The lab focuses on the diagnosis and treatment of cancer using advanced imaging techniques. There are 3 main projects: breast tomosynthesis, computer aided diagnosis, and radiation treatment planning.

First, while mammography remains the gold standard in breast cancer screening, it has many well known limitations. Dr. Lo leads a team from the Ravin Advanced Imaging Laboratories (see website above) that collaborates closely with Siemens Healthcare to develop breast tomosynthesis, a form of limited-angle tomography. Using a modified digital mammography system, tomosynthesis can acquire a 3D image quickly, easily, and at comparable dose as conventional mammography. Tomosynthesis may improve sensitivity of breast cancer diagnosis by helping radiologists to detect subtle lesions which would otherwise be obscured. In addition, tomosynthesis can also improve specificity since radiologists can better characterize benign cases and thus avoid unnecessary follow-up studies and procedures. For these reasons, tomosynthesis is the most exciting recent development in breast imaging, and the only technology with the potential to replace mammography in the near future. We have concluded an NIH-sponsored clinical trial that accrued nearly 400 subjects, and are now participating in a multi-center trial for Siemens to collect data for their FDA submission.

Second, since the 1990s, we have been a leader in computer aided diagnosis (CAD), which is an interdisciplinary field combining elements of medical physics, engineering, statistics, and bioinformatics. We have developed automated detection algorithms that use computer vision techniques to localize suspicious mammographic lesions. We have also designed predictive models that use machine learning and statistical analysis in order to classify mammograms as benign versus malignant. During these studies, we compiled one of the largest multi-institution breast cancer databases with approximately 5000 cases. We are in the process of translating these techniques to the new modality of breast tomosynthesis.

Finally, we are extending image retrieval techniques from radiology toward the problem of intensity modulated radiation therapy (IMRT), specifically to improve treatment planning for prostate and head & neck cancer. Our goal is to improve the efficiency and safety of treatment plans. The idea is simple - to match a new patient against a large database of existing patients based on similarities in their CT data, and then to use the existing treatment parameters from the best match to develop a new treatment plan with acceptable clinical quality. We have developed a database of several hundred prostate IMRT cases from Duke as well as two outside independent clinics, and we are now beginning to investigate the much more challenging problem of head & neck cancer.

Baker JA, Lo JY. Breast tomosynthesis: state-of-the-art and review of the literature. Acad Radiol. 2011 Oct;18(10):1298-310. Abstract

Chanyavanich V, Das SK, Lee WR, Lo JY. Knowledge-based IMRT treatment planning for prostate cancer. Med Phys. 2011 May;38(5):2515-22. Abstract

Mazurowski MA, Lo JY, Harrawood BP, Tourassi GD. Mutual information-based template matching scheme for detection of breast masses: From mammography to digital breast tomosynthesis. J Biomed Inform. 2011 May 1. Abstract

Singh S, Maxwell J, Baker JA, Nicholas JL, Lo JY. Computer-aided classification of breast masses: performance and interobserver variability of expert radiologists versus residents. Radiology. 2011 Jan;258(1):73-80. Abstract

Shafer CM, Samei E, Lo JY. The quantitative potential for breast tomosynthesis imaging. Med Phys. 2010 Mar;37(3):1004-16. Abstract

Jesneck JL, Mukherjee S, Yurkovetsky Z, Clyde M, Marks JR, Lokshin AE, Lo JY. Do serum biomarkers really measure breast cancer? BMC Cancer. 2009;9:164. Abstract

Singh S, Tourassi GD, Baker JA, Samei E, Lo JY. Automated breast mass detection in 3D reconstructed tomosynthesis volumes: a featureless approach. Med Phys. 2008 Aug;35(8):3626-36. Abstract