Clinical applications of artificial intelligence in liver imaging
Akira Yamada1, Koji Kamagata2, Kenji Hirata3
1Department of Radiology, Shinshu University School of Medicine, Matsumoto, Nagano, Japan. a_yamada@shinshu-u.ac.jp.
La Radiologia Medica
|May 10, 2023
Summary
Artificial intelligence (AI) in liver imaging shows promise for segmentation and reconstruction, aiding in disease diagnosis and prognosis. However, further validation is needed for widespread clinical adoption of these AI tools.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence
Background:
- Clinical applications of AI in liver imaging are rapidly evolving.
- Computed tomography (CT) and magnetic resonance imaging (MRI) are key modalities.
- Latent Dirichlet Allocation (LDA) is used for topic analysis of research trends.
Purpose of the Study:
- To review the current status of AI in liver imaging.
- To identify challenges in clinical applications.
- To analyze research trends using LDA.
Main Methods:
- Topic analysis of PubMed search results using Latent Dirichlet Allocation (LDA).
- Review of clinical applications of AI in liver CT and MRI.
Main Results:
- Key topics identified: segmentation, hepatocellular carcinoma (HCC) and radiomics, metastasis, fibrosis, and reconstruction.
- Deep learning for liver segmentation is crucial for body composition analysis and biomarker development.
- Deep learning reconstruction shows potential in reducing contrast and radiation doses.
Conclusions:
- AI technologies in liver imaging, particularly deep learning, are advancing rapidly.
- Significant challenges remain, including external validation of models and disease-specific diagnostic performance evaluation.
- Clinical application of AI in liver imaging is still in developmental stages.
