Optimization and Performance Evaluation of Deep Learning Algorithm in Medical Image Processing

• DOI: 10.54097/de0qx980
• Publication Date: 2024-04-26T07:27:05Z
• Authors: Jingbo Zhang; Lingxi Xiao; Yuwei Zhang; Jiatao Lai; Yutian Yang
• Keywords: Optimization algorithm

In this paper, the optimization and performance evaluation of deep learning algorithm in medical image processing are studied. Firstly, paper introduces importance challenges processing, expounds application prospect field. Subsequently, discusses methods detail, including model structure design, data preprocessing, super parameter adjustment so on. terms evaluation, study selected classic models such as U-Net, DeepLab DenseNet, compared them with ROC curve AUC value to evaluate their predictive ability classification. The results show that DenseNet shows high prediction accuracy, while U-Net is slightly average. Finally, advantages disadvantages each analyzed, future research direction prospected. This great significance promote development technology, provides important theoretical technical support for diagnosis treatment.