Hybrid Approach for Enhancing Lesion Segmentation in Fundus Images
Source: arXiv:2509.25549v2
Type: replace-cross
Abstract
Choroidal nevi are common benign pigmented lesions in the eye, with a small risk of transforming into melanoma. Early detection is critical to improving survival rates, but misdiagnosis or delayed diagnosis can lead to poor outcomes. Despite advancements in AI-based image analysis, diagnosing choroidal nevi in colour fundus images remains challenging, particularly for clinicians without specialized expertise.
Existing datasets often suffer from low resolution and inconsistent labelling, limiting the effectiveness of segmentation models. This paper addresses the challenge of achieving precise segmentation of fundus lesions, a critical step toward developing robust diagnostic tools.
Challenges in Current Segmentation Techniques
While deep learning models like U-Net have demonstrated effectiveness, their accuracy heavily depends on the quality and quantity of annotated data. Previous mathematical and clustering segmentation methods, though accurate, required extensive human input, making them impractical for medical applications.
Proposed Hybrid Model
This paper proposes a novel approach that combines mathematical and clustering segmentation models with insights from U-Net, leveraging the strengths of both methods. This hybrid model improves accuracy, reduces the need for large-scale training data, and achieves significant performance gains on high-resolution fundus images.
Performance Metrics
The proposed model achieves a Dice coefficient of 89.7% and an IoU of 80.01% on 1024*1024 fundus images, outperforming the Attention U-Net model, which achieved 51.3% and 34.2%, respectively. It also demonstrated better generalizability on external datasets.
Implications for Clinical Practice
This work forms a part of a broader effort to develop a decision support system for choroidal nevus diagnosis, with potential applications in automated lesion annotation to enhance the speed and accuracy of diagnosis and monitoring.
Conclusion
The integration of mathematical and deep learning techniques presents a promising pathway for improving lesion segmentation in fundus images. By reducing the reliance on extensive annotated datasets while maintaining high accuracy, this hybrid approach could significantly enhance the diagnostic capabilities available to clinicians, ultimately leading to better patient outcomes in the management of choroidal nevi.