Meeting Banner
Abstract #2626

Computerized Classification of Benign & Malignant Breast Lesions on DCE-MRI Utilizing Novel Shape Descriptors

Rachel Evonne Sparks1, Anant Madabhushi1

1Biomedical Engineering, Rutgers University, Piscataway, NJ, United States

The abstract presents a computerized decision support tool which utilizes novel morphologic descriptors for distinguishing benign from malignant breast lesions as they appear on DCE-MRI. The computerized decision support tools were evaluated on 41 suspicious breast lesions and gave a classification accurate of 83.0 4.5 %.