Search

Aims: Machine learning (ML) binary classification in diagnostic histopathology is an area of intense investigation. Several assumptions, including training image quality/format and the number of training images required, appear to be similar in many studies irrespective of the paucity of supporting evidence. We empirically compared training image file type, training set size, and two common convolutional neural networks (CNNs) using transfer learning (ResNet50 and SqueezeNet)., Methods and Results: Thirty haematoxylin and eosin (H&E)-stained slides with carcinoma or normal tissue from three tissue types (breast, colon, and prostate) were photographed, generating 3000 partially overlapping images (1000 per tissue type). These lossless Portable Networks Graphics (PNGs) images were converted to lossy Joint Photographic Experts Group (JPG) images. Tissue type-specific binary classification ML models were developed by the use of all PNG or JPG images, and repeated with a subset of 500, 200, 100, 50, 30 and 10 images. Eleven models were generated for each tissue type, at each quantity of training images, for each file type, and for each CNN, resulting in 924 models. Internal accuracies and generalisation accuracies were compared. There was no meaningful significant difference in accuracies between PNG and JPG models. Models trained with more images did not invariably perform better. ResNet50 typically outperformed SqueezeNet. Models were generalisable within a tissue type but not across tissue types., Conclusions: Lossy JPG images were not inferior to lossless PNG images in our models. Large numbers of unique H&E-stained slides were not required for training optimal ML models. This reinforces the need for an evidence-based approach to best practices for histopathological ML., (© 2019 John Wiley & Sons Ltd.)

High resolution, multispectral, and multimodal imagery of tissue biopsies is an indispensable source of information for diagnosis and prognosis of diseases. Automatic extraction of relevant features from these imagery is a valuable assistance for medical experts. A primary step in computational histology is accurate image segmentation to detect the number and spatial distribution of cell nuclei in the tissue, along with segmenting other structures such as lumen and epithelial regions which together make up a gland structure. This chapter presents an automatic segmentation system for histopathology imaging. Spatial constraint fuzzy C-means provides an unsupervised initialization. An active contour algorithm that combines multispectral edge and region informations through a vector multiphase level set framework and Beltrami color metric tensors refines the segmentation. An improved iterative kernel filtering approach detects individual nuclei centers and decomposes densely clustered nuclei structures. The obtained results show high performances for nuclei detection compared to the human annotation.

Context: The Clinical Laboratory Improvement Amendments of 1988 require that laboratories perform cytologic-histologic correlation, although the optimal methods and the value of performing correlation have not been determined., Objective: To determine the similarities and differences in how laboratories perform cytologic-histologic correlation., Design: One hundred sixty-two American laboratories were sent a letter requesting copies of the materials they used in the cytologic-histologic correlation process. The returned materials were classified into the categories of forms, logs, and tally sheets. A checklist (derived from the College of American Pathologists Laboratory Accreditation Cytopathology Checklist) was developed to classify the "minimum expected" (15) and "additional" data points that laboratories collected when they performed a correlation., Participants: American pathology laboratories., Main Outcome Measures: Measures were percentage of laboratories that recorded minimum expected and additional data points and the frequency with which specific minimum expected data points were recorded., Results: The response frequency was 32.1%, and a total of 84 cytologic-histologic correlation materials were obtained. The only minimum expected variables recorded on forms or logs by more than 50% of laboratories were cytology case number, sign-out cytology diagnosis, surgical pathology case number, and sign-out surgical pathology diagnosis. Nine (17.3%) laboratories did not record data on forms, logs, or tally sheets. The mean number of minimum expected and additional variables recorded on forms was 6.5 and 8.7, respectively., Conclusions: Laboratories record data from the cytologic-histologic correlation process in a number of ways, indicating the lack of standardization of the data collection process.