The Journal of pathology | 2021
Weakly supervised annotation-free cancer detection and prediction of genotype in routine histopathology.
Abstract
Deep Learning is a powerful tool in computational pathology: it can be used for tumor detection and for predicting genetic alterations based on histopathology images alone. Conventionally, tumor detection and prediction of genetic alterations are two separate workflows. Newer methods have combined them, but require complex, manually engineered computational pipelines, restricting reproducibility and robustness. To address these issues, we present a new method for simultaneous tumor detection and prediction of genetic alterations: The Slide-Level Assessment Model (SLAM) uses a single off-the-shelf neural network to predict molecular alterations directly from routine pathology slides without any manual annotations, improving upon previous methods by automatically excluding normal and non-informative tissue regions. SLAM requires only standard programming libraries and is conceptually simpler than previous approaches. We have extensively validated SLAM for clinically relevant tasks using two large multicentric cohorts of colorectal cancer patients, DACHS from Germany and YCR-BCIP from the United Kingdom. We show that SLAM yields reliable slide-level classification of tumor presence with an area under the receiver operating curve (AUROC) of 0.980 (confidence interval 0.975, 0.984; N\u2009=\u20092297 tumor and N\u2009=\u20091281 normal slides). In addition, SLAM can detect microsatellite instability (MSI) / mismatch repair deficiency (dMMR) or microsatellite stability (MSS) /mismatch repair proficiency (pMMR) with an AUROC of 0.909 (0.888, 0.929; N\u2009=\u20092039 patients) and BRAF mutational status with an AUROC of 0.821 (0.786, 0.852; N\u2009=\u20092075 patients). The improvement with respect to previous methods was validated in a large external testing cohort in which MSI/dMMR status was detected with an AUROC of 0.900 (0.864, 0.931; N\u2009=\u2009805 patients). In addition, SLAM provides human-interpretable visualization maps, enabling the analysis of multiplexed network predictions by human experts. In summary, SLAM is a new simple and powerful method for computational pathology which could be applied to multiple disease contexts. This article is protected by copyright. All rights reserved.