Research on Medical Image Diagnosis Models Based on Convolutional Neural Networks

Authors
  • Marina Caldwell

Abstract

This study explored the application of convolutional neural network (CNN) based medical image classification and compared the performance of five common deep learning models, including CNN, VGG16, ResNet-50, Inception-v3 and MobileNet. Through experiments on public medical image datasets, the classification accuracy, recall and F1 score of each model were evaluated. The experimental results show that VGG16 performs best in all evaluation indicators, with an accuracy of 90.2%, a recall of 88.5% and an F1 score of 89.3%. Although ResNet-50 also has high performance, it is slightly inferior to VGG16 in accuracy and recall. Inception-v3 and MobileNet perform relatively poorly in processing medical images, with lower accuracy and recall than the former two. CNN, as a baseline model, performs the worst in various indicators, showing its limitations in medical image analysis. Through comparative experiments, this study provides a theoretical basis for selecting appropriate deep learning models for medical image classification and provides guidance for the future application of deep learning in medical image diagnosis.

References
Cover Image
Downloads
Published
2025-02-28
Section
Articles
License
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Academic Scholar Hub Press also operates under the Creative Commons Licence CC-BY 4.0. This allows for copy and redistribute the material in any medium or format for any purpose, even commercially. The premise is that you must provide appropriate citation information.

How to Cite

Caldwell, M. (2025). Research on Medical Image Diagnosis Models Based on Convolutional Neural Networks. Journal of Computer Technology and Software, 4(2). https://doi.org/10.5281/zenodo.14984979