Scholarlink

<p>Given the high global incidence of breast cancer, developing early and accurate diagnostic methods for mammography is increasingly urgent. In this paper, we propose a new method (MoFCM), which is improved from the Fuzzy C-Means (FCM) algorithm by applying fuzzy logic techniques during image preprocessing, thereby preserving important features and enhancing segmentation efficiency. In addition, the proposed method integrates spatial information and spatial coefficients to help the FCM algorithm improve clustering ability in cases where noise appears in the input image. To evaluate the effectiveness of the proposed method, we conduct experiments, comparing it with the FCM and Enhanced Fuzzy C-Means (EnFCM) algorithms through clustering and segmentation efficiency evaluation indexes. The experimental results demonstrate that the proposed method achieves accurate mass segmentation efficiency on the DDSM, MVinDr and our datasets with the highest accuracy of 94.7%. This research contributes to the field of computer vision in medicine, paving the way for intelligent solutions to support breast cancer diagnosis, improve treatment efficiency, and enhance the quality of healthcare.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Novel Proposal For Mammogram Mass Segmentation Using an Improved Fuzzy C-means

  • 期刊论文

  • 发表日期：

    2025年10月16日

查看全文
  • Ho-Dat Tran,
  • Thuong-Cang Phan,
  • Anh-Cang Phan

摘要

Given the high global incidence of breast cancer, developing early and accurate diagnostic methods for mammography is increasingly urgent. In this paper, we propose a new method (MoFCM), which is improved from the Fuzzy C-Means (FCM) algorithm by applying fuzzy logic techniques during image preprocessing, thereby preserving important features and enhancing segmentation efficiency. In addition, the proposed method integrates spatial information and spatial coefficients to help the FCM algorithm improve clustering ability in cases where noise appears in the input image. To evaluate the effectiveness of the proposed method, we conduct experiments, comparing it with the FCM and Enhanced Fuzzy C-Means (EnFCM) algorithms through clustering and segmentation efficiency evaluation indexes. The experimental results demonstrate that the proposed method achieves accurate mass segmentation efficiency on the DDSM, MVinDr and our datasets with the highest accuracy of 94.7%. This research contributes to the field of computer vision in medicine, paving the way for intelligent solutions to support breast cancer diagnosis, improve treatment efficiency, and enhance the quality of healthcare.