Abstract: A generalizable and interpretable deep learning model for predicting biomarker status and biomarker metrics from histopathology slide images is provided.

Abstract: Methods and systems for determining microsatellite instability (MSI) directly from microsatellite region mappings for specific loci in the genome are provided. Techniques include an MSI assay that may be deployed in a paired form, that is, as tumor sample and matched normal sample MSI assay, or an unpaired form, that is, as a tumor-only MSI assay. The techniques provide an automated process for MSI determination by mapping read counts in tumor samples and normal samples and comparing the two, for an identified set of 43 microsatellite loci.

Abstract: Multi-modal approaches to predict tumor immune infiltration are based on integrating gene expression data and imaging features in a neural network-based framework. This framework is configured to estimate percent composition, and thus immune infiltration score, of a patient tumor biopsy sample. Multi-modal approaches may also be used to predict cell composition beyond immune cells via integrated multi-layer neural network frameworks.

Abstract: A generalizable and interpretable deep learning model for predicting microsatellite instability from histopathology slide images is provided. Microsatellite instability (MSI) is an important genomic phenotype that can direct clinical treatment decisions, especially in the context of cancer immunotherapies. A deep learning framework is provided to predict MSI from histopathology images, to improve the generalizability of the predictive model using adversarial training to new domains, such as on new data sources or tumor types, and to provide techniques to visually interpret the topological and morphological features that influence the MSI predictions.