SLIViT: a general AI framework for clinical-feature diagnosis from limited 3D biomedical-imaging data

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Healthcare
• Biomedical Data Science

Background:

• Accurate measurement of disease-related risk factors in medical imaging is crucial for early diagnosis and treatment.
• Current deep learning models often require large annotated datasets, which are scarce in medical applications.
• Domain-specific models may lack generalizability across different imaging modalities and tasks.

Conclusions:

• SLIViT offers a robust and accurate solution for disease risk factor measurement in volumetric biomedical imaging.
• The framework demonstrates high efficacy even with limited annotated training data, addressing a key challenge in medical AI.
• SLIViT has the potential to automate and accelerate medical research and clinical workflows.