Article
Universal encoding of pan-cancer histology by deep texture representations.
Cell reports - 1 Mar 2022
Komura Daisuke, Kawabe Akihiro, Fukuta Keisuke, Sano Kyohei, Umezaki Toshikazu, Koda Hirotomo, Suzuki Ryohei, Tominaga Ken, Ochi Mieko, Konishi Hiroki, Masakado Fumiya, Saito Noriyuki, Sato Yasuyoshi, Onoyama Takumi, Nishida Shu, Furuya Genta, Katoh Hiroto, Yamashita Hiroharu, Kakimi Kazuhiro, Seto Yasuyuki, Ushiku Tetsuo, Fukayama Masashi, Ishikawa Shumpei
Abstract excerpt
Cancer histological images contain rich biological and clinical information, but quantitative representation can be problematic and has prevented the direct comparison and accumulation of large-scale datasets. Here, we show successful universal encoding of cancer histology by deep texture representations (DTRs) produced by a bilinear convolutional neural network. DTR-based, unsupervised histological profiling,...
Read the complete abstract on PubMedTopics
Share this publication in a Topic to start or enrich a Post.