{"title":"QuickQual:轻量级,方便的视网膜图像质量评分与现成的预训练模型","authors":"Justin Engelmann, A. Storkey, M. Bernabeu","doi":"10.48550/arXiv.2307.13646","DOIUrl":null,"url":null,"abstract":"Image quality remains a key problem for both traditional and deep learning (DL)-based approaches to retinal image analysis, but identifying poor quality images can be time consuming and subjective. Thus, automated methods for retinal image quality scoring (RIQS) are needed. The current state-of-the-art is MCFNet, composed of three Densenet121 backbones each operating in a different colour space. MCFNet, and the EyeQ dataset released by the same authors, was a huge step forward for RIQS. We present QuickQual, a simple approach to RIQS, consisting of a single off-the-shelf ImageNet-pretrained Densenet121 backbone plus a Support Vector Machine (SVM). QuickQual performs very well, setting a new state-of-the-art for EyeQ (Accuracy: 88.50% vs 88.00% for MCFNet; AUC: 0.9687 vs 0.9588). This suggests that RIQS can be solved with generic perceptual features learned on natural images, as opposed to requiring DL models trained on large amounts of fundus images. Additionally, we propose a Fixed Prior linearisation scheme, that converts EyeQ from a 3-way classification to a continuous logistic regression task. For this task, we present a second model, QuickQual MEga Minified Estimator (QuickQual-MEME), that consists of only 10 parameters on top of an off-the-shelf Densenet121 and can distinguish between gradable and ungradable images with an accuracy of 89.18% (AUC: 0.9537). Code and model are available on GitHub: https://github.com/justinengelmann/QuickQual . QuickQual is so lightweight, that the entire inference code (and even the parameters for QuickQual-MEME) is already contained in this paper.","PeriodicalId":261989,"journal":{"name":"OMIA@MICCAI","volume":"23 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"QuickQual: Lightweight, convenient retinal image quality scoring with off-the-shelf pretrained models\",\"authors\":\"Justin Engelmann, A. Storkey, M. Bernabeu\",\"doi\":\"10.48550/arXiv.2307.13646\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Image quality remains a key problem for both traditional and deep learning (DL)-based approaches to retinal image analysis, but identifying poor quality images can be time consuming and subjective. Thus, automated methods for retinal image quality scoring (RIQS) are needed. The current state-of-the-art is MCFNet, composed of three Densenet121 backbones each operating in a different colour space. MCFNet, and the EyeQ dataset released by the same authors, was a huge step forward for RIQS. We present QuickQual, a simple approach to RIQS, consisting of a single off-the-shelf ImageNet-pretrained Densenet121 backbone plus a Support Vector Machine (SVM). QuickQual performs very well, setting a new state-of-the-art for EyeQ (Accuracy: 88.50% vs 88.00% for MCFNet; AUC: 0.9687 vs 0.9588). This suggests that RIQS can be solved with generic perceptual features learned on natural images, as opposed to requiring DL models trained on large amounts of fundus images. Additionally, we propose a Fixed Prior linearisation scheme, that converts EyeQ from a 3-way classification to a continuous logistic regression task. For this task, we present a second model, QuickQual MEga Minified Estimator (QuickQual-MEME), that consists of only 10 parameters on top of an off-the-shelf Densenet121 and can distinguish between gradable and ungradable images with an accuracy of 89.18% (AUC: 0.9537). Code and model are available on GitHub: https://github.com/justinengelmann/QuickQual . QuickQual is so lightweight, that the entire inference code (and even the parameters for QuickQual-MEME) is already contained in this paper.\",\"PeriodicalId\":261989,\"journal\":{\"name\":\"OMIA@MICCAI\",\"volume\":\"23 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-07-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"OMIA@MICCAI\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2307.13646\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"OMIA@MICCAI","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2307.13646","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
对于传统的和基于深度学习(DL)的视网膜图像分析方法来说,图像质量仍然是一个关键问题,但是识别质量差的图像既耗时又主观。因此,需要视网膜图像质量评分(RIQS)的自动化方法。目前最先进的是MCFNet,由三个Densenet121主干网组成,每个主干网在不同的色彩空间中运行。MCFNet和由同一作者发布的EyeQ数据集是RIQS向前迈出的一大步。我们提出QuickQual,一种简单的RIQS方法,由一个现成的imagenet预训练的Densenet121骨干加上一个支持向量机(SVM)组成。QuickQual表现非常好,为EyeQ设定了新的技术水平(准确率:88.50% vs MCFNet的88.00%;AUC: 0.9687 vs 0.9588)。这表明RIQS可以通过在自然图像上学习到的一般感知特征来解决,而不需要在大量眼底图像上训练DL模型。此外,我们提出了一个固定先验线性化方案,该方案将EyeQ从3-way分类转换为连续逻辑回归任务。对于这项任务,我们提出了第二个模型,quickquality MEga Minified Estimator (quickquality - meme),它在现有的Densenet121之上仅由10个参数组成,可以区分可分级和不可分级的图像,准确率为89.18% (AUC: 0.9537)。代码和模型可在GitHub: https://github.com/justinengelmann/QuickQual。QuickQual是如此轻量级,以至于本文中已经包含了整个推理代码(甚至QuickQual- meme的参数)。
Image quality remains a key problem for both traditional and deep learning (DL)-based approaches to retinal image analysis, but identifying poor quality images can be time consuming and subjective. Thus, automated methods for retinal image quality scoring (RIQS) are needed. The current state-of-the-art is MCFNet, composed of three Densenet121 backbones each operating in a different colour space. MCFNet, and the EyeQ dataset released by the same authors, was a huge step forward for RIQS. We present QuickQual, a simple approach to RIQS, consisting of a single off-the-shelf ImageNet-pretrained Densenet121 backbone plus a Support Vector Machine (SVM). QuickQual performs very well, setting a new state-of-the-art for EyeQ (Accuracy: 88.50% vs 88.00% for MCFNet; AUC: 0.9687 vs 0.9588). This suggests that RIQS can be solved with generic perceptual features learned on natural images, as opposed to requiring DL models trained on large amounts of fundus images. Additionally, we propose a Fixed Prior linearisation scheme, that converts EyeQ from a 3-way classification to a continuous logistic regression task. For this task, we present a second model, QuickQual MEga Minified Estimator (QuickQual-MEME), that consists of only 10 parameters on top of an off-the-shelf Densenet121 and can distinguish between gradable and ungradable images with an accuracy of 89.18% (AUC: 0.9537). Code and model are available on GitHub: https://github.com/justinengelmann/QuickQual . QuickQual is so lightweight, that the entire inference code (and even the parameters for QuickQual-MEME) is already contained in this paper.