Background:The mitotic figure recognition contest at the 2012 International Conference on Pattern Recognition (ICPR) challenges a system to identify all mitotic figures in a region of interest of hematoxylin and eosin stained tissue, using each of three scanners (Aperio, Hamamatsu, and multispectral).Methods:Our approach combines manually designed nuclear features with the learned features extracted by convolutional neural networks (CNN). The nuclear features capture color, texture, and shape information of segmented regions around a nucleus. The use of a CNN handles the variety of appearances of mitotic figures and decreases sensitivity to the manually crafted features and thresholds.Results:On the test set provided by the contest, the trained system achieves F1 scores up to 0.659 on color scanners and 0.589 on multispectral scanner.Conclusions:We demonstrate a powerful technique combining segmentation-based features with CNN, identifying the majority of mitotic figures with a fair precision. Further, we show that the approach accommodates information from the additional focal planes and spectral bands from a multi-spectral scanner without major redesign.

Abstract. Despite the prognostic importance of mitotic count as one of the components of the Bloom -Richardson grade [3], several studies [2, 9, 10] have found that pathologists' agreement on the mitotic grade is fairly modest. Collecting a set of more than 4,200 candidate mitotic figures, we evaluate pathologists' agreement on individual figures, and train a computerized system for mitosis detection, comparing its performance to the classifications of three pathologists. The system's and the pathologists' classifications are based on evaluation of digital micrographs of hematoxylin and eosin stained breast tissue. On figures where the majority of pathologists agree on a classification, we compare the performance of the trained system to that of the individual pathologists. We find that the level of agreement of the pathologists ranges from slight to moderate, with strong biases, and that the system performs competitively in rating the ground truth set. This study is a step towards automatic mitosis count to accelerate a pathologist's work and improve reproducibility.

Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms of data, such as images and videos. However, the discrete nature of natural language makes the disentangling of textual representations more challenging (e.g., the manipulation over the data space cannot be easily achieved). Inspired by information theory, we propose a novel method that effectively manifests disentangled representations of text, without any supervision on semantics. A new mutual information upper bound is derived and leveraged to measure dependence between style and content. By minimizing this upper bound, the proposed method induces style and content embeddings into two independent low-dimensional spaces. Experiments on both conditional text generation and text-style transfer demonstrate the high quality of our disentangled representation in terms of content and style preservation.

We develop a system for the FEVER fact extraction and verification challenge that uses a high precision entailment classifier based on transformer networks pretrained with language modeling, to classify a broad set of potential evidence. The precision of the entailment classifier allows us to enhance recall by considering every statement from several articles to decide upon each claim. We include not only the articles best matching the claim text by TFIDF score, but read additional articles whose titles match named entities and capitalized expressions occurring in the claim text. The entailment module evaluates potential evidence one statement at a time, together with the title of the page the evidence came from (providing a hint about possible pronoun antecedents). In preliminary evaluation, the system achieves .5736 FEVER score, .6108 label accuracy, and .6485 evidence F1 on the FEVER shared task test set.

No abstract