Low Cost Automated Ultrasound for Breast Cancer Detection and Diagnosis

项目总结/文摘

Project Summary/Abstract Breast cancer is the most common cause of cancer death among women worldwide and the numbers are disproportionately high for women in developing countries. Developing countries demand low-cost, portable diagnostic tools that are easy to use and do not require additional professional staff to be effective. We propose a low-cost, portable, automated ultrasound device containing software to improve the image quality and provide computer-aided detection and diagnosis (CAD) to locate and distinguish between clearly benign and potentially malignant palpable breast lumps. While screening has been the focus for diagnosis in western countries, breast cancer most commonly presents in women less than 50 years of age and as a palpable lump in developing countries. The common types of breast lumps in this age group (cysts, fibroadenomas, fibrocystic change and cancer) are usually distinct on ultrasound with fewer than 10% representing malignancy in young women. Our goal is to develop a device that could be used by a local health aid to determine which breast lumps need to be biopsied because they are suspicious and which can be left alone. This would enable stressed healthcare delivery systems to focus resources on the women most likely to benefit from their efforts. In order to provide an effective tool for breast cancer triage in developing countries we propose to do the following: 1. Introduce novel algorithms for ultrasound image enhancement and computer-aided detection and diagnosis (CAD). These algorithms take advantage of temporal information available in a live ultrasound scan to determine the probability of malignancy. 2. Clinical trial in California to determine the sensitivity and specificity of the ultrasound device. A clinical trial on a cohort group that matches the women in developing countries will determine the effectiveness of the technology. 3. Clinical trial to validate effectiveness, acceptability and feasibility of technology in an LMIC environment. The clinical trial will be performed in Mexico This work is a close collaboration with breast cancer expert and surgeon, Dr. Susan Love, breast imaging radiologist, clinical trial expert Dr. Wendie Berg, medical and software product development and commercialization expert, Christine Podilchuk, PhD, technology and commercialization expert, Professor Richard Mammone, and professor of surgery and global health medicine expert, Dr. Ben Anderson.