A dual-stage method for lesion segmentation on digital mammograms

• Published in: Medical physics (Lancaster) Vol. 34; no. 11; pp. 4180 - 4193
• Main Authors: Yuan, Yading, Giger, Maryellen L., Li, Hui, Suzuki, Kenji, Sennett, Charlene
• Format: Journal Article
• Language: English
• Published: United States American Association of Physicists in Medicine 01.11.2007
• Subjects: ALGORITHMS; BIOMEDICAL RADIOGRAPHY; breast cancer; Breast Neoplasms - diagnosis; Breast Neoplasms - diagnostic imaging; cancer; CARCINOMAS; Computer‐aided diagnosis; DIAGNOSIS; Diagnosis, Computer-Assisted; Diagnosis, Differential; Diagnostic Imaging - methods; diagnostic radiography; Digital mammography; Digital radiography; edge detection; Female; geometric active contour model; gradient methods; Humans; IMAGE PROCESSING; image segmentation; Iteration theory; MAMMARY GLANDS; Mammography; Mammography - instrumentation; Mammography - methods; Mass lesion segmentation; Mass Screening - methods; Medical image contrast; medical image processing; Medical image segmentation; Medical imaging; Models, Statistical; Partial differential equations; Pattern Recognition, Automated - methods; Radiographic Image Enhancement - methods; Radiologists; RADIOLOGY AND NUCLEAR MEDICINE; Reproducibility of Results; tumours; geometric active contour model; breast cancer; Mass lesion segmentation; computer-aided diagnosis
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.2790837

Mass lesion segmentation on mammograms is a challenging task since mass lesions are usually embedded and hidden in varying densities of parenchymal tissue structures. In this article, we present a method for automatic delineation of lesion boundaries on digital mammograms. This method utilizes a geometric active contour model that minimizes an energy function based on the homogeneities inside and outside of the evolving contour. Prior to the application of the active contour model, a radial gradient index (RGI)-based segmentation method is applied to yield an initial contour closer to the lesion boundary location in a computationally efficient manner. Based on the initial segmentation, an automatic background estimation method is applied to identify the effective circumstance of the lesion, and a dynamic stopping criterion is implemented to terminate the contour evolution when it reaches the lesion boundary. By using a full-field digital mammography database with 739 images, we quantitatively compare the proposed algorithm with a conventional region-growing method and an RGI-based algorithm by use of the area overlap ratio between computer segmentation and manual segmentation by an expert radiologist. At an overlap threshold of 0.4, 85% of the images are correctly segmented with the proposed method, while only 69% and 73% of the images are correctly delineated by our previous developed region-growing and RGI methods, respectively. This resulting improvement in segmentation is statistically significant.