Artificial intelligence-assisted photodynamic diagnosis and photodynamic therapy against cancer

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Artificial intelligence (AI) comprises advanced computational systems designed to simulate selected aspects of human cognitive functions, such as learning, reasoning, and perception. While AI does not fully replicate human cognition, its analytical processing of large datasets through “machine learning (ML)” and “deep learning (DL)” has increasingly been applied in medicine, including photodynamic diagnosis (PDD) and photodynamic therapy (PDT). In oncology, these modalities are increasingly used for early cancer detection, tumor margin delineation, and personalized therapeutic planning. These applications involve AI-PDD/PDT workflows, dosimetry, predictive modeling, and translational opportunities across Western and Chinese medicine. The working process of AI-PDD/PDT systems encompasses treatment workflows, dosimetry optimization, real-time monitoring, and outcome prediction. AI-driven systems standardize photosensitizer preparation, reduce human error, and enhance reproducibility in PDT. Findings are synthesized conceptually across diverse studies, focusing on thematic integration of AI-assisted PDD/PDT mechanisms rather than quantitative pooling or statistical comparison of outcomes. Cancer applications include bladder, gastric, colorectal, and breast tumors, where AI-PDD/PDT improves diagnostic precision and enhances therapeutic efficacy. The rapid development of AI in recent years has extended into PDT, supporting fundamental research, clinical translation, and therapeutic innovation. AI-PDD/PDT demonstrates potential to enhance accuracy, safety, and efficiency through workflow optimization and predictive modeling. However, current evidence remains preliminary, and further systematic validation and clinical trials are required to substantiate these proposed benefits. Future milestones include refining the interface between AI and PDD/PDT and conducting prospective, multicenter trials to establish clinical utility.