A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation
Purpose: Many semi-automated and fully-automated approaches have been proposed in literature to improve the objectivity of the estimation of conjunctival hyperemia, based on image processing analysis of eyes’ photographs. The purpose is to improve its evaluation using faster fully-automated systems...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/11/7/2978 |
| _version_ | 1827696706036170752 |
|---|---|
| author | Nico Curti Enrico Giampieri Fabio Guaraldi Federico Bernabei Laura Cercenelli Gastone Castellani Piera Versura Emanuela Marcelli |
| author_facet | Nico Curti Enrico Giampieri Fabio Guaraldi Federico Bernabei Laura Cercenelli Gastone Castellani Piera Versura Emanuela Marcelli |
| author_sort | Nico Curti |
| collection | DOAJ |
| description | Purpose: Many semi-automated and fully-automated approaches have been proposed in literature to improve the objectivity of the estimation of conjunctival hyperemia, based on image processing analysis of eyes’ photographs. The purpose is to improve its evaluation using faster fully-automated systems and independent by the human subjectivity. Methods: In this work, we introduce a fully-automated analysis of the redness grading scales able to completely automatize the clinical procedure, starting from the acquired image to the redness estimation. In particular, we introduce a neural network model for the conjunctival segmentation followed by an image processing pipeline for the vessels network segmentation. From these steps, we extract some features already known in literature and whose correlation with the conjunctival redness has already been proved. Lastly, we implemented a predictive model for the conjunctival hyperemia using these features. Results: In this work, we used a dataset of images acquired during clinical practice.We trained a neural network model for the conjunctival segmentation, obtaining an average accuracy of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.94</mn></mrow></semantics></math></inline-formula> and a corresponding IoU score of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.88</mn></mrow></semantics></math></inline-formula> on a test set of images. The set of features extracted on these ROIs is able to correctly predict the Efron scale values with a Spearman’s correlation coefficient of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.701</mn></mrow></semantics></math></inline-formula> on a set of not previously used samples. Conclusions: The robustness of our pipeline confirms its possible usage in a clinical practice as a viable decision support system for the ophthalmologists. |
| first_indexed | 2024-03-10T12:53:13Z |
| format | Article |
| id | doaj.art-63587df70a7340b4b2cab9707f885068 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T12:53:13Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-63587df70a7340b4b2cab9707f8850682023-11-21T12:09:31ZengMDPI AGApplied Sciences2076-34172021-03-01117297810.3390/app11072978A Fully Automated Pipeline for a Robust Conjunctival Hyperemia EstimationNico Curti0Enrico Giampieri1Fabio Guaraldi2Federico Bernabei3Laura Cercenelli4Gastone Castellani5Piera Versura6Emanuela Marcelli7eDIMESLab, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyeDIMESLab, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyOphthalmic Unit, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyOphthalmic Unit, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyeDIMESLab, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyDepartment of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyOphthalmic Unit, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyeDIMESLab, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, 40138 Bologna, ItalyPurpose: Many semi-automated and fully-automated approaches have been proposed in literature to improve the objectivity of the estimation of conjunctival hyperemia, based on image processing analysis of eyes’ photographs. The purpose is to improve its evaluation using faster fully-automated systems and independent by the human subjectivity. Methods: In this work, we introduce a fully-automated analysis of the redness grading scales able to completely automatize the clinical procedure, starting from the acquired image to the redness estimation. In particular, we introduce a neural network model for the conjunctival segmentation followed by an image processing pipeline for the vessels network segmentation. From these steps, we extract some features already known in literature and whose correlation with the conjunctival redness has already been proved. Lastly, we implemented a predictive model for the conjunctival hyperemia using these features. Results: In this work, we used a dataset of images acquired during clinical practice.We trained a neural network model for the conjunctival segmentation, obtaining an average accuracy of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.94</mn></mrow></semantics></math></inline-formula> and a corresponding IoU score of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.88</mn></mrow></semantics></math></inline-formula> on a test set of images. The set of features extracted on these ROIs is able to correctly predict the Efron scale values with a Spearman’s correlation coefficient of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.701</mn></mrow></semantics></math></inline-formula> on a set of not previously used samples. Conclusions: The robustness of our pipeline confirms its possible usage in a clinical practice as a viable decision support system for the ophthalmologists.https://www.mdpi.com/2076-3417/11/7/2978artificial intelligencecomputer aided diagnosiscomputer visionconjunctivahyperemiaEfron scale |
| spellingShingle | Nico Curti Enrico Giampieri Fabio Guaraldi Federico Bernabei Laura Cercenelli Gastone Castellani Piera Versura Emanuela Marcelli A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation Applied Sciences artificial intelligence computer aided diagnosis computer vision conjunctiva hyperemia Efron scale |
| title | A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation |
| title_full | A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation |
| title_fullStr | A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation |
| title_full_unstemmed | A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation |
| title_short | A Fully Automated Pipeline for a Robust Conjunctival Hyperemia Estimation |
| title_sort | fully automated pipeline for a robust conjunctival hyperemia estimation |
| topic | artificial intelligence computer aided diagnosis computer vision conjunctiva hyperemia Efron scale |
| url | https://www.mdpi.com/2076-3417/11/7/2978 |
| work_keys_str_mv | AT nicocurti afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT enricogiampieri afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT fabioguaraldi afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT federicobernabei afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT lauracercenelli afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT gastonecastellani afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT pieraversura afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT emanuelamarcelli afullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT nicocurti fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT enricogiampieri fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT fabioguaraldi fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT federicobernabei fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT lauracercenelli fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT gastonecastellani fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT pieraversura fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation AT emanuelamarcelli fullyautomatedpipelineforarobustconjunctivalhyperemiaestimation |