Shengzhan Wang ^1* Wenyue Shen ² Zhiyuan Gao ^2* Xiaoyu Jiang ³ Yaqi Wang ⁴ Yunxiang Li ^4* Xiaoyu Ma ^4* Wenhao Wang ^1* Shuanghua Xin ^1* Weina Ren ¹ Kai Jin ^2* Juan Ye ^2*

• ¹ Affiliated Hospital, Ningbo University, Ningbo, Zhejiang Province, China
• ² Eye Center, Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China
• ³ College of Control Science and Engineering, Zhejiang University, Hangzhou, Zhejiang Province, China
• ⁴ Communication University of Zhejiang, Hangzhou, Zhejiang Province, China

The assessment of image quality (IQA) plays a pivotal role in the realm of imagebased computer-aided diagnosis techniques, with fundus imaging standing as the primary method for the screening and diagnosis of ophthalmic diseases. Conventional studies on fundus IQA tend to rely on simplistic datasets for evaluation, predominantly focusing on either local or global information, rather than a synthesis of both. Moreover, the interpretability of these studies often lacks compelling evidence. In order to address these issues, this study introduces the Local and Global Attention Aggregated Deep Neural Network (LGAANet), an innovative approach that integrates both local and global information for enhanced analysis.The LGAANet was developed and validated using a Multi-Source Heterogeneous Fundus (MSHF) database, encompassing a diverse collection of images. This dataset includes 802 color fundus photography (CFP) images (302 from portable cameras), and 500 ultrawidefield (UWF) images from 904 patients with diabetic retinopathy (DR) and glaucoma, as well as healthy individuals.This dataset includes normal and pathological images from color fundus photography (CFP), images captured with a portable camera from healthy volunteers, and ultrawide-field (UWF) images from patients with diabetic retinopathy. The assessment of image quality was meticulously carried out by a trio of ophthalmologists, leveraging the human visual system as a benchmark. Furthermore, the model employs attention mechanisms and saliency maps to bolster its interpretability.In testing with the CFP dataset, LGAANet demonstrated remarkable accuracy in three critical dimensions of image quality (illumination, clarity and contrast), recording scores of 0.947, 0.924, and 0.947, respectively. Similarly, when applied to the UWF dataset, the model achieved accuracies of 0.889, 0.913, and 0.923, respectively. These results underscore the efficacy of LGAANet in distinguishing between varying degrees of image quality with high precision.To our knowledge, LGAANet represents the inaugural algorithm trained on an MSHF dataset specifically for fundus IQA, marking a significant milestone in the advancement of computer-aided diagnosis in ophthalmology. This research significantly contributes to the Abstract field, offering a novel methodology for the assessment and interpretation of fundus images in the detection and diagnosis of ocular diseases.