Early cancer diagnosis is one of the most effective approaches for improving disease prognosis and patient survival. Current cancer diagnostic methods often suffer from limitations such as invasiveness, high cost, time-consuming process and limited sensitivity at early stages. Cancer biomarkers such as genetic, epigenetic, transcriptomic, proteomic, metabolomic indicators and extracellular vesicles serve as significant measurable molecular signatures reflecting the presence and progression of malignancy and also disease response to treatments. Luminescent biosensors, working based on light-emitting phenomena including fluorescence, chemiluminescence, bioluminescence, and electrochemiluminescence, provide rapid, highly sensitive, user-friendly and non-invasive platforms for detecting these biomarkers. These systems transduce resulting events of biomarker recognition into various luminescence signals that can be quantified conventiently. The integration of artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), further advances this field by enabling automated feature extraction, complex pattern recognition which is a highly promising possibility in detection of multiple biomarkers, and robust data interpretation, thereby enhancing accuracy, reproducibility, and device portability. This review presents a comprehensive overview of AI-assisted luminescent biosensors for cancer diagnosis, discussing the molecular basis of key biomarkers, recent developments in luminescent detection technologies, and the application of AI algorithms for analytical optimization. Finally, it outlines current challenges and future prospects toward the development of intelligent, point-of-care biosensors for real-time cancer screening and precision diagnostics.
Shalileh et al. (Mon,) studied this question.