For prostate cancer patients, the Gleason score is one of the most important prognostic factors, potentially determining treatment independent of the stage. However, Gleason scoring is based on subjective microscopic examination of tumor morphology and suffers from poor reproducibility. Here we present a deep learning system (DLS) for Gleason scoring whole-slide images of prostatectomies. Our system was developed using 112 million pathologist-annotated image patches from 1226 slides, and evaluated on an independent validation dataset of 331 slides. Compared to a reference standard provided by genitourinary pathology experts, the mean accuracy among 29 general pathologists was 0.61 on the validation set. The DLS achieved a significantly higher diagnostic accuracy of 0.70 ( p = 0.002) and trended towards better patient risk stratification in correlations to clinical follow-up data. Our approach could improve the accuracy of Gleason scoring and subsequent therapy decisions, particularly where specialist expertise is unavailable. The DLS also goes beyond the current Gleason system to more finely characterize and quantitate tumor morphology, providing opportunities for refinement of the Gleason system itself.

Supplemental Digital Content is available in the text.

Human stem-cell derivatives are likely to play an important role in the future of regenerative medicine. Evaluation and comparison to their in vivo counterparts is critical for assessment of their therapeutic potential. Transcriptomics was used to compare a new differentiation derivative of human embryonic stem (hES) cells--retinal pigment epithelium (RPE)--to human fetal RPE. Several hES cell lines were differentiated into putative RPE, which expressed RPEspecific molecular markers and was capable of phagocytosis, an important RPE function. Isolated hES cell-derived RPE was able to transdifferentiate into cells of neuronal lineage and redifferentiate into RPE-like cells through multiple passages (>30 Population doublings). Gene expression profiling demonstrated their higher similarity to primary RPE tissue than of existing human RPE cell lines D407 and ARPE-19, which has been shown to attenuate loss of visual function in animals. This is the first report of the isolation and characterization of putative RPE cells from hES cells, as well as the first application of transcriptomics to assess embryonic stem-cell derivatives and their in vivo counterparts--a "differentiomics" outlook. We describe for the first time, a differentiation system that does not require coculture with animal cells or factors, thus allowing the production of zoonoses-free RPE cells suitable for subretinal transplantation in patients with retinal degenerative diseases. With the further development of therapeutic cloning, or the creation of the banks of homozygous human leucocyte antigen (HLA) hES cells using parthenogenesis, RPE lines could be generated to overcome the problem of immune rejection and could be one of the nearest term applications of stem-cell technology.

Context.-Nodal metastasis of a primary tumor influences therapy decisions for a variety of cancers. Histologic identification of tumor cells in lymph nodes can be laborious and error-prone, especially for small tumor foci.Objective.-To evaluate the application and clinical implementation of a state-of-the-art deep learning-based artificial intelligence algorithm (LYmph Node Assistant or LYNA) for detection of metastatic breast cancer in sentinel lymph node biopsies.Design.-Whole slide images were obtained from hematoxylin-eosin-stained lymph nodes from 399 patients (publicly available Camelyon16 challenge dataset). LYNA was developed by using 270 slides and evaluated on the remaining 129 slides. We compared the findings to those obtained from an independent laboratory (108 slides from 20 patients/86 blocks) using a different scanner to measure reproducibility.Results.-LYNA achieved a slide-level area under the receiver operating characteristic (AUC) of 99% and a tumor-level sensitivity of 91% at 1 false positive per patient on the Camelyon16 evaluation dataset. We also identified 2 ''normal'' slides that contained micrometastases. When applied to our second dataset, LYNA achieved an AUC of 99.6%. LYNA was not affected by common histology artifacts such as overfixation, poor staining, and air bubbles.Conclusions.-Artificial intelligence algorithms can exhaustively evaluate every tissue patch on a slide, achieving higher tumor-level sensitivity than, and comparable slidelevel performance to, pathologists. These techniques may improve the pathologist's productivity and reduce the number of false negatives associated with morphologic detection of tumor cells. We provide a framework to aid practicing pathologists in assessing such algorithms for adoption into their workflow (akin to how a pathologist assesses immunohistochemistry results).