Topological Data Analysis for Eye Fundus Image Quality Assessment
The objective of this work is to perform image quality assessment (IQA) of eye fundus images in the context of digital fundoscopy with topological data analysis (TDA) and machine learning methods. Eye health remains inaccessible for a large amount of the global population. Digital tools that automiz...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/11/8/1322 |
| _version_ | 1827685516790726656 |
|---|---|
| author | Gener José Avilés-Rodríguez Juan Iván Nieto-Hipólito María de los Ángeles Cosío-León Gerardo Salvador Romo-Cárdenas Juan de Dios Sánchez-López Patricia Radilla-Chávez Mabel Vázquez-Briseño |
| author_facet | Gener José Avilés-Rodríguez Juan Iván Nieto-Hipólito María de los Ángeles Cosío-León Gerardo Salvador Romo-Cárdenas Juan de Dios Sánchez-López Patricia Radilla-Chávez Mabel Vázquez-Briseño |
| author_sort | Gener José Avilés-Rodríguez |
| collection | DOAJ |
| description | The objective of this work is to perform image quality assessment (IQA) of eye fundus images in the context of digital fundoscopy with topological data analysis (TDA) and machine learning methods. Eye health remains inaccessible for a large amount of the global population. Digital tools that automize the eye exam could be used to address this issue. IQA is a fundamental step in digital fundoscopy for clinical applications; it is one of the first steps in the preprocessing stages of computer-aided diagnosis (CAD) systems using eye fundus images. Images from the EyePACS dataset were used, and quality labels from previous works in the literature were selected. Cubical complexes were used to represent the images; the grayscale version was, then, used to calculate a persistent homology on the simplex and represented with persistence diagrams. Then, 30 vectorized topological descriptors were calculated from each image and used as input to a classification algorithm. Six different algorithms were tested for this study (SVM, decision tree, k-NN, random forest, logistic regression (LoGit), MLP). LoGit was selected and used for the classification of all images, given the low computational cost it carries. Performance results on the validation subset showed a global accuracy of 0.932, precision of 0.912 for label “quality” and 0.952 for label “no quality”, recall of 0.932 for label “quality” and 0.912 for label “no quality”, AUC of 0.980, F1 score of 0.932, and a Matthews correlation coefficient of 0.864. This work offers evidence for the use of topological methods for the process of quality assessment of eye fundus images, where a relatively small vector of characteristics (30 in this case) can enclose enough information for an algorithm to yield classification results useful in the clinical settings of a digital fundoscopy pipeline for CAD. |
| first_indexed | 2024-03-10T08:53:51Z |
| format | Article |
| id | doaj.art-a8136a305081483987792ac3eec73533 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T08:53:51Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-a8136a305081483987792ac3eec735332023-11-22T07:18:51ZengMDPI AGDiagnostics2075-44182021-07-01118132210.3390/diagnostics11081322Topological Data Analysis for Eye Fundus Image Quality AssessmentGener José Avilés-Rodríguez0Juan Iván Nieto-Hipólito1María de los Ángeles Cosío-León2Gerardo Salvador Romo-Cárdenas3Juan de Dios Sánchez-López4Patricia Radilla-Chávez5Mabel Vázquez-Briseño6Facultad de Ingeniería Arquitectura y Diseño, Universidad Autónoma de Baja California, Carretera Transpeninsular Ensenada-Tijuana #3917, Playitas, Ensenada 22860, MexicoFacultad de Ingeniería Arquitectura y Diseño, Universidad Autónoma de Baja California, Carretera Transpeninsular Ensenada-Tijuana #3917, Playitas, Ensenada 22860, MexicoDirección de Investigación, Innovación y Posgrado, Universidad Politécnica de Pachuca, Carretera Ciudad Sahagún-Pachuca Km. 20, Ex-Hacienda de Santa Bárbara, Hidalgo 43830, MexicoFacultad de Ingeniería Arquitectura y Diseño, Universidad Autónoma de Baja California, Carretera Transpeninsular Ensenada-Tijuana #3917, Playitas, Ensenada 22860, MexicoFacultad de Ingeniería Arquitectura y Diseño, Universidad Autónoma de Baja California, Carretera Transpeninsular Ensenada-Tijuana #3917, Playitas, Ensenada 22860, MexicoEscuela de Ciencias de la Salud, Universidad Autónoma de Baja California, Carretera Transpeninsular S/N, Valle Dorado, Ensenada 22890, MexicoFacultad de Ingeniería Arquitectura y Diseño, Universidad Autónoma de Baja California, Carretera Transpeninsular Ensenada-Tijuana #3917, Playitas, Ensenada 22860, MexicoThe objective of this work is to perform image quality assessment (IQA) of eye fundus images in the context of digital fundoscopy with topological data analysis (TDA) and machine learning methods. Eye health remains inaccessible for a large amount of the global population. Digital tools that automize the eye exam could be used to address this issue. IQA is a fundamental step in digital fundoscopy for clinical applications; it is one of the first steps in the preprocessing stages of computer-aided diagnosis (CAD) systems using eye fundus images. Images from the EyePACS dataset were used, and quality labels from previous works in the literature were selected. Cubical complexes were used to represent the images; the grayscale version was, then, used to calculate a persistent homology on the simplex and represented with persistence diagrams. Then, 30 vectorized topological descriptors were calculated from each image and used as input to a classification algorithm. Six different algorithms were tested for this study (SVM, decision tree, k-NN, random forest, logistic regression (LoGit), MLP). LoGit was selected and used for the classification of all images, given the low computational cost it carries. Performance results on the validation subset showed a global accuracy of 0.932, precision of 0.912 for label “quality” and 0.952 for label “no quality”, recall of 0.932 for label “quality” and 0.912 for label “no quality”, AUC of 0.980, F1 score of 0.932, and a Matthews correlation coefficient of 0.864. This work offers evidence for the use of topological methods for the process of quality assessment of eye fundus images, where a relatively small vector of characteristics (30 in this case) can enclose enough information for an algorithm to yield classification results useful in the clinical settings of a digital fundoscopy pipeline for CAD.https://www.mdpi.com/2075-4418/11/8/1322persistent homologyeye fundus imagestopological data analysisimage quality assessmentcomputational ophthalmology |
| spellingShingle | Gener José Avilés-Rodríguez Juan Iván Nieto-Hipólito María de los Ángeles Cosío-León Gerardo Salvador Romo-Cárdenas Juan de Dios Sánchez-López Patricia Radilla-Chávez Mabel Vázquez-Briseño Topological Data Analysis for Eye Fundus Image Quality Assessment Diagnostics persistent homology eye fundus images topological data analysis image quality assessment computational ophthalmology |
| title | Topological Data Analysis for Eye Fundus Image Quality Assessment |
| title_full | Topological Data Analysis for Eye Fundus Image Quality Assessment |
| title_fullStr | Topological Data Analysis for Eye Fundus Image Quality Assessment |
| title_full_unstemmed | Topological Data Analysis for Eye Fundus Image Quality Assessment |
| title_short | Topological Data Analysis for Eye Fundus Image Quality Assessment |
| title_sort | topological data analysis for eye fundus image quality assessment |
| topic | persistent homology eye fundus images topological data analysis image quality assessment computational ophthalmology |
| url | https://www.mdpi.com/2075-4418/11/8/1322 |
| work_keys_str_mv | AT generjoseavilesrodriguez topologicaldataanalysisforeyefundusimagequalityassessment AT juanivannietohipolito topologicaldataanalysisforeyefundusimagequalityassessment AT mariadelosangelescosioleon topologicaldataanalysisforeyefundusimagequalityassessment AT gerardosalvadorromocardenas topologicaldataanalysisforeyefundusimagequalityassessment AT juandediossanchezlopez topologicaldataanalysisforeyefundusimagequalityassessment AT patriciaradillachavez topologicaldataanalysisforeyefundusimagequalityassessment AT mabelvazquezbriseno topologicaldataanalysisforeyefundusimagequalityassessment |