Deep learning for the detection of anatomical tissue structures and neoplasms of the skin on scanned histopathological tissue sections

Frontiers in Oncology

Oncology

Charlotte Janßen¹  Frithjof Lobers²  Ulrich Sack²  Katharina Kriegsmann³  Georg Steinbuss³  Rolf Rüdinger Meliß⁴  Mark Kriegsmann⁵  Christiane Zgorzelski⁵  Jörg Kriegsmann⁶  Thomas Muley⁷ 

[1] Center for Industrial Mathematics (ZeTeM), University of Bremen, Bremen, Germany;Department of Clinical Immunology, Medical Faculty, University of Leipzig, Leipzig, Germany;Department of Hematology, Oncology and Rheumatology, Heidelberg University, Heidelberg, Germany;Institute for Dermatopathology, Hannover, Germany;Institute of Pathology, Heidelberg University, Heidelberg, Germany;MVZ Histology, Cytology and Molecular Diagnostics Trier, Trier, Germany;Proteopath Trier, Trier, Germany;Translational Lung Research Centre (TLRC) Heidelberg, Member of the German Centre for Lung Research (DZL), Heidelberg, Germany;

关键词: deep learning;    pathology;    artificial intelligence;    dermatopathology;    digital pathology;    deep learning - artificial neural network;   

DOI  :  10.3389/fonc.2022.1022967

 received in 2022-08-19, accepted in 2022-10-24,  发布年份 2022

来源: Frontiers

【 摘 要 】

Basal cell carcinoma (BCC), squamous cell carcinoma (SqCC) and melanoma are among the most common cancer types. Correct diagnosis based on histological evaluation after biopsy or excision is paramount for adequate therapy stratification. Deep learning on histological slides has been suggested to complement and improve routine diagnostics, but publicly available curated and annotated data and usable models trained to distinguish common skin tumors are rare and often lack heterogeneous non-tumor categories. A total of 16 classes from 386 cases were manually annotated on scanned histological slides, 129,364 100 x 100 µm (~395 x 395 px) image tiles were extracted and split into a training, validation and test set. An EfficientV2 neuronal network was trained and optimized to classify image categories. Cross entropy loss, balanced accuracy and Matthews correlation coefficient were used for model evaluation. Image and patient data were assessed with confusion matrices. Application of the model to an external set of whole slides facilitated localization of melanoma and non-tumor tissue. Automated differentiation of BCC, SqCC, melanoma, naevi and non-tumor tissue structures was possible, and a high diagnostic accuracy was achieved in the validation (98%) and test (97%) set. In summary, we provide a curated dataset including the most common neoplasms of the skin and various anatomical compartments to enable researchers to train, validate and improve deep learning models. Automated classification of skin tumors by deep learning techniques is possible with high accuracy, facilitates tumor localization and has the potential to support and improve routine diagnostics.

【 授权许可】

Unknown   
Copyright © 2022 Kriegsmann, Lobers, Zgorzelski, Kriegsmann, Janßen, Meliß, Muley, Sack, Steinbuss and Kriegsmann