Surface-Enhanced Raman Scattering (SERS) combined with machine learning enables accurate diagnosis of cervical cancer: From molecule to cell to tissue level

Abstract

The rising number of cervical cancer cases is placing a heavy economic strain on the country and its people. Improving survival rates hinges on early detection, precise diagnosis, and thorough treatment. Common screening and diagnostic methods like Pap smears, HPV testing, colposcopy, and histopathological exams are used in clinical practice, but they are often costly, time-consuming, invasive, subjective, and may lack the necessary sensitivity and specificity for accurate diagnosis. Developing a quick, non-invasive, and precise method for cervical cancer screening is crucial. Raman spectroscopy offers structural insights without damaging samples, but its weak signals and interference from biological fluorescence limit its clinical use. Surface-Enhanced Raman Scattering (SERS) overcomes these challenges, and recent advances, especially when combined with machine learning, enhance cervical cancer diagnosis by enabling precise detection of tumor. This paper comprehensively reviews and summarizes the application of SERS in cervical cancer diagnosis, ranging from molecular biomarker detection to live cell level and then to tissue level diagnosis. By integrating with machine learning, it facilitates the development of accurate, non-invasive diagnosis of cervical cancer.