Pelvic floor pressure distribution profile in urinary incontinence: a classification study with feature selection

Imagem de Miniatura
Arquivos
art_CARAFINI_Pelvic_floor_pressure_distribution_profile_in_urinary_incontinence_2019.PDF (1.57 MB)
Citações na Scopus
3
Tipo de produção
article
Data de publicação
2019
DOI
Scopus
Web of Science
PubMed
Título da Revista
ISSN da Revista
Título do Volume
Editora
PEERJ INC
Autores
CARAFINI, Adriano
VIEIRA, Marcus Fraga
Citação
PEERJ, v.7, article ID e8207, 22p, 2019
Projetos de Pesquisa
Unidades Organizacionais
Fascículo
Resumo
Background. Pelvic floor pressure distribution profiles, obtained by a novel instrumented non-deformable probe, were used as the input to a feature extraction, selection, and classification approach to test their potential for an automatic diagnostic system for objective female urinary incontinence assessment. We tested the performance of different feature selection approaches and different classifiers, as well as sought to establish the group of features that provides the greatest discrimination capability between continent and incontinent women. Methods. The available data for evaluation consisted of intravaginal spatiotemporal pressure profiles acquired from 24 continent and 24 incontinent women while performing four pelvic floor maneuvers: the maximum contraction maneuver, Valsalva maneuver, endurance maneuver, and wave maneuver. Feature extraction was guided by previous studies on the characterization of pressure profiles in the vaginal canal, where the extracted features were tested concerning their repeatability. Feature selection was achieved through a combination of a ranking method and a complete non-exhaustive subset search algorithm: branch and bound and recursive feature elimination. Three classifiers were tested: k-nearest neighbors (k-NN), support vector machine, and logistic regression. Results. Of the classifiers employed, there was not one that outperformed the others; however, k-NN presented statistical inferiority in one of the maneuvers. The best result was obtained through the application of recursive feature elimination on the features extracted from all the maneuvers, resulting in 77.1% test accuracy, 74.1% precision, and 83.3 recall, using SVM. Moreover, the best feature subset, obtained by observing the selection frequency of every single feature during the application of branch and bound, was directly employed on the classification, thus reaching 95.8% accuracy. Although not at the level required by an automatic system, the results show the potential use of pelvic floor pressure distribution profiles data and provide insights into the pelvic floor functioning aspects that contribute to urinary incontinence.
Palavras-chave
Urinary Incontinence, Pelvic Floor, Pressure Distribution, Classification, Feature Selection, Branch and Bound, k-Nearest Neighbors, Support Vector Machine, Logistic Regression, Recursive Feature Elimination
URI
https://observatorio.fm.usp.br/handle/OPI/34395
Referências
  1. Bo K, 2005, PHYS THER, V85, P269
  2. Bo K, 2004, INT UROGYNECOL J, V15, P76, DOI 10.1007/s00192-004-1125-0
  3. Cacciari LP, 2017, CLIN BIOMECH, V47, P53, DOI 10.1016/j.clinbiomech.2017.05.015
  4. Cacciari LP, 2017, J BIOMECH, V58, P139, DOI 10.1016/j.jbiomech.2017.04.035
  5. Caetano Aletha Silva, 2007, Rev Bras Med Esporte, V13, P270, DOI 10.1590/S1517-86922007000400012
  6. Cawley GC, 2010, J MACH LEARN RES, V11, P2079
  7. Chamochumbi CCM, 2012, BRAZ J PHYS THER, V16, P314, DOI 10.1590/S1413-35552012005000020
  8. DELANCEY JOL, 1988, OBSTET GYNECOL, V72, P296
  9. Devreese A, 2004, NEUROUROL URODYNAM, V23, P190, DOI 10.1002/nau.20018
  10. Duc TT, 1998, GEOCARTO INT, V13, P89, DOI [10.1080/10106049809354645, DOI 10.1080/10106049809354645]
  11. Dumoulin C, 2017, INCONTINENCE, P1443
  12. Guaderrama NM, 2005, NEUROUROL URODYNAM, V24, P243, DOI 10.1002/nau.20112
  13. Hafner M, 2009, LECT NOTES COMPUT SC, V5761, P247
  14. Haylen BT, 2016, INT UROGYNECOL J, V27, P165, DOI 10.1007/s00192-015-2932-1
  15. Kao WC, 2011, EXPERT SYST APPL, V38, P6458, DOI 10.1016/j.eswa.2010.11.100
  16. Krishnan S, 2018, BIOMED SIGNAL PROCES, V43, P41, DOI 10.1016/j.bspc.2018.02.008
  17. Langkvist M, 2014, PATTERN RECOGN LETT, V42, P11, DOI 10.1016/j.patrec.2014.01.008
  18. Laurikkala J, 1999, METHOD INFORM MED, V38, P125
  19. Lenth RV., 2008, ENCY STAT QUAL RELIA, DOI [10.1002/9780470061572.eqr014, DOI 10.1002/9780470061572.EQR014]
  20. Lim R, 2016, J UROLOGY, V196, P153, DOI 10.1016/j.juro.2016.01.090
  21. Miller JM, 2001, OBSTET GYNECOL, V97, P255, DOI 10.1016/S0029-7844(00)01132-7
  22. Milsom I, 2014, EUR UROL, V65, P79, DOI 10.1016/j.eururo.2013.08.031
  23. Morin M, 2004, NEUROUROL URODYNAM, V23, P668, DOI 10.1002/nau.20069
  24. Nambiar AK, 2018, EUR UROL, V73, P596, DOI 10.1016/j.eururo.2017.12.031
  25. Nygaard IE, 2015, AM J OBSTET GYNECOL, V213, DOI 10.1016/j.ajog.2015.01.044
  26. Shaikhina T, 2019, BIOMED SIGNAL PROCES, V52, P456, DOI 10.1016/j.bspc.2017.01.012
  27. Shao ZF, 2016, IEEE T CYBERNETICS, V46, P1939, DOI 10.1109/TCYB.2015.2458177
  28. Shishido K, 2008, J UROLOGY, V179, P1917, DOI 10.1016/j.juro.2008.01.020
  29. Stapor K, 2018, ADV INTELL SYST, V578, P12, DOI 10.1007/978-3-319-59162-9_2
  30. Tamanini JTN, 2003, REV SAUDE PUBL, V37, P203, DOI 10.1590/S0034-89102003000200007
  31. Verelst M, 2007, NEUROUROL URODYNAM, V26, P852, DOI 10.1002/nau.20415
  32. Webb A, 2002, STAT PATTERN RECOGNI
  33. Wyndaele M, 2018, EUR UROL SUPPL, V17, P91, DOI 10.1016/j.eursup.2017.12.001

Coleções
Artigos e Materiais de Revistas Científicas - FM/MFT
Artigos e Materiais de Revistas Científicas - LIM/54