Forecasting depressive relapse in Bipolar Disorder from clinical data

Imagem de Miniatura
Arquivos
art_BORGES-JUNIOR_Forecasting_depressive_relapse_in_Bipolar_Disorder_from_clinical_2018.PDF (104.34 KB)
Citações na Scopus
6
Tipo de produção
conferenceObject
Data de publicação
2018
Scopus
Web of Science
Título da Revista
ISSN da Revista
Título do Volume
Editora
IEEE
Autores
BORGES-JUNIOR, Renato
SALVINI, Rogerio
NIERENBERG, Andrew A.
SACHS, Gary S.
Citação
PROCEEDINGS 2018 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM), p.613-616, 2018
Projetos de Pesquisa
Unidades Organizacionais
Fascículo
Resumo
Bipolar disorder (BD) is a mood disorder characterized by recurrent episodes of depression and mania/hypomania. Depressive relapse in BD reach rates close to 50% in 1 year and 70% in up to 4 years of treatment. Several studies have been developed to discover more efficient treatments for BD and prevent relapses. However, most of relapse studies used only statistical methods. We aim to analyze the performance of machine learning algorithms in predicting depressive relapse using only clinical data from patients. Five well-used machine learning algorithms (Support Vector Machines, Random Forests, Naive Bayes and Multilayer Perceptron) were applied to the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) dataset of a cohort of 800 patients who became euthymic during the study and were followed up for 1 year: 507 presented a depressive relapse and 293 did not. The algorithms showed reasonable performance in the prediction task, ranging from 61% to 80% in the F-measure. Random Forest algorithm had a higher average of performance (Relapse Group 68%; No Relapse Group 74%), although, the performance between classifiers showed no significant difference. Random Forest analysis demonstrated that the three most important mood symptoms observed were: interest, depression mood and energy. Results show that the machine learning algorithms could be seen as a sensible approach to better support medical decision-making in the BD treatment and prevention of future relapses.
Palavras-chave
bipolar disorder, mental health, depressive relapse, machine learning, artificial intelligence
URI
https://observatorio.fm.usp.br/handle/OPI/31462
Referências
  1. Chawla NV, 2002, J ARTIF INTELL RES, V16, P321, DOI 10.1613/jair.953
  2. Deckersbach T., 2016, J AFFECTIVE DISORDER
  3. Dem J., 2006, J MACH LEARN RES, V7, P1
  4. Ferrari AJ, 2016, BIPOLAR DISORD, V18, P440, DOI 10.1111/bdi.12423
  5. Goldstein BA, 2017, J AM MED INFORM ASSN, V24, P198, DOI 10.1093/jamia/ocw042
  6. Goodwin FK, 2007, MANIC DEPRESSIVE ILL, V1
  7. Hall M., 2009, SIGKDD EXPLORATIONS, V11, P10, DOI [10.1145/1656274.1656278, DOI 10.1145/1656274.1656278]
  8. Kitchenham B., 2004, KEELE U TECH REP, V33, P1
  9. Kohavi R., 1995, IJCAI-95. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, P1137
  10. Librenza-Garcia D, 2017, NEUROSCI BIOBEHAV R, V80, P538, DOI 10.1016/j.neubiorev.2017.07.004
  11. Perlis RH, 2006, AM J PSYCHIAT, V163, P217, DOI 10.1176/appi.ajp.163.2.217
  12. Perlis RH, 2009, BIPOLAR DISORD, V11, P391, DOI 10.1111/j.1399-5618.2009.00686.x
  13. Sachs GS, 2003, BIOL PSYCHIAT, V53, P1028, DOI 10.1016/S0006-3223(03)00165-3
  14. Salvini R, 2015, STUD HEALTH TECHNOL, V216, P741, DOI 10.3233/978-1-61499-564-7-741
  15. Shim IH, 2015, J AFFECT DISORDERS, V173, P120, DOI 10.1016/j.jad.2014.10.061
  16. Vazquez GH, 2015, EUR NEUROPSYCHOPHARM, V25, P1501, DOI 10.1016/j.euroneuro.2015.07.013

Coleções
Comunicações em Eventos - FM/MPS
Comunicações em Eventos - HC/IPq
Comunicações em Eventos - LIM/21