Automated detection and categorization of genital injuries using digital colposcopy

Kelwin Fernandes*, Jaime S. Cardoso, Birgitte Schmidt Astrup

*Kontaktforfatter for dette arbejde

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Resumé

Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g. a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Thereby, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we compare traditional handcrafted features and deep learning techniques in the automated processing of colposcopic images for genital injury detection. Positive results where achieved by both paradigms in segmentation and classification subtasks, being traditional and deep models the best strategy for each subtask type respectively.

OriginalsprogEngelsk
TitelPattern Recognition and Image Analysis : 8th Iberian Conference, IbPRIA 2017, Proceedings
RedaktørerLuís A. Alexandre, José Salvador Sánchez, João M. F. Rodrigues
ForlagSpringer
Publikationsdato2017
Sider251-258
ISBN (Trykt)9783319588377
ISBN (Elektronisk)978-3-319-58838-4
DOI
StatusUdgivet - 2017
Begivenhed8th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2017 - Faro, Portugal
Varighed: 20. jun. 201723. jun. 2017

Konference

Konference8th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2017
LandPortugal
ByFaro
Periode20/06/201723/06/2017
SponsorCRESC ALGARVE 2020, Portugal 2020 and FEEI, EVA Hotel, SPIC - Creative Solutions
NavnLecture Notes in Computer Science
Vol/bind10255
ISSN0302-9743

Fingeraftryk

Processing
Deep learning

Citer dette

Fernandes, K., Cardoso, J. S., & Astrup, B. S. (2017). Automated detection and categorization of genital injuries using digital colposcopy. I L. A. Alexandre, J. Salvador Sánchez, & J. M. F. Rodrigues (red.), Pattern Recognition and Image Analysis: 8th Iberian Conference, IbPRIA 2017, Proceedings (s. 251-258). Springer. Lecture Notes in Computer Science, Bind. 10255 https://doi.org/10.1007/978-3-319-58838-4_28
Fernandes, Kelwin ; Cardoso, Jaime S. ; Astrup, Birgitte Schmidt. / Automated detection and categorization of genital injuries using digital colposcopy. Pattern Recognition and Image Analysis: 8th Iberian Conference, IbPRIA 2017, Proceedings. red. / Luís A. Alexandre ; José Salvador Sánchez ; João M. F. Rodrigues. Springer, 2017. s. 251-258 (Lecture Notes in Computer Science, Bind 10255).
@inproceedings{8b6f52cd9c8d4809b237819504cdcfda,
title = "Automated detection and categorization of genital injuries using digital colposcopy",
abstract = "Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g. a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Thereby, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we compare traditional handcrafted features and deep learning techniques in the automated processing of colposcopic images for genital injury detection. Positive results where achieved by both paradigms in segmentation and classification subtasks, being traditional and deep models the best strategy for each subtask type respectively.",
keywords = "Deep learning, Digital colposcopy, Genital injury, Handcrafted features, Image processing",
author = "Kelwin Fernandes and Cardoso, {Jaime S.} and Astrup, {Birgitte Schmidt}",
year = "2017",
doi = "10.1007/978-3-319-58838-4_28",
language = "English",
isbn = "9783319588377",
pages = "251--258",
editor = "{A. Alexandre}, Lu{\'i}s and {Salvador S{\'a}nchez}, Jos{\'e} and {M. F. Rodrigues}, Jo{\~a}o",
booktitle = "Pattern Recognition and Image Analysis",
publisher = "Springer",
address = "Germany",

}

Fernandes, K, Cardoso, JS & Astrup, BS 2017, Automated detection and categorization of genital injuries using digital colposcopy. i L A. Alexandre, J Salvador Sánchez & J M. F. Rodrigues (red), Pattern Recognition and Image Analysis: 8th Iberian Conference, IbPRIA 2017, Proceedings. Springer, Lecture Notes in Computer Science, bind 10255, s. 251-258, 8th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2017, Faro, Portugal, 20/06/2017. https://doi.org/10.1007/978-3-319-58838-4_28

Automated detection and categorization of genital injuries using digital colposcopy. / Fernandes, Kelwin; Cardoso, Jaime S.; Astrup, Birgitte Schmidt.

Pattern Recognition and Image Analysis: 8th Iberian Conference, IbPRIA 2017, Proceedings. red. / Luís A. Alexandre; José Salvador Sánchez; João M. F. Rodrigues. Springer, 2017. s. 251-258 (Lecture Notes in Computer Science, Bind 10255).

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - Automated detection and categorization of genital injuries using digital colposcopy

AU - Fernandes, Kelwin

AU - Cardoso, Jaime S.

AU - Astrup, Birgitte Schmidt

PY - 2017

Y1 - 2017

N2 - Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g. a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Thereby, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we compare traditional handcrafted features and deep learning techniques in the automated processing of colposcopic images for genital injury detection. Positive results where achieved by both paradigms in segmentation and classification subtasks, being traditional and deep models the best strategy for each subtask type respectively.

AB - Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g. a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Thereby, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we compare traditional handcrafted features and deep learning techniques in the automated processing of colposcopic images for genital injury detection. Positive results where achieved by both paradigms in segmentation and classification subtasks, being traditional and deep models the best strategy for each subtask type respectively.

KW - Deep learning

KW - Digital colposcopy

KW - Genital injury

KW - Handcrafted features

KW - Image processing

U2 - 10.1007/978-3-319-58838-4_28

DO - 10.1007/978-3-319-58838-4_28

M3 - Article in proceedings

SN - 9783319588377

SP - 251

EP - 258

BT - Pattern Recognition and Image Analysis

A2 - A. Alexandre, Luís

A2 - Salvador Sánchez, José

A2 - M. F. Rodrigues, João

PB - Springer

ER -

Fernandes K, Cardoso JS, Astrup BS. Automated detection and categorization of genital injuries using digital colposcopy. I A. Alexandre L, Salvador Sánchez J, M. F. Rodrigues J, red., Pattern Recognition and Image Analysis: 8th Iberian Conference, IbPRIA 2017, Proceedings. Springer. 2017. s. 251-258. (Lecture Notes in Computer Science, Bind 10255). https://doi.org/10.1007/978-3-319-58838-4_28