John T. Nardini, Charles W. J. Pugh, Helen M. Byrne
{"title":"统计和拓扑总结有助于疾病检测的分割视网膜血管图像","authors":"John T. Nardini, Charles W. J. Pugh, Helen M. Byrne","doi":"10.1111/micc.12799","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Objective</h3>\n \n <p>Disease complications can alter vascular network morphology and disrupt tissue functioning. Microvascular diseases of the retina are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>We compute 13 separate descriptor vectors (5 statistical, 8 topological) to summarize the morphology of retinal vessel segmentation images and train support vector machines to predict each image's disease classification from the summary vectors. We assess the performance of each descriptor vector, using five-fold cross validation to estimate their accuracy. We apply these methods to four datasets that were assembled from four existing data repositories; three datasets contain segmented retinal vascular images from one of the repositories, whereas the fourth “All” dataset combines images from four repositories.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>Among the 13 total descriptor vectors considered, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels. On the combined “All” dataset, the Box-counting vector outperforms all other descriptors, including the topological Flooding vector which is sensitive to differences in the annotation styles between the different datasets.</p>\n </section>\n \n <section>\n \n <h3> Conclusion</h3>\n \n <p>Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease and assessing their current limitations. These methods could be incorporated into automated disease assessment tools.</p>\n </section>\n </div>","PeriodicalId":18459,"journal":{"name":"Microcirculation","volume":"30 4","pages":""},"PeriodicalIF":1.9000,"publicationDate":"2023-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Statistical and topological summaries aid disease detection for segmented retinal vascular images\",\"authors\":\"John T. Nardini, Charles W. J. Pugh, Helen M. Byrne\",\"doi\":\"10.1111/micc.12799\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div>\\n \\n \\n <section>\\n \\n <h3> Objective</h3>\\n \\n <p>Disease complications can alter vascular network morphology and disrupt tissue functioning. Microvascular diseases of the retina are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Methods</h3>\\n \\n <p>We compute 13 separate descriptor vectors (5 statistical, 8 topological) to summarize the morphology of retinal vessel segmentation images and train support vector machines to predict each image's disease classification from the summary vectors. We assess the performance of each descriptor vector, using five-fold cross validation to estimate their accuracy. We apply these methods to four datasets that were assembled from four existing data repositories; three datasets contain segmented retinal vascular images from one of the repositories, whereas the fourth “All” dataset combines images from four repositories.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Results</h3>\\n \\n <p>Among the 13 total descriptor vectors considered, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels. On the combined “All” dataset, the Box-counting vector outperforms all other descriptors, including the topological Flooding vector which is sensitive to differences in the annotation styles between the different datasets.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Conclusion</h3>\\n \\n <p>Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease and assessing their current limitations. These methods could be incorporated into automated disease assessment tools.</p>\\n </section>\\n </div>\",\"PeriodicalId\":18459,\"journal\":{\"name\":\"Microcirculation\",\"volume\":\"30 4\",\"pages\":\"\"},\"PeriodicalIF\":1.9000,\"publicationDate\":\"2023-01-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Microcirculation\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1111/micc.12799\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"HEMATOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Microcirculation","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/micc.12799","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"HEMATOLOGY","Score":null,"Total":0}
Statistical and topological summaries aid disease detection for segmented retinal vascular images
Objective
Disease complications can alter vascular network morphology and disrupt tissue functioning. Microvascular diseases of the retina are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images.
Methods
We compute 13 separate descriptor vectors (5 statistical, 8 topological) to summarize the morphology of retinal vessel segmentation images and train support vector machines to predict each image's disease classification from the summary vectors. We assess the performance of each descriptor vector, using five-fold cross validation to estimate their accuracy. We apply these methods to four datasets that were assembled from four existing data repositories; three datasets contain segmented retinal vascular images from one of the repositories, whereas the fourth “All” dataset combines images from four repositories.
Results
Among the 13 total descriptor vectors considered, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels. On the combined “All” dataset, the Box-counting vector outperforms all other descriptors, including the topological Flooding vector which is sensitive to differences in the annotation styles between the different datasets.
Conclusion
Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease and assessing their current limitations. These methods could be incorporated into automated disease assessment tools.
期刊介绍:
The journal features original contributions that are the result of investigations contributing significant new information relating to the vascular and lymphatic microcirculation addressed at the intact animal, organ, cellular, or molecular level. Papers describe applications of the methods of physiology, biophysics, bioengineering, genetics, cell biology, biochemistry, and molecular biology to problems in microcirculation.
Microcirculation also publishes state-of-the-art reviews that address frontier areas or new advances in technology in the fields of microcirculatory disease and function. Specific areas of interest include: Angiogenesis, growth and remodeling; Transport and exchange of gasses and solutes; Rheology and biorheology; Endothelial cell biology and metabolism; Interactions between endothelium, smooth muscle, parenchymal cells, leukocytes and platelets; Regulation of vasomotor tone; and Microvascular structures, imaging and morphometry. Papers also describe innovations in experimental techniques and instrumentation for studying all aspects of microcirculatory structure and function.