Deep learning-based brain metastatic detection and treatment response assessment system on 3D MRI

It is very important to evaluate the exact treatment response in patients with metastatic brain cancer (BM), but it is labor-intensive. The goal of this study is to develop a computer-assisted detection (CAD) system for BM detection and treatment response evaluation based on deep learning. The detection system in this study accurately detected a BM nodule greater than 5 mm. In addition, the proposed deep learning system automatically compared the BM nodules between the two MM scans, measured the size change, and evaluated the response to treatment. The difference between the evaluation of the proposed system and the evaluation of experienced neuroradiologists was very significant. This deep learning approach will help accurately evaluate treatment responses and facilitate transition to precision medicine.