Artificial intelligence in acute appendicitis: A comprehensive review of machine learning and deep learning...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Clinical Diagnostics

Background:

• Acute appendicitis (AAp) is a common surgical emergency with diagnostic challenges.
• Conventional methods have limitations in sensitivity and specificity, particularly in atypical presentations.
• AI, including machine learning (ML) and deep learning (DL), shows promise for enhancing AAp diagnosis.

Purpose of the Study:

• To review current AI applications in diagnosing adult and pediatric AAp.
• To evaluate AI models using clinical data, radiological imaging, and decision support systems.
• To assess the impact of explainable AI (XAI) on clinical trust.

Main Methods:

• Systematic review of 65 studies on AI in AAp diagnosis.
• Analysis of ML models (e.g., random forest, SVM) and DL models (e.g., CNNs).
• Evaluation of AI integrating multimodal data (clinical, laboratory, imaging) and XAI techniques.

Main Results:

• ML and DL models achieved >90% sensitivity and specificity for AAp diagnosis.
• DL models, especially CNNs, outperformed radiologists in interpreting imaging.
• Multimodal AI models demonstrated superior diagnostic precision.
• XAI methods enhanced model transparency and clinician trust.

Conclusions:

• AI significantly enhances AAp diagnostic accuracy and complication differentiation.
• Current evidence is largely retrospective; prospective, multicenter studies are needed for clinical implementation.
• Integrating AI into emergency care workflows is a future direction.