Identification of cervical cancer using laser-induced breakdown spectroscopy coupled with principal component analysis and support vector machine

Cervical cancer is one of the most widespread diseases in women. Traditional cancer diagnosis is extremely complicated and relies on subjective interpretation of biopsy material. In this work, laser-induced breakdown spectroscopy (LIBS) was used in cervical cancer recognition. In order to improve identification accuracy of cervical cancer by LIBS, the chemometric methods of principal component analysis (PCA) and support vector machine (SVM) were combined. The results show that the content of trace elements in normal tissues and cervical cancer tissues was significantly different. Normalized peak intensities of Na, Mg, and K in the cervical cancer tissues were significantly higher than normal tissues, and the normalized peak intensities of Ca in the normal tissues were higher than cervical cancer tissues. The identification accuracies of PCA-SVM are better than SVM, with the achieved accuracies of 94.44% and 93.06%, respectively. It can be concluded that LIBS techniques coupled with chemometric method is a potential in cancer tissue identification, which provides a preliminary research basis for real-time diagnosis of cancer tissues using LIBS.