Detection of Hepatocellular Carcinoma in Contrast-Enhanced Magnetic Resonance Imaging Using Deep Learning Classifier: A Multi-Center Retrospective Study
Junmo Kim, Ji Hye Min, Seon Kyoung Kim, Soo-Yong Shin, Min Woo Lee
Abstract
Hepatocellular carcinoma (HCC) is one of the most common malignant tumors and a leading cause of cancer-related death worldwide. We propose a fully automated deep learning model to detect HCC using hepatobiliary phase magnetic resonance images from 549 patients who underwent surgical resection. Our model used a fine-tuned convolutional neural network and achieved 87% sensitivity and 93% specificity for the detection of HCCs with an external validation data set (54 patients). We also confirmed whether the lesion detected by our deep learning model is a true lesion using a class activation map.
Topics & Concepts
Hepatocellular carcinomaMagnetic resonance imagingConvolutional neural networkDeep learningHCCSMedicineRadiologyLesionClassifier (UML)Artificial intelligenceCarcinomaPathologyComputer scienceInternal medicineHepatocellular Carcinoma Treatment and PrognosisLiver Disease Diagnosis and TreatmentSpectroscopy Techniques in Biomedical and Chemical Research