面向口腔癌早期诊断的光电联合一体化定量检测关键技术研究

口腔癌是一种具有严重危害的口腔疾病，由于缺乏行之有效的早期检测手段，导致口腔癌患者的5年生存率仍停留在约50%，且近几十年没有变化，因而迫切需要发展新的无创检测技术来解决这一全球性问题。本研究针对口腔癌早期检测中亟待解决的关键科学问题，致力于发展一种可提高检测灵敏度和特异性的光电联合定量检测技术。主要包括：搭建可实现多波长-多源探距离漫反射光谱测量与生物电阻抗测量的一体化系统，以提高系统的灵敏度、空间分辨以及测量数据的精确性和信息完备性；将CUDA技术引入可视为“金标准”的蒙特卡洛模拟，并加入表征组织体微结构的相位函数高阶参量进行精确、快速建模，在此基础上利用人工神经网络实现多参数准确、高效和高鲁棒性反构，并结合生物电阻抗参数，实现双模态联合诊断以提高现有方法的特异性和灵敏度等主要性能，为早期口腔癌检测和筛查提供新的有力工具。另外，该技术对其它粘膜病变组织检测也具有重要的借鉴意义。

Oral cancer is a serious oral disease，the overall 5-year survival rates for oral cancer have remained low at approximately 50% for the past decades due to the lack of an efficient way for an early detection. Therefore, it is urgent to develop new detection technology to solve this global problem. This project is devoted to developing a quantitative detection technique combined diffuse reflectance spectroscopy and bioelectrical impedance measurement to improve the sensitivity and specificity of detection. It mainly includes: an integrated system combined diffuse reflectance spectroscopy with multi-wave and multi-source-detector distance measurment, and bioelectrical impedance measurement is established to improve the sensitivity and spatial resolution of the system and the accuracy and information completeness of the measured data. CUDA technology is implemented in Monte Carlo simulation regarded as the "gold standard", and the higher-order parameters of the phase function representing the microstructure information of the tissue are added for accurately and rapidly modeling the process of the photon migration in tissue. On this basis, the artificial neural network is used to inversely reconstruct multi-parameters with high accuracy, efficiency and robustness, Furtherly combining with bioelectrical impedance parameters, a joint diagnosis is implemented to improve the specificity and sensitivity, and provide a new powerful tool for the detection and screening of early oral cancer. In addition, this novel method also has important reference significance for the detection of other mucosal pathological tissues.