Fiber Optic SPR Sensor and Deep Learning-Driven Embedded Detection Method for Clinical Trace Biomarkers

Clinical testing, with over four decades of development, encounters hurdles in detecting trace biomarkers such as tumor markers and inflammatory factors, owing to low sensitivity, lengthy procedures, and reliance on cumbersome equipment. This paper introduces an interdisciplinary strategy that integrates SPR optical fiber sensing, deep learning algorithms, and embedded systems. The proposed system adopts an integrated “physical sensing-intelligent analysis-embedded decision-making” framework, providing an efficient approach for rapid trace biomarker detection and enabling immediate clinical diagnosis, the study delves into the potential of SPR optical fiber sensors for highly sensitive trace biomarker capture, while also addressing acknowledged limitations and challenges that impede their widespread adoption.