Search

Background Microscopic colitis (MC) is a common cause of chronic watery diarrhea. Biopsies with characteristic histological features are crucial for establishing the diagnosis. The two main subtypes are collagenous colitis (CC) and lymphocytic colitis (LC) but incomplete forms exist. The disease course remains unpredictable varying from spontaneous remission to a relapsing course. Aim To identify possible histological predictors of course of disease. Methods Sixty patients from the European prospective MC registry (PRO-MC Collaboration) were included. Digitised histological slides stained with CD3 and Van Gieson were available for all patients. Total cell density and proportion of CD3 positive lymphocytes in lamina propria and surface epithelium were estimated by automated image analysis, and measurement of the subepithelial collagenous band was performed. Histopathological features were correlated to the number of daily stools and daily watery stools at time of endoscopy and at baseline as well as the clinical disease course (quiescent, achieved remission after treatment, relapsing or chronic active) at 1-year follow-up. Results Neither total cell density in lamina propria, proportion of CD3 positive lymphocytes in lamina propria or surface epithelium, or thickness of collagenous band showed significant correlation to the number of daily stools or daily watery stools at any point of time. None of the assessed histological parameters at initial diagnosis were able to predict clinical disease course at 1-year follow-up. Conclusions Our data indicate that the evaluated histological parameters were neither markers of disease activity at the time of diagnosis nor predictors of disease course.

Automated detection of cancer metastases in lymph nodes has the potential to improve the assessment of prognosis for patients. To enable fair comparison between the algorithms for this purpose, we set up the CAMELYON17 challenge in conjunction with the IEEE International Symposium on Biomedical Imaging 2017 Conference in Melbourne. Over 300 participants registered on the challenge website, of which 23 teams submitted a total of 37 algorithms before the initial deadline. Participants were provided with 899 whole-slide images (WSIs) for developing their algorithms. The developed algorithms were evaluated based on the test set encompassing 100 patients and 500 WSIs. The evaluation metric used was a quadratic weighted Cohen's kappa. We discuss the algorithmic details of the 10 best pre-conference and two post-conference submissions. All these participants used convolutional neural networks in combination with pre- and postprocessing steps. Algorithms differed mostly in neural network architecture, training strategy, and pre- and postprocessing methodology. Overall, the kappa metric ranged from 0.89 to -0.13 across all submissions. The best results were obtained with pre-trained architectures such as ResNet. Confusion matrix analysis revealed that all participants struggled with reliably identifying isolated tumor cells, the smallest type of metastasis, with detection rates below 40%. Qualitative inspection of the results of the top participants showed categories of false positives, such as nerves or contamination, which could be targeted for further optimization. Last, we show that simple combinations of the top algorithms result in higher kappa metric values than any algorithm individually, with 0.93 for the best combination.

Simple Summary Around 15% of breast cancer patients are diagnosed as triple-negative (TNBC), which have significantly lower 5-year survival rates (77%) than other types of breast cancer (93%). Our study aimed at developing an image analysis-based biomarker to assess how the immune system interacts with the tumor and investigate the potential added value of stromal tumor-infiltrating lymphocytes (sTIL) for the prognosis of overall survival compared to the manual approach. In a large retrospective cohort of 257 patients, we found that our fully automated hematoxylin and eosin (H&E) image analysis pipeline can quantify sTIL density showing both high concordance with manual scoring and association with the prognosis of patients with TNBC. It also overcomes natural limitations of manual assessment that hinder clinical adoption of the immune biomarker. We conclude that sTIL scoring by automated image analysis has prognostic potential comparable to manual scoring and should be further investigated for future use in a clinical setting. Abstract Triple-negative breast cancer (TNBC) is an aggressive and difficult-to-treat cancer type that represents approximately 15% of all breast cancers. Recently, stromal tumor-infiltrating lymphocytes (sTIL) resurfaced as a strong prognostic biomarker for overall survival (OS) for TNBC patients. Manual assessment has innate limitations that hinder clinical adoption, and the International Immuno-Oncology Biomarker Working Group (TIL-WG) has therefore envisioned that computational assessment of sTIL could overcome these limitations and recommended that any algorithm should follow the manual guidelines where appropriate. However, no existing studies capture all the concepts of the guideline or have shown the same prognostic evidence as manual assessment. In this study, we present a fully automated digital image analysis pipeline and demonstrate that our hematoxylin and eosin (H&E)-based pipeline can provide a quantitative and interpretable score that correlates with the manual pathologist-derived sTIL status, and importantly, can stratify a retrospective cohort into two significant distinct prognostic groups. We found our score to be prognostic for OS (HR: 0.81 CI: 0.72–0.92 p = 0.001) independent of age, tumor size, nodal status, and tumor type in statistical modeling. While prior studies have followed fragments of the TIL-WG guideline, our approach is the first to follow all complex aspects, where appropriate, supporting the TIL-WG vision of computational assessment of sTIL in the future clinical setting.

Deep learning-based algorithms have shown great promise for assisting pathologists in detecting lymph node metastases when evaluated based on their predictive accuracy. However, for clinical adoption, we need to know what happens when the test set dramatically changes from the training distribution. In such settings, we should estimate the uncertainty of the predictions, so we know when to trust the model (and when not to). Here, we i) investigate current popular methods for improving the calibration of predictive uncertainty, and ii) compare the performance and calibration of the methods under clinically relevant in-distribution dataset shifts. Furthermore, we iii) evaluate their performance on the task of out-of-distribution detection of a different histological cancer type not seen during training. Of the investigated methods, we show that deep ensembles are more robust in respect of both performance and calibration for in-distribution dataset shifts and allows us to better detect incorrect predictions. Our results also demonstrate that current methods for uncertainty quantification are not necessarily able to detect all dataset shifts, and we emphasize the importance of monitoring and controlling the input distribution when deploying deep learning for digital pathology.