B Schmauch1, P Herent1, P Jehanno2, O Dehaene3, C Saillard1, C Aubé4, A Luciani5, N Lassau6, S Jégou1. 1. Owkin Inc, Research and Development Laboratory, 75, rue de Turbigo, 75003 Paris, France. 2. Owkin Inc, Research and Development Laboratory, 75, rue de Turbigo, 75003 Paris, France. Electronic address: paul.jehanno@owkin.com. 3. École Centrale d'Electronique (ECE), 75015 Paris, France. 4. Radiology Department, CHU Angers, 49933 Angers, France. 5. Radiology Department, AP-HP, Hôpitaux Universitaires Henri-Mondor, 94010 Creteil, France. 6. Radiology Department, Institut Gustave Roussy, 94805 Villejuif, France; IR4M, UMR8081CNRS, Université Paris-Sud, Université Paris-Saclay, 94805 Villejuif, France.
Abstract
PURPOSE: The purpose of this study was to create an algorithm that simultaneously detects and characterizes (benign vs. malignant) focal liver lesion (FLL) using deep learning. MATERIALS AND METHODS: We trained our algorithm on a dataset proposed during a data challenge organized at the 2018 Journées Francophones de Radiologie. The dataset was composed of 367 two-dimensional ultrasound images from 367 individual livers, captured at various institutions. The algorithm was guided using an attention mechanism with annotations made by a radiologist. The algorithm was then tested on a new data set from 177 patients. RESULTS: The models reached mean ROC-AUC scores of 0.935 for FLL detection and 0.916 for FLL characterization over three shuffled three-fold cross-validations performed with the training data. On the new dataset of 177 patients, our models reached a weighted mean ROC-AUC scores of 0.891 for seven different tasks. CONCLUSION: This study that uses a supervised-attention mechanism focused on FLL detection and characterization from liver ultrasound images. This method could prove to be highly relevant for medical imaging once validated on a larger independent cohort.
PURPOSE: The purpose of this study was to create an algorithm that simultaneously detects and characterizes (benign vs. malignant) focal liver lesion (FLL) using deep learning. MATERIALS AND METHODS: We trained our algorithm on a dataset proposed during a data challenge organized at the 2018 Journées Francophones de Radiologie. The dataset was composed of 367 two-dimensional ultrasound images from 367 individual livers, captured at various institutions. The algorithm was guided using an attention mechanism with annotations made by a radiologist. The algorithm was then tested on a new data set from 177 patients. RESULTS: The models reached mean ROC-AUC scores of 0.935 for FLL detection and 0.916 for FLL characterization over three shuffled three-fold cross-validations performed with the training data. On the new dataset of 177 patients, our models reached a weighted mean ROC-AUC scores of 0.891 for seven different tasks. CONCLUSION: This study that uses a supervised-attention mechanism focused on FLL detection and characterization from liver ultrasound images. This method could prove to be highly relevant for medical imaging once validated on a larger independent cohort.