Automated tumor assessment of squamous cell carcinoma on tongue cancer patients with hyperspectral imaging

摘要

Head and neck cancer (HNC) includes cancers in the oralasal cavity, pharynx, larynx, etc., and it is the sixthmost common cancer worldwide. The principal treatment is surgical removal where a complete tumor resection iscrucial to reduce the recurrence and mortality rate. Intraoperative tumor imaging enables surgeons to objectivelyvisualize the malignant lesion to maximize the tumor removal with healthy safe margins. Hyperspectral imaging(HSI) is an emerging imaging modality for cancer detection, which can augment surgical tumor inspection,currently limited to subjective visual inspection. In this paper, we aim to investigate HSI for automated cancerdetection during image-guided surgery, because it can provide quantitative information about light interactionwith biological tissues and exploit the potential for malignant tissue discrimination. The proposed solution formsa novel framework for automated tongue-cancer detection, explicitly exploiting HSI, which particularly uses thespectral variations in specic bands describing the cancerous tissue properties. The method follows a machinelearningbased classification, employing linear support vector machine (SVM), and offers a superior sensitivityand a significant decrease in computation time. The model evaluation is on 7 ex-vivo specimens of squamouscell carcinoma of the tongue, with known histology. The HIS combined with the proposed classification reachesa sensitivity of 94%, specificity of 68% and area under the curve (AUC) of 92%. This feasibility study paves theway for introducing HIS as a non-invasive imaging aid for cancer detection and increase of the efiectiveness ofsurgical oncology.