Pathological Myopia detection from selective fundus image features

Zhang, Zhuo; Cheng, Jun; Liu, Jiang; Kong, Chui Chee; Mei, Saw Seang

doi:10.1109/iciea.2012.6361007

Cited by 6 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When it comes to specific diagnostic methods for pathologic myopia, Liu et al proposed the PAMELA (Pathological Myopia Detection Through Peri-papillary Atrophy) system, which automatically receives retinal fundus images, performs region of interest (ROI) extraction and optic disc segmentation, and uses support vector machine (SVM) to automatically diagnose pathological myopia based on the feature of peripapillary atrophy (PPA) in a dataset containing 80 fundus images [8]. Zhang et al utilized the Minimum Redundancy-Maximum Relevancy (mRMR) feature selection technique to select and rank candidate features, and then used SVM classifier to diagnose pathological myopia [9][10]. However, these approaches belong to machine learning, which requires manual feature extraction and selection, resulting in relatively high workload.…”

Section: Related Workmentioning

confidence: 99%

Research on pathologic myopia recognition based on vision transformer

Yang

2023

ACE

View full text Add to dashboard Cite

Currently, the diagnosis of pathological myopia is mostly done through manual diagnosis, which not only requires experienced ophthalmologists but is also time-consuming and labour-intensive. In order to improve the diagnostic efficiency and accuracy, and to prevent irreversible visual impairment caused by missed diagnosis, misdiagnosis, and delayed treatment, this paper presents a fine-grained image analysis task of classifying fundus images of patients with pathological myopia and non-pathological myopia. To accurately identify subtle differences in features among similar fundus images, a pathological myopia recognition model based on Vision Transformer (ViT) is proposed. The model incorporates a feature selection module using self-attention mechanism that can effectively select important features in the fundus images, thereby eliminating the influence of irrelevant regions on recognition. Experimental results demonstrate that this method outperforms traditional ViT models, achieving high accuracy in pathological myopia recognition.

show abstract

Section: Related Workmentioning

confidence: 99%