Topic: Machine Learning Models to Predict Clinical Outcomes: Promises and Perils

The surgical and clinical communities are witnessing a growing integration of Artificial Intelligence (AI) and Machine Learning (ML) into everyday practice — from predicting postoperative complications to estimating survival, treatment response, and functional outcomes. Across specialties, surgeons and clinicians are now actively developing and applying predictive algorithms to improve decision making, personalize care, and simulate trial scenarios in silico. The human desire to foresee the future is ancient - dating back to early civilizations observing the stars from atop Ziggurats - and today, for perhaps the first time, we possess the tools and knowledge to make accurate, data-driven predictions.

This Special Issue of Artificial Intelligence in Surgery invites contributions from the clinical and surgical communities to showcase their experiences with ML-based prediction models. We especially encourage submissions describing the development, validation, or clinical implementation of models designed to forecast outcomes in real patients or trial simulations. Whether rooted in imaging, clinical data, pathology, genomics, or multimodal inputs — we welcome studies that apply AI to solve real-world clinical challenges.

We also seek contributions that explore the methodological foundations of predictive modeling. What defines a robust and reproducible predictive model in surgery? How should data be selected, preprocessed, and validated? How can we navigate the risks of overfitting, bias, and limited generalizability? Articles focusing on model transparency, interpretability, calibration, and clinical utility are particularly welcome.

By bringing together applied and methodological perspectives, this Special Issue aims to define current standards, highlight best practices, and stimulate critical discussion on both the promises and pitfalls of AI-based clinical prediction.

We look forward to receiving contributions from surgeons, clinicians, data scientists, and interdisciplinary teams committed to shaping the future of predictive medicine.