Machine learning algorithms hold the promise to effectively automate the analysis of histopathological images that are routinely generated in clinical practice. Any machine learning method used in the clinical diagnostic process has to be extremely accurate and, ideally, provide a measure of uncertainty for its predictions. Such accurate and reliable classifiers need enough labelled data for training, which requires time-consuming and costly manual annotation by pathologists. Thus, it is critical to minimise the amount of data needed to reach the desired accuracy by maximising the efficiency of training. We propose an accurate, reliable and active (ARA) image classification framework and introduce a new Bayesian Convolutional Neural Network (ARA-CNN) for classifying histopathological images of colorectal cancer. The model achieves exceptional classification accuracy, outperforming other models trained on the same dataset. The network outputs an uncertainty measurement for each tested image. We show that uncertainty measures can be used to detect mislabelled training samples and can be employed in an efficient active learning workflow. Using a variational dropout-based entropy measure of uncertainty in the workflow speeds up the learning process by roughly 45%. Finally, we utilise our model to segment whole-slide images of colorectal tissue and compute segmentation-based spatial statistics.

Machine learning algorithms hold the promise to effectively automate the analysis of histopathological images that are routinely generated in clinical practice. Any machine learning method used in the clinical diagnostic process has to be extremely accurate and, ideally, provide a measure of uncertainty for its predictions. Such accurate and reliable classifiers need enough labelled data for training, which requires time-consuming and costly manual annotation by pathologists. Thus, it is critical to minimise the amount of data needed to reach the desired accuracy by maximising the efficiency of training. We propose an accurate, reliable and active (ARA) image classification framework and introduce a new Bayesian Convolutional Neural Network (ARA-CNN) for classifying histopathological images of colorectal cancer. The model achieves exceptional classification accuracy, outperforming other models trained on the same dataset. The network outputs an uncertainty measurement for each tested image. We show that uncertainty measures can be used to detect mislabelled training samples and can be employed in an efficient active learning workflow. Using a variational dropout-based entropy measure of uncertainty in the workflow speeds up the learning process by roughly 45%. Finally, we utilise our model to segment whole-slide images of colorectal tissue and compute segmentation-based spatial statistics.

Diarrhea constitutes a frequent and often debilitating complication of allogeneic hematopoietic cell transplantation (alloHCT). Especially when accompanied by jaundice, skin rash, and symptoms of the upper gastrointestinal tract, diarrhea strongly suggests emergence of acute graft-versus-host disease (GvHD), which is a serious immune complication of the procedure, with possible fatal consequences. However, especially when diarrhea occurs as an isolated symptom, the other causes must be excluded before initiation of GvHD treatment with immune-suppressive drugs. In this article, a broad overview of the literature of different causes of diarrhea in the setting of alloHCT is provided, revealing causes and presentations different from those observed in the general population. We discuss gastrointestinal GvHD with a special focus on biomarkers, but also uncover underlying infectious: viral, bacterial, and parasitic as well as toxic causes of diarrhea. Finally, we suggest a practical algorithm of approach to patients with diarrhea after alloHCT, which may help to establish a proper diagnosis and initiate a causative treatment.