以數位影像相關法實現乳癌觸診裝置可行性分析

Abstract

乳癌為全球女性發生人數最多的癌症疾病，且發生率持續上升，由統計資料可知乳癌的死亡率和發現期數息息相關，故早期診斷為乳癌防治的重要發展目標。在實驗室過去的研究中，已建立出一套結合超音波影像及力量感測器的乳房腫瘤形狀力學模型，此模型藉由機器學習方式預測腫瘤的生物力學特性，以此特性協助醫師做乳癌診斷。 本研究應用數位影像相關法同時量測腫瘤外型及探頭與腫瘤之間的力學關係，將過去乳房腫瘤形狀力學模型所需的超音波設備及荷重元系統簡化至雙眼視覺模組，藉由此方法預期可以將腫瘤檢測以行動裝置執行，成為家用型醫療器材。 本研究以驗證數位影像相關法是否能夠用以量測乳房腫瘤特性（腫瘤尺寸、腫瘤與周圍組織硬度比），系統中包含自製橡膠探頭、雙眼視覺，實驗結果可得知探頭的變形與乳房腫瘤的尺寸、硬度皆有正向的關係，並且能夠透過探頭面上各點變形的情形，透過其變型比例及梯度等影像的變化來判讀腫瘤狀況。

Breast cancer is the most common cancer among women worldwide, and the incident continued to rise. From the statistics, breast cancer mortality is related to the cancer’s stage. Therefore, early detection in order to improve breast cancer outcome and survival remains the cornerstone of breast cancer control. In the past research in the laboratory, the breast tumor shape-biomechanical model that combined with ultrasound imaging and force sensors has been developed to predict the biomechanical properties of the tumor by machine learning. In this research, digital-imaging correlation method was used to measure tumor shape and the force reaction between the probe and the tumor. The ultrasound equipment and the load cell system needed in the past research were simplified to the binocular vision module. This method is expected to enable tumor detection to be performed as a mobile device and become a home type medical device. The system includes rubber probe, binocular vision. Experimental results can be informed that the deformation of the probe and breast tumor size, stiffness are positive relationship. Then, can read the tumor’s properties by the variants and gradient changes in the probe’s deformation image.