Identification of Breast Cancer Using Integrated Information from MRI and Mammography

摘要: Objectives Integration of information from corresponding regions between the breast MRI and an X-ray mammogram could benefit detection cancer in clinical diagnosis. We aimed to provide a framework registration mammography evaluate diagnosis using combined information. Materials Methods 43 patients with 46 lesions underwent both scans, interval two examinations was around one month. The distribution malignant benign 31/46 based on histological results. Maximum intensity projection thin-plate spline methods were applied for image mammography. integrated evaluated results histology as reference. assessment annotations statistical analysis performed by radiologists. Results For cranio-caudal view, mean post-registration error 2.2±1.9 mm. For medio-lateral oblique proposed approach even better 3.0±2.4 In mammography, sensitivity 91.9±2.3% (29/31, 28/31), specificity 70.0±4.7% (11/15, 10/15), accuracy 84.8±3.1% (40/46, 38/46), positive predictive value 86.4±2.1% (29/33, 28/33) negative 80.8±5.4% (11/13, 10/13). Conclusion MRI aid shows potential improvements sensitivity, specificity, accuracy, PPV NPV compared use alone.