Applications of artificial intelligence in imaging diagnosis and classification of tumors(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2026年第4期
- Page:
- 547-552
- Research Field:
- 医学人工智能
- Publishing date:
Info
- Title:
- Applications of artificial intelligence in imaging diagnosis and classification of tumors
- Author(s):
- CHEN Huayuan1; WU Xun2; CHEN Chaomin2; FENG Danqian3
- 1. Guangzhou Southern Medical Equipment Comprehensive Testing Co., Ltd., Guangzhou 510515, China 2. School of Biomedical Engineering, Southern Medical University, Guangzhou 510515, China 3. Guangdong Medical Devices Quality Surveillance and Test Institute, Guangzhou 510663, China
- Keywords:
- Keywords: artificial intelligence deep learning bone tumor brain tumor
- PACS:
- R318
- DOI:
- DOI:10.3969/j.issn.1005-202X.2026.04.019
- Abstract:
- Abstract: With the development of artificial intelligence (AI), its advantages in tumor diagnosis and classification are becoming increasingly pronounced. This review summarizes the applications of AI in imaging diagnosis and classification of tumors, with an emphasis on the recent advances of deep learning in the imaging diagnosis of bone tumors and brain tumors. To begin with, an introduction to the basic concepts of AI and deep learning and their application background in medical imaging is provided. The applications of AI in imaging diagnosis of bone tumors, including tumor detection, benign and malignant classification, and diagnosis of metastatic bone tumors are elaborated in details, demonstrating the potential of deep learning models in improving the diagnostic accuracy and efficiency. Furthermore, the application of AI in brain tumor classification is discussed, highlighting the progress in convolutional neural network and depth residual network for brain tumor MRI image classification, and the results show that AI achieves performance comparable to or even superior to experienced radiologists in brain tumor classification. Ultimately, the advantages and challenges of AI in tumor imaging diagnosis are analyzed, and it is pointed out that model interpretability, data availability, and generalization ability represent important research directions for future research. Although AI exhibits broad application prospects in tumor imaging diagnosis, further research is still required to address the existing challenges.
Last Update: 2026-04-29