) of ≥30% are reviewed and discussed.

Results

The rate of inaccurate diagnoses (assessed as a major discordance) ranged from 3% to 9% among the different specimen groups, with highest mean percentage of inaccurate diagnoses in gynaecology, dermatopathology and gastrointestinal specimens.

Conclusions

It was possible to ascertain that gynaecology, dermatopathology and gastrointestinal specimens had presented the greatest diagnostic challenge to the participant pathologists, determined as highest rate of diagnostic inaccuracy, that is, major discordance with respective case target diagnoses.

Through a combination of routine second opinions, directed retrospective peer review and participation in appropriate external quality assurance schemes, the risk associated with these diagnoses can be minimised.

Interpathologist Diagnostic Agreement for Non-Small Cell Lung Carcinomas Using Current and Recent Classifications

CONTEXT.—

Measurement of interpathologist diagnostic agreement (IPDA) should allow pathologists to improve current diagnostic criteria and disease classifications.

OBJECTIVES.—

To determine how IPDA for pathologists' diagnoses of non-small cell lung carcinoma (NSCLC) is affected by the addition of a set of mucin and immunohistochemical (IHC) stains to hematoxylin-eosin (H&E) alone, by recent NSCLC reclassifications, by simplification of these classifications, and by pathologists' practice location, pulmonary pathology expertise, practice duration, and lung carcinoma case exposure.

DESIGN.—

We used a Web-based survey to present core images of 54 NSCLC cases to 22 practicing pathologists for diagnosis, initially as H&E only, then as H&E plus mucin and 4 IHC stains. Each case was diagnosed according to published 2004, 2011, and 2015 NSCLC classifications. Cohen's kappa was calculated for the 231 pathologist pairs as a measure of IPDA.

RESULTS.—

Twenty-two pathologists diagnosed 54 NSCLC cases by using 4 published classifications. IPDA is significantly higher for H&E/mucin/IHC diagnoses than for H&E-only diagnoses. IPDA for H&E/mucin/IHC diagnoses is highest with the 2015 classification. IPDA is estimated higher after collapse of stated diagnoses into subhead or dichotomized classes. IPDA for H&E/mucin/IHC diagnoses with the 2015 World Health Organization classification is similar for community and academic pathologists, and is higher when pathologists have pulmonary pathology expertise, have more than 6 years of practice experience, or diagnose more than 100 new lung carcinoma cases per year.

CONCLUSIONS.—

Higher IPDA is associated with use of mucin and IHC stains, with the 2015 NSCLC classification, and with pathologists' pulmonary pathology expertise, practice duration, and frequency of lung carcinoma cases.

Pediatric cancer pathology review from a single institution: Neuropathology expert opinion is essential for accurate diagnosis of pediatric brain tumors

BACKGROUND AND OBJECTIVES

Second pathology review has been reported to improve accuracy in oncologic diagnoses, including pediatric malignancies. We assessed the impact of second review on the diagnosis of pediatric malignancies at a tertiary care referral center in Beirut, Lebanon.

METHODS

Pathology reports of patients treated at the Children's Cancer Institute in Lebanon were retrospectively reviewed for the period 2008-2016 and compared with same samples' diagnoses at St. Jude Children's Research Hospital. Diagnostic disagreements were divided into major, minor, and none based on their effect on diagnosis and/or patient management.

RESULTS

Second review was requested for 171 cases, accounting for 19% of all cases during that period. Second opinion was mostly requested for brain tumors (62% of all brain tumor cases) and neuroblastoma for NMYC testing (65% of all neuroblastoma), while hematologic malignancies had the fewest referrals (3% of all hematologic cases). Major disagreements in second review occurred in 20 cases (12% of total), and minor disagreements in 21 cases (12% of total). The largest proportion of major disagreements (71%) occurred in pediatric brain tumors, and novel molecular tests contributed to the diagnosis in 55% of these cases.

CONCLUSIONS

The availability of a specialized pediatric neuropathologist and a basic panel of relevant molecular testing are essential for appropriate diagnosis of pediatric brain tumors. Centers that do not have the available infrastructure in place can benefit greatly from second review referrals for this challenging subset of tumors.

Association of Omics Features with Histopathology Patterns in Lung Adenocarcinoma

Adenocarcinoma accounts for more than 40% of lung malignancy, and microscopic pathology evaluation is indispensable for its diagnosis. However, how histopathology findings relate to molecular abnormalities remains largely unknown. Here, we obtained H&E-stained whole-slide histopathology images, pathology reports, RNA sequencing, and proteomics data of 538 lung adenocarcinoma patients from The Cancer Genome Atlas and used these to identify molecular pathways associated with histopathology patterns. We report cell-cycle regulation and nucleotide binding pathways underpinning tumor cell dedifferentiation, and we predicted histology grade using transcriptomics and proteomics signatures (area under curve >0.80). We built an integrative histopathology-transcriptomics model to generate better prognostic predictions for stage I patients (p = 0.0182 ± 0.0021) compared with gene expression or histopathology studies alone, and the results were replicated in an independent cohort (p = 0.0220 ± 0.0070). These results motivate the integration of histopathology and omics data to investigate molecular mechanisms of pathology findings and enhance clinical prognostic prediction.

Deep Learning for Classification of Colorectal Polyps on Whole-slide Images

CONTEXT

Histopathological characterization of colorectal polyps is critical for determining the risk of colorectal cancer and future rates of surveillance for patients. However, this characterization is a challenging task and suffers from significant inter- and intra-observer variability.

AIMS

We built an automatic image analysis method that can accurately classify different types of colorectal polyps on whole-slide images to help pathologists with this characterization and diagnosis.

SETTING AND DESIGN

Our method is based on deep-learning techniques, which rely on numerous levels of abstraction for data representation and have shown state-of-the-art results for various image analysis tasks.

SUBJECTS AND METHODS

Our method covers five common types of polyps (i.e., hyperplastic, sessile serrated, traditional serrated, tubular, and tubulovillous/villous) that are included in the US Multisociety Task Force guidelines for colorectal cancer risk assessment and surveillance. We developed multiple deep-learning approaches by leveraging a dataset of 2074 crop images, which were annotated by multiple domain expert pathologists as reference standards.

STATISTICAL ANALYSIS

We evaluated our method on an independent test set of 239 whole-slide images and measured standard machine-learning evaluation metrics of accuracy, precision, recall, and F1 score and their 95% confidence intervals.

RESULTS

Our evaluation shows that our method with residual network architecture achieves the best performance for classification of colorectal polyps on whole-slide images (overall accuracy: 93.0%, 95% confidence interval: 89.0%-95.9%).

CONCLUSIONS

Our method can reduce the cognitive burden on pathologists and improve their efficacy in histopathological characterization of colorectal polyps and in subsequent risk assessment and follow-up recommendations.

Quality assurance of anatomic pathology diagnoses: Comparison of alternate approaches

OBJECTIVES

Traditionally, a 10% review has been the basis for quality assurance programs in anatomic pathology. The effectiveness of such reviews has been questioned and alternative methodologies suggested. The study investigates the error detection rates for four quality assurance protocols.

METHODS

The detection rate for diagnostic errors in surgical pathology was calculated over a one year period using four different review procedures comprising: random 10% review, correlation of internal and external diagnoses following solicited external expert opinion, correlation of internal diagnoses with outside diagnoses in cases sent for review at a second institution treating the patient along with a focused review of dermatopathology cases over a 3 month period. Error rate was expressed as percentage of reviewed cases where the initial diagnosis differed from the review diagnosis. Error rates detected by each method were compared among the methods RESULTS: The 10% random review detected seventeen errors in 2147 cases (0.8%). Solicited case consultations requested by clinicians or internal pathologists detected five diagnostic errors in seventy cases (7.1%). Unsolicited reviews by outside institutions in the course of patient care detected three diagnostic errors in 190 cases (1.6%). Review of the dermatopathology material disclosed 5 diagnostic errors in 59 cases (8.5%).

CONCLUSIONS

Focused reviews initiated by diagnostic concerns of a clinician or pathologist, unsolicited reviews because of treatment at another institution and sub-specialty based reviews appear to be more effective in detecting diagnostic errors than the 10% random review. Quality assurance programs should include focused reviews in addition to 10% random review to maximize error detection.

A multisite validation of whole slide imaging for primary diagnosis using standardized data collection and analysis

Context:

Text-based reporting and manual arbitration for whole slide imaging (WSI) validation studies are labor intensive and do not allow for consistent, scalable, and repeatable data collection or analysis.

Objective:

The objective of this study was to establish a method of data capture and analysis using standardized codified checklists and predetermined synoptic discordance tables and to use these methods in a pilot multisite validation study.

Methods and Study Design:

Fifteen case report form checklists were generated from the College of American Pathology cancer protocols. Prior to data collection, all hypothetical pairwise comparisons were generated, and a level of harm was determined for each possible discordance. Four sites with four pathologists each generated 264 independent reads of 33 cases. Preestablished discordance tables were applied to determine site by site and pooled accuracy, intrareader/intramodality, and interreader intramodality error rates.

Results:

Over 10,000 hypothetical pairwise comparisons were evaluated and assigned harm in discordance tables. The average difference in error rates between WSI and glass, as compared to ground truth, was 0.75% with a lower bound of 3.23% (95% confidence interval). Major discordances occurred on challenging cases, regardless of modality. The average inter-reader agreement across sites for glass was 76.5% (weighted kappa of 0.68) and for digital it was 79.1% (weighted kappa of 0.72).

Conclusion:

These results demonstrate the feasibility and utility of employing standardized synoptic checklists and predetermined discordance tables to gather consistent, comprehensive diagnostic data for WSI validation studies. This method of data capture and analysis can be applied in large-scale multisite WSI validations.

Pediatric Neuropathology in Africa: Local Experience in Nigeria and Challenges and Prospects for the Continent

The present state of pediatric neuropathology practice is in rudimentary developmental stages in most parts of sub-Saharan Africa. We sought to determine the pattern of neurosurgical lesions in children diagnosed in southwestern Nigeria and briefly address issues surrounding the practice of this aspect of pathology in Africa. We performed a retrospective review of histopathologic results of biopsies obtained from pediatric patients with neurosurgical lesions at the Department of Pathology, Obafemi Awolowo University Teaching Hospitals Complex, Ile-Ife, Nigeria, between January 2001 and December 2011. Demographic and clinical data were extracted from the Ife-Ijesha cancer registry and histopathological diagnoses were confirmed. A total of 111 biopsies were reviewed with a maximum of 17 in 2001 and minimum of 3 in 2005. Patient ages ranged between 1 day and 16 years with a male:female ratio of 1.02:1. There were 53 spinal lesions, 15 intracranial lesions, 36 scalp masses, 6 skull lesions and 1 muscle biopsy. Most of the specimens were from myelomeningoceles. This documentation of the major types of pediatric neurological conditions encountered in clinical practice in this relatively resource-limited setting indicate the need for collaboration with better developed centers to improve training in neurosurgery and neuropathology to enhance the quality of clinical care for young patients in Africa.

Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features

Lung cancer is the most prevalent cancer worldwide, and histopathological assessment is indispensable for its diagnosis. However, human evaluation of pathology slides cannot accurately predict patients' prognoses. In this study, we obtain 2,186 haematoxylin and eosin stained histopathology whole-slide images of lung adenocarcinoma and squamous cell carcinoma patients from The Cancer Genome Atlas (TCGA), and 294 additional images from Stanford Tissue Microarray (TMA) Database. We extract 9,879 quantitative image features and use regularized machine-learning methods to select the top features and to distinguish shorter-term survivors from longer-term survivors with stage I adenocarcinoma (P<0.003) or squamous cell carcinoma (P=0.023) in the TCGA data set. We validate the survival prediction framework with the TMA cohort (P<0.036 for both tumour types). Our results suggest that automatically derived image features can predict the prognosis of lung cancer patients and thereby contribute to precision oncology. Our methods are extensible to histopathology images of other organs.

Diagnosis of lung cancer through manual histopathology evaluation is insufficient to predict patient survival. Here, the authors use computerized image processing to identify diagnostically relevant image features and use these features to distinguish lung cancer patients with different prognoses.

A Case of Malignant Gastrointestinal Stromal Tumor Initially Misdiagnosed as Malignant B-Cell Lymphoma

Errors that occur in anatomic pathology influence the treatment strategy of patients with malignancy. There are four general types of error with three subtypes in the category of defective interpretation. The first subtype is a false-negative diagnosis or undercall of the extent or severity of the lesion, the second is a false-positive diagnosis, and the third is misclassification. We herein report a 65-year-old female patient with malignant gastrointestinal stromal tumor that was diagnosed after reevaluation of the lesion at our hospital – and treated with proximal gastrectomy – after initial diagnosis as malignant B-cell lymphoma on esophagogastroduodenoscopy biopsy of a small gastric fundic mass and subsequent treatment with six cycles of CHOP chemotherapy with aggravation of the mass at another hospital.

Multicenter Assessment of Gram Stain Error Rates

Gram stains remain the cornerstone of diagnostic testing in the microbiology laboratory for the guidance of empirical treatment prior to availability of culture results. Incorrectly interpreted Gram stains may adversely impact patient care, and yet there are no comprehensive studies that have evaluated the reliability of the technique and there are no established standards for performance. In this study, clinical microbiology laboratories at four major tertiary medical care centers evaluated Gram stain error rates across all nonblood specimen types by using standardized criteria. The study focused on several factors that primarily contribute to errors in the process, including poor specimen quality, smear preparation, and interpretation of the smears. The number of specimens during the evaluation period ranged from 976 to 1,864 specimens per site, and there were a total of 6,115 specimens. Gram stain results were discrepant from culture for 5% of all specimens. Fifty-eight percent of discrepant results were specimens with no organisms reported on Gram stain but significant growth on culture, while 42% of discrepant results had reported organisms on Gram stain that were not recovered in culture. Upon review of available slides, 24% (63/263) of discrepant results were due to reader error, which varied significantly based on site (9% to 45%). The Gram stain error rate also varied between sites, ranging from 0.4% to 2.7%. The data demonstrate a significant variability between laboratories in Gram stain performance and affirm the need for ongoing quality assessment by laboratories. Standardized monitoring of Gram stains is an essential quality control tool for laboratories and is necessary for the establishment of a quality benchmark across laboratories.

Interpretive Diagnostic Error Reduction in Surgical Pathology and Cytology: Guideline From the College of American Pathologists Pathology and Laboratory Quality Center and the Association of Directors of Anatomic and Surgical Pathology

CONTEXT

Additional reviews of diagnostic surgical and cytology cases have been shown to detect diagnostic discrepancies.

OBJECTIVE

To develop, through a systematic review of the literature, recommendations for the review of pathology cases to detect or prevent interpretive diagnostic errors.

DESIGN

The College of American Pathologists Pathology and Laboratory Quality Center in association with the Association of Directors of Anatomic and Surgical Pathology convened an expert panel to develop an evidence-based guideline to help define the role of case reviews in surgical pathology and cytology. A literature search was conducted to gather data on the review of cases in surgical pathology and cytology.

RESULTS

The panel drafted 5 recommendations, with strong agreement from open comment period participants ranging from 87% to 93%. The recommendations are: (1) anatomic pathologists should develop procedures for the review of selected pathology cases to detect disagreements and potential interpretive errors; (2) anatomic pathologists should perform case reviews in a timely manner to avoid having a negative impact on patient care; (3) anatomic pathologists should have documented case review procedures that are relevant to their practice setting; (4) anatomic pathologists should continuously monitor and document the results of case reviews; and (5) if pathology case reviews show poor agreement within a defined case type, anatomic pathologists should take steps to improve agreement.

CONCLUSIONS

Evidence exists that case reviews detect errors; therefore, the expert panel recommends that anatomic pathologists develop procedures for the review of pathology cases to detect disagreements and potential interpretive errors, in order to improve the quality of patient care.