Technical Name 深度強化學習框架使用超音波影像診斷腋窩淋巴結狀態
Project Operator National Taiwan University
Project Host 張瑞峰
The RL model develops a control policy directly from experience to predict statesrewards during a learning procedure. Hence, we designed a medical image environment including US images, different actions,rewards, agent learns in this environment to extract the ALN regionevaluates the status. The performance of our proposed method achieves an accuracy of 83.6, a sensitivity of 88.6,a specificity of 89.0.
Scientific Breakthrough
In some previous studies, several nomogramscomputer-aided diagnosis (CAD) systems by use of primary tumor characteristics such as tumor size, tumor typegrade, lymphovascular invasion,hormonal status have been developed to predict ALN status in breast cancer. However, those method needs human interventionextra exams,the performance was not high that needed further development. The purpose of this study is to develop a CAD system to determine the metastasis status in breast cancer using the RL method in 2D US ALN images.
Industrial Applicability
RL methods can tailor for achieving precise treatment for individual tasks, which are suitable in the clinicalmedical decision-making process in real-life healthcare applications. Although the axillary lymph node metastasis status is a significant factor in evaluating the breast cancer patient, there is still no clinical usage with maturecommercial products. This study will produce a real clinical practice tool that can help physicians evaluates the axillary lymph node metastasis status in the futurethe treatment direction of prognosis.
Matching Needs
Keyword lymph node status reinforcement learning computer-aided diagnosis ultrasound breast cancer diagnosis deep learning precision medicine tumor cancer
  • Contact
other people also saw