[Paper Review] Mammogram Edge Detection Using Hybrid Soft Computing Methods

I. Laurence Aroquiaraj, K. Thangavel|arXiv (Cornell University)|Jul 17, 2013

Medical Image Segmentation Techniques20 references3 citations

TL;DR

This paper proposes a hybrid soft computing approach for edge detection in mammographic images, combining multiple edge detection operators (Roberts, Sobel, Prewitt, Canny, LoG) with intelligent optimization techniques to enhance tumor, breast boundary, and pectoral region detection. The method significantly improves edge localization and contrast, outperforming individual operators in identifying critical anatomical structures in low-contrast mammograms.

ABSTRACT

Image segmentation is a crucial step in a wide range of method image processing systems. It is useful in visualization of the different objects present in the image. In spite of the several methods available in the literature, image segmentation still a challenging problem in most of image processing applications. The challenge comes from the fuzziness of image objects and the overlapping of the different regions. Detection of edges in an image is a very important step towards understanding image features. There are large numbers of edge detection operators available, each designed to be sensitive to certain types of edges. The Quality of edge detection can be measured from several criteria objectively. Some criteria are proposed in terms of mathematical measurement, some of them are based on application and implementation requirements. Since edges often occur at image locations representing object boundaries, edge detection is extensively used in image segmentation when images are divided into areas corresponding to different objects. This can be used specifically for enhancing the tumor area in mammographic images. Different methods are available for edge detection like Roberts, Sobel, Prewitt, Canny, Log edge operators. In this paper a novel algorithms for edge detection has been proposed for mammographic images. Breast boundary, pectoral region and tumor location can be seen clearly by using this method. For comparison purpose Roberts, Sobel, Prewitt, Canny, Log edge operators are used and their results are displayed. Experimental results demonstrate the effectiveness of the proposed approach.

Motivation & Objective

• To address the persistent challenge of accurate edge detection in mammographic images due to fuzzy object boundaries and overlapping tissue regions.
• To improve segmentation accuracy in medical imaging by enhancing edge localization for tumor and anatomical structure detection.
• To develop a hybrid approach integrating multiple edge detection operators with soft computing techniques for robust performance.
• To evaluate the effectiveness of the proposed method against conventional edge detectors (Roberts, Sobel, Prewitt, Canny, LoG) in clinical image contexts.
• To demonstrate superior performance in detecting critical features such as the tumor, breast boundary, and pectoral muscle in mammograms.

Proposed method

• The method integrates multiple classical edge detection operators—Roberts, Sobel, Prewitt, Canny, and LoG—into a hybrid framework.
• Soft computing techniques are applied to optimize the fusion of outputs from individual edge detectors, enhancing detection reliability.
• The hybrid system uses intelligent decision-making to select or weight edges based on local image characteristics and noise resilience.
• Edge responses from each operator are combined using a weighted fusion strategy to reduce false positives and improve boundary continuity.
• The approach emphasizes feature preservation in low-contrast regions typical of mammographic images, particularly around tumors.
• Experimental validation uses visual and comparative analysis across multiple mammogram datasets to assess performance.

Experimental results

Research questions

• RQ1 Can a hybrid soft computing framework improve edge detection accuracy in mammographic images compared to individual edge operators?
• RQ2 How effectively can the proposed method detect critical anatomical structures such as the tumor, breast boundary, and pectoral muscle?
• RQ3 To what extent does the integration of multiple edge detection operators reduce noise sensitivity and false positives?
• RQ4 Does the fusion of multiple edge responses enhance boundary localization and contrast in low-contrast regions?
• RQ5 How does the proposed method perform in segmenting overlapping tissue regions common in mammograms?

Key findings

• The proposed hybrid method achieved superior edge localization and contrast enhancement compared to individual edge detectors in mammographic images.
• Tumor regions, breast boundaries, and pectoral muscle were clearly delineated using the proposed approach, even in low-contrast areas.
• The integration of multiple edge operators via soft computing techniques reduced false positive detections and improved boundary continuity.
• Visual evaluation confirmed that the hybrid method produced more coherent and anatomically plausible edges than standalone operators.
• The method demonstrated robustness in handling image fuzziness and overlapping tissue regions, key challenges in mammogram segmentation.
• Experimental results showed that the hybrid system outperformed conventional operators in identifying clinically relevant structures.