A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images
Correspondence: rodrigo.parra@ua.edu.py; Tel.: +595-981-433-908
Furkejuvvon:
| Váldodahkki: | |
|---|---|
| Eará dahkkit: | , , , , , , , , , |
| Materiálatiipa: | article |
| Giella: | eaŋgalasgiella |
| Almmustuhtton: |
2021
|
| Fáttát: | |
| Liŋkkat: | https://doi.org/10.3390/diagnostics11111951 http://hdl.handle.net/20.500.14066/3791 |
| Fáddágilkorat: |
Eai fáddágilkorat, Lasit vuosttaš fáddágilkora!
|
| _version_ | 1870612072799141888 |
|---|---|
| author | Parra, Rodrigo |
| author2 | Ojeda, Verena Vázquez Noguera, José Luis García Torres, Miguel Mello Román, Julio César Villalba Cardozo, Cynthia Emilia Facon, Jacques Divina, Federico Cardozo, Olivia Castillo, Verónica Elisa Castro Matto, Ingrid |
| author2_role | author author author author author author author author author author |
| author_browse | Cardozo, Olivia Castillo, Verónica Elisa Castro Matto, Ingrid Divina, Federico Facon, Jacques García Torres, Miguel Mello Román, Julio César Ojeda, Verena Parra, Rodrigo Villalba Cardozo, Cynthia Emilia Vázquez Noguera, José Luis |
| author_facet | Parra, Rodrigo Ojeda, Verena Vázquez Noguera, José Luis García Torres, Miguel Mello Román, Julio César Villalba Cardozo, Cynthia Emilia Facon, Jacques Divina, Federico Cardozo, Olivia Castillo, Verónica Elisa Castro Matto, Ingrid |
| author_role | author |
| bitstream.checksum.fl_str_mv | 339176871dd448851fbbd5e2a0b76fb4 858b22fda432bd774e469302988c1974 6ed998122107fb54109c124b552e8a72 6ed998122107fb54109c124b552e8a72 f086f61d89565a23e4555cde09632943 |
| bitstream.checksumAlgorithm.fl_str_mv | MD5 MD5 MD5 MD5 MD5 |
| bitstream.url.fl_str_mv | http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/1/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models%20for%20automatic%20diagnosis%20of%20ocular%20toxoplasmosis%20from%20fundus%20images.pdf http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/2/license.txt http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/3/PINV18-1293art1.pdf.txt http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/5/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models%20for%20automatic%20diagnosis%20of%20ocular%20toxoplasmosis%20from%20fundus%20images.pdf.txt http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/4/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models.jpg |
| dc.contributor.other.es.fl_str_mv | Universidad Americana/INCADE S.A.E |
| dc.creator.none.fl_str_mv | Parra, Rodrigo Ojeda, Verena Vázquez Noguera, José Luis García Torres, Miguel Mello Román, Julio César Villalba Cardozo, Cynthia Emilia Facon, Jacques Divina, Federico Cardozo, Olivia Castillo, Verónica Elisa Castro Matto, Ingrid |
| dc.date.accessioned.none.fl_str_mv | 2022-04-29T23:09:50Z |
| dc.date.available.none.fl_str_mv | 2022-04-29T23:09:50Z |
| dc.date.issued.none.fl_str_mv | 2021-10-21 |
| dc.format.extent.es.fl_str_mv | 15 páginas |
| dc.identifier.citation.en.fl_str_mv | Parra, R., Ojeda, V., Vázquez Noguera, J. L., García Torres, M., Mello Román, J. C., Villalba, C., Facon, J., Divina, F., Cardozo, O., Castillo, V. E., & Castro Matto, I. (2021). A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images. Diagnostics, 11(11), Artículo 1951. https://doi.org/10.3390/diagnostics11111951 |
| dc.identifier.essn.es.fl_str_mv | 2075-4418 |
| dc.identifier.other.es.fl_str_mv | https://doi.org/10.3390/diagnostics11111951 |
| dc.identifier.uri.none.fl_str_mv | http://hdl.handle.net/20.500.14066/3791 |
| dc.language.iso.es.fl_str_mv | eng |
| dc.publisher.en.fl_str_mv | Multidisciplinary Digital Publishing Institute |
| dc.relation.projectCONACYT.es.fl_str_mv | PINV18-1293 |
| dc.rights.*.fl_str_mv | Atribución 4.0 Internacional |
| dc.rights.accessRights.es.fl_str_mv | info:eu-repo/semantics/openAccess |
| dc.rights.uri.*.fl_str_mv | http://creativecommons.org/licenses/by/4.0/ |
| dc.subject.classification.es.fl_str_mv | 7. Salud 7.3. Prevención, vigilancia y control de enfermedades transmisibles y no transmisibles |
| dc.subject.ocde.es.fl_str_mv | 2. Ingeniería y Tecnología 2.2. Ingeniería eléctrica, electrónica [ingeniería eléctrica, electrónica, ingeniería y sistemas de comunicación, ingeniería informática (sólo equipos) y otras disciplinas afines] |
| dc.subject.other.es.fl_str_mv | Deep learning Machine learning interpretability Ocular toxoplasmosis Trust |
| dc.title.es.fl_str_mv | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| dc.type.es.fl_str_mv | info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
| description | Correspondence: rodrigo.parra@ua.edu.py; Tel.: +595-981-433-908 |
| eu_rights_str_mv | openAccess |
| format | article |
| id | CONACYT_e07ae695480ff23673a509eba7ea85fb |
| identifier_str_mv | Parra, R., Ojeda, V., Vázquez Noguera, J. L., García Torres, M., Mello Román, J. C., Villalba, C., Facon, J., Divina, F., Cardozo, O., Castillo, V. E., & Castro Matto, I. (2021). A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images. Diagnostics, 11(11), Artículo 1951. https://doi.org/10.3390/diagnostics11111951 2075-4418 |
| language | eng |
| network_acronym_str | CONACYT |
| network_name_str | Repositorio Institucional CONACYT |
| oai_identifier_str | oai:repositorio.conacyt.gov.py:20.500.14066/3791 |
| publishDate | 2021 |
| publishDateSort | 2021 |
| repository.mail.fl_str_mv | repositorio.institucional@conacyt.gov.py |
| repository.name.fl_str_mv | Repositorio Institucional CONACYT |
| repository_id_str | |
| rights_invalid_str_mv | Atribución 4.0 Internacional http://creativecommons.org/licenses/by/4.0/ |
| spelling | 75d0d64f-0e5d-42e4-ade6-c878e26a025360037a0b11f-e3cb-4093-be02-bf78fd2f85126001144600386292aa-adcb-4977-a6aa-08b85eb571e060011bddad9-f33b-43c9-8169-2cb6feaa29de600117660026eba037-b731-48fb-bfd0-fb206319897d60032a48873-af97-439e-bf26-eb8c2883849b6007c221dbb-5e01-4a6e-b13a-3428060503eb60084782119-1768-4c44-838e-440839f303d8600e22e0e61-768d-40bd-8ba2-641719f6ac1a600Universidad Americana/INCADE S.A.E2022-04-29T23:09:50Z2022-04-29T23:09:50Z2021-10-21Parra, R., Ojeda, V., Vázquez Noguera, J. L., García Torres, M., Mello Román, J. C., Villalba, C., Facon, J., Divina, F., Cardozo, O., Castillo, V. E., & Castro Matto, I. (2021). A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images. Diagnostics, 11(11), Artículo 1951. https://doi.org/10.3390/diagnostics11111951https://doi.org/10.3390/diagnostics11111951http://hdl.handle.net/20.500.14066/37912075-4418Correspondence: rodrigo.parra@ua.edu.py; Tel.: +595-981-433-908This article belongs to the Special Issue Eye Diseases: Diagnosis and Management.In the automatic diagnosis of ocular toxoplasmosis (OT), Deep Learning (DL) has arisen as a powerful and promising approach for diagnosis. However, despite the good performance of the models, decision rules should be interpretable to elicit trust from the medical community. Therefore, the development of an evaluation methodology to assess DL models based on interpretability methods is a challenging task that is necessary to extend the use of AI among clinicians. In this work, we propose a novel methodology to quantify the similarity between the decision rules used by a DL model and an ophthalmologist, based on the assumption that doctors are more likely to trust a prediction that was based on decision rules they can understand. Given an eye fundus image with OT, the proposed methodology compares the segmentation mask of OT lesions labeled by an ophthalmologist with the attribution matrix produced by interpretability methods. Furthermore, an open dataset that includes the eye fundus images and the segmentation masks is shared with the community. The proposal was tested on three different DL architectures. The results suggest that complex models tend to perform worse in terms of likelihood to be trusted while achieving better results in sensitivity and specificity.Consejo Nacional de Ciencia y TecnologíaPrograma Paraguayo para el Desarrollo de la Ciencia y Tecnología. Proyectos de investigación y desarrollo15 páginasengMultidisciplinary Digital Publishing InstituteAtribución 4.0 Internacionalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccess© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).7. Salud7.3. Prevención, vigilancia y control de enfermedades transmisibles y no transmisiblesDeep learningMachine learning interpretabilityOcular toxoplasmosisTrust2. Ingeniería y Tecnología2.2. Ingeniería eléctrica, electrónica [ingeniería eléctrica, electrónica, ingeniería y sistemas de comunicación, ingeniería informática (sólo equipos) y otras disciplinas afines]A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus imagesinfo:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersion11DiagnosticsPINV18-129311Parra, RodrigoOjeda, VerenaVázquez Noguera, José LuisGarcía Torres, MiguelMello Román, Julio CésarVillalba Cardozo, Cynthia EmiliaFacon, JacquesDivina, FedericoCardozo, OliviaCastillo, Verónica ElisaCastro Matto, IngridORIGINALA trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images.pdfA trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images.pdfArtículo científicoapplication/pdf5065931http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/1/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models%20for%20automatic%20diagnosis%20of%20ocular%20toxoplasmosis%20from%20fundus%20images.pdf339176871dd448851fbbd5e2a0b76fb4MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-81698http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/2/license.txt858b22fda432bd774e469302988c1974MD52TEXTPINV18-1293art1.pdf.txtPINV18-1293art1.pdf.txtExtracted texttext/plain40698http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/3/PINV18-1293art1.pdf.txt6ed998122107fb54109c124b552e8a72MD53A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images.pdf.txtA trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images.pdf.txtExtracted texttext/plain40698http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/5/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models%20for%20automatic%20diagnosis%20of%20ocular%20toxoplasmosis%20from%20fundus%20images.pdf.txt6ed998122107fb54109c124b552e8a72MD55THUMBNAILA trust-based methodology to evaluate deep learning models.jpgA trust-based methodology to evaluate deep learning models.jpgVista de primera página de artículo científicoimage/jpeg947964http://repositorio.conacyt.gov.py/bitstream/20.500.14066/3791/4/A%20trust-based%20methodology%20to%20evaluate%20deep%20learning%20models.jpgf086f61d89565a23e4555cde09632943MD5420.500.14066/3791oai:repositorio.conacyt.gov.py:20.500.14066/37912026-02-12 19:30:32.294Repositorio Institucional CONACYTrepositorio.institucional@conacyt.gov.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 |
| spellingShingle | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images Parra, Rodrigo 7. Salud 7.3. Prevención, vigilancia y control de enfermedades transmisibles y no transmisibles Deep learning Machine learning interpretability Ocular toxoplasmosis Trust 2. Ingeniería y Tecnología 2.2. Ingeniería eléctrica, electrónica [ingeniería eléctrica, electrónica, ingeniería y sistemas de comunicación, ingeniería informática (sólo equipos) y otras disciplinas afines] |
| status_str | publishedVersion |
| title | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| title_full | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| title_fullStr | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| title_full_unstemmed | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| title_short | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| title_sort | A trust-based methodology to evaluate deep learning models for automatic diagnosis of ocular toxoplasmosis from fundus images |
| topic | 7. Salud 7.3. Prevención, vigilancia y control de enfermedades transmisibles y no transmisibles Deep learning Machine learning interpretability Ocular toxoplasmosis Trust 2. Ingeniería y Tecnología 2.2. Ingeniería eléctrica, electrónica [ingeniería eléctrica, electrónica, ingeniería y sistemas de comunicación, ingeniería informática (sólo equipos) y otras disciplinas afines] |
| url | https://doi.org/10.3390/diagnostics11111951 http://hdl.handle.net/20.500.14066/3791 |