Exploiting Precision Mapping and Component-Specific Feature Enhancement for Breast Cancer Segmentation and Identification

TL;DR

This work tackles the challenge of accurate breast cancer detection in ultrasound imagery by proposing two deep learning frameworks: PMAD-LinkNet for precise tumor segmentation using a novel Precision Mapping Mechanism and CSFEC-Net for tissue-specific classification leveraging a Component-Specific Feature Enhancement Module with multi-level attention. The PMAD-LinkNet achieves a Dice coefficient of 97.2% and an IoU of 96.9%, while CSFEC-Net attains 99.2% accuracy with precision 99.3%, recall 99.1%, and F1-score 99.1% on a Kaggle ultrasound dataset, demonstrating strong boundary delineation and discriminative feature amplification between benign, malignant, and normal tissues. Key methodological contributions include dynamic spatial mapping conditioned on morphological variation and multi-scale attention-based feature enhancement for classification, validated against multiple baselines. The study highlights promising potential for automated, boundary-accurate breast cancer diagnosis with prospects for multimodal integration, explainable AI, and real-time deployment in varied clinical settings.

Abstract

Breast cancer is one of the leading causes of death globally, and thus there is an urgent need for early and accurate diagnostic techniques. Although ultrasound imaging is a widely used technique for breast cancer screening, it faces challenges such as poor boundary delineation caused by variations in tumor morphology and reduced diagnostic accuracy due to inconsistent image quality. To address these challenges, we propose novel Deep Learning (DL) frameworks for breast lesion segmentation and classification. We introduce a precision mapping mechanism (PMM) for a precision mapping and attention-driven LinkNet (PMAD-LinkNet) segmentation framework that dynamically adapts spatial mappings through morphological variation analysis, enabling precise pixel-level refinement of tumor boundaries. Subsequently, we introduce a component-specific feature enhancement module (CSFEM) for a component-specific feature-enhanced classifier (CSFEC-Net). Through a multi-level attention approach, the CSFEM magnifies distinguishing features of benign, malignant, and normal tissues. The proposed frameworks are evaluated against existing literature and a diverse set of state-of-the-art Convolutional Neural Network (CNN) architectures. The obtained results show that our segmentation model achieves an accuracy of 98.1%, an IoU of 96.9%, and a Dice Coefficient of 97.2%. For the classification model, an accuracy of 99.2% is achieved with F1-score, precision, and recall values of 99.1%, 99.3%, and 99.1%, respectively.

Figures (18)

• Figure 1: Overall Workflow Diagram of Proposed Work
• Figure 2: Samples of Breast Ultrasound Images and Masks (overlap) from the Dataset
• Figure 3: Training and Validation Dice Coefficient Curves of Proposed Localization Framework
• Figure 4: Training and Validation IoU Score Curves of Proposed Localization Framework
• Figure 5: Breast Ultrasound Images Observed After Each Preprocessing Step