Applications of artificial intelligence in abdominal imaging.

Area of Science:

• Radiology and Medical Imaging
• Artificial Intelligence in Medicine
• Deep Learning and Radiomics

Background:

• Artificial intelligence (AI) is rapidly advancing, offering potential solutions for abdominal imaging challenges.
• AI, particularly deep learning and radiomics, demonstrates high accuracy in detecting various abdominal conditions across imaging modalities.

Purpose of the Study:

• To review the current state and future directions of AI applications in abdominal imaging.
• To highlight AI's potential in disease detection, classification, prognostication, and personalized medicine within abdominal radiology.

Main Methods:

• Review of AI applications, including deep learning and radiomics, in abdominal imaging.
• Analysis of AI's performance in detecting liver, pancreatic, renal, and bowel pathologies.
• Identification of challenges and future strategies for AI integration in clinical practice.

Main Results:

• AI models show high accuracy in automating segmentation, classification, and prognostication for abdominal conditions.
• AI surpasses traditional methods in specific diagnostic tasks but faces adoption barriers.
• Key challenges include data heterogeneity, lack of validation, and model interpretability.

Conclusions:

• AI has transformative potential to enhance diagnostic accuracy and personalize care in abdominal imaging.
• Overcoming challenges through multi-center collaboration, explainable AI, and standardized data is crucial.
• Successful integration of AI will redefine abdominal radiology and improve patient outcomes.