Detection of Hepatocellular Carcinoma in Contrast-Enhanced Magnetic Resonance Imaging Using Deep Learning Classifier: A Multi-Center Retrospective Study.
Junmo KimJi Hye MinSeon Kyoung KimSoo-Yong ShinMin Woo LeePublished in: Scientific reports (2020)
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.
Keyphrases
- deep learning
- convolutional neural network
- magnetic resonance imaging
- magnetic resonance
- contrast enhanced
- end stage renal disease
- artificial intelligence
- newly diagnosed
- computed tomography
- ejection fraction
- chronic kidney disease
- machine learning
- prognostic factors
- diffusion weighted
- peritoneal dialysis
- air pollution
- big data
- diffusion weighted imaging
- optical coherence tomography
- patient reported
- clinical evaluation