BREAST CANCER DIAGNOSIS USING IMAGE RETRIEVAL FOR DIFFERENT ULTRASONIC SYSTEMS (WA-P8)

Author(s) : Yu-Len Huang (Department of Computer Science and Information Engineering Tunghai University, Taiwan) Dar-Ren Chen (Department of General Surgery China Medical College & Hospital, Taiwan) Ya-Kuang Liu (Department of Computer Science and Information Engineering Tunghai University, Taiwan)

Abstract : This study employs the image retrieval technique to evaluate a series of pathologically proven breast tumors, and classified tumors as benign or malignant lesions. We evaluated 600 ultrasound (US) images of solid breast nod-ules including 230 malignant and 370 benign tumors. The US images were acquired from four different ultrasound systems. Firstly, the physician located regions-of-interest (ROI) of ultrasound images. The textual features from ROI sub-image are utilized to classify breast tumors. The principal component analysis (PCA) is used to reduce the dimension of textual feature vector and then the image retrieval technique was utilized to differentiate between benign and malignant tumors. Historical cases can be di-rectly added into the database and training of the diagno-sis system again is not needed. The accuracy of the pro-posed computer-aided diagnosis system (CAD) was 91.2%, the sensitivity was 97.0% and the specificity was 87.6%. This system differentiates solid breast nodules with a relatively high accuracy in the different US systems and helps inexperienced operators avoid misdiagnosis.

Menu