{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Kather JN"],"funding":["NIDCR NIH HHS","European Research Council","NCI NIH HHS"],"pagination":["789-799"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC7610412"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["1(8)"],"pubmed_abstract":["Molecular alterations in cancer can cause phenotypic changes in tumor cells and their micro-environment. Routine histopathology tissue slides - which are ubiquitously available - can reflect such morphological changes. Here, we show that deep learning can consistently infer a wide range of genetic mutations, molecular tumor subtypes, gene expression signatures and standard pathology biomarkers directly from routine histology. We developed, optimized, validated and publicly released a one-stop-shop workflow and applied it to tissue slides of more than 5000 patients across multiple solid tumors. Our findings show that a single deep learning algorithm can be trained to predict a wide range of molecular alterations from routine, paraffin-embedded histology slides stained with hematoxylin and e"],"journal":["Nature cancer"],"pubmed_title":["Pan-cancer image-based detection of clinically actionable genetic alterations."],"pmcid":["PMC7610412"],"funding_grant_id":["771083","K08 DE026500","208237","P30 CA014599"],"pubmed_authors":["Kochanny S","Kather JN","Trautwein C","Speirs V","Ortiz-Bruchle NN","Sommer KAJ","Pearson AT","Buelow RD","Srisuwananukorn A","Grabsch HI","Krause J","Loeffler C","Luedde T","Bankhead P","Hanby AM","Niehues JM","van den Brandt PA","Echle A","Jager D","Boor P","Schulte JJ","Muti HS","Brenner H","Heij LR","Cipriani NA","Kooreman LFS","Patnaik A","Hoffmeister M"],"additional_accession":[]},"is_claimable":false,"name":"Pan-cancer image-based detection of clinically actionable genetic alterations.","description":"Molecular alterations in cancer can cause phenotypic changes in tumor cells and their micro-environment. Routine histopathology tissue slides - which are ubiquitously available - can reflect such morphological changes. Here, we show that deep learning can consistently infer a wide range of genetic mutations, molecular tumor subtypes, gene expression signatures and standard pathology biomarkers directly from routine histology. We developed, optimized, validated and publicly released a one-stop-shop workflow and applied it to tissue slides of more than 5000 patients across multiple solid tumors. Our findings show that a single deep learning algorithm can be trained to predict a wide range of molecular alterations from routine, paraffin-embedded histology slides stained with hematoxylin and e","dates":{"release":"2020-01-01T00:00:00Z","publication":"2020 Aug","modification":"2026-06-02T18:32:49.806Z","creation":"2025-04-06T18:04:44.579Z"},"accession":"S-EPMC7610412","cross_references":{"pubmed":["33763651"],"doi":["10.1038/s43018-020-0087-6"]}}