A Comprehensive Review on the Application of Artificial Intelligence for Predicting Postsurgical Recurrence Risk in...

Area of Science:

• Oncology
• Radiology
• Medical Informatics

Background:

• Non-small cell lung cancer (NSCLC) is a leading cause of cancer mortality worldwide.
• Postsurgical recurrence affects 30%-55% of early-stage NSCLC patients within five years.
• Predicting recurrence risk is crucial for improving patient management and outcomes.

Purpose of the Study:

• To review the utilization of artificial intelligence (AI) for predicting postsurgical recurrence risk in early-stage NSCLC.
• To analyze studies incorporating CT, PET, and clinical data for recurrence prediction.
• To identify current challenges and future directions in AI-driven recurrence prediction.

Main Methods:

• Literature search of studies published between 2018-2024.
• Focus on radiomics, machine learning, and deep learning using preoperative PET, CT, and PET/CT data.
• Inclusion of studies with or without clinical data integration; methodological quality assessed using METRICS.

Main Results:

• AI and radiomics models demonstrate potential in predicting postsurgical recurrence risk.
• Promising results observed with handcrafted radiomics features, deep learning, and multimodal models.
• Identified challenges include small sample sizes, lack of external validation, and interpretability issues.

Conclusions:

• Future research requires larger, prospective, multicenter studies for robust AI model development.
• Improving data integration, interpretability, and fusion of imaging modalities is essential.
• Standardizing methodologies and fostering collaboration will advance clinical utility and patient care.