Improving preoperative risk-of-death prediction in surgery congenital heart defects using artificial intelligence model: A pilot study
Carregando...
Citações na Scopus
22
Tipo de produção
article
Data de publicação
2020
Título da Revista
ISSN da Revista
Título do Volume
Editora
PUBLIC LIBRARY SCIENCE
Autores
CHANG JUNIOR, Joao
BINUESA, Fabio
DOSSOU, Paul-Eric
Citação
PLOS ONE, v.15, n.9, article ID e0238199, 21p, 2020
Resumo
Background Congenital heart disease accounts for almost a third of all major congenital anomalies. Congenital heart defects have a significant impact on morbidity, mortality and health costs for children and adults. Research regarding the risk of pre-surgical mortality is scarce. Objectives Our goal is to generate a predictive model calculator adapted to the regional reality focused on individual mortality prediction among patients with congenital heart disease undergoing cardiac surgery. Methods Two thousand two hundred forty CHD consecutive patients' data from InCor's heart surgery program was used to develop and validate the preoperative risk-of-death prediction model of congenital patients undergoing heart surgery. There were six artificial intelligence models most cited in medical references used in this study: Multilayer Perceptron (MLP), Random Forest (RF), Extra Trees (ET), Stochastic Gradient Boosting (SGB), Ada Boost Classification (ABC) and Bag Decision Trees (BDT). Results The top performing areas under the curve were achieved using Random Forest (0.902). Most influential predictors included previous admission to ICU, diagnostic group, patient's height, hypoplastic left heart syndrome, body mass, arterial oxygen saturation, and pulmonary atresia. These combined predictor variables represent 67.8% of importance for the risk of mortality in the Random Forest algorithm. Conclusions The representativeness of ""hospital death"" is greater in patients up to 66 cm in height and body mass index below 13.0 for InCor's patients. The proportion of ""hospital death"" declines with the increased arterial oxygen saturation index. Patients with prior hospitalization before surgery had higher ""hospital death"" rates than who did not required such intervention. The diagnoses groups having the higher fatal outcomes probability are aligned with the international literature. A web application is presented where researchers and providers can calculate predicted mortality based on the CgntSCORE on any web browser or smartphone.
Palavras-chave
Referências
- [Anonymous], 2011, J AM COLL CARDIOL, V58, pe44, DOI 10.1016/j.jacc.2011.08.007
- [Anonymous], 2017, KIDNEY INT S, V7, P1, DOI 10.1016/j.kisu.2017.04.001
- [Anonymous], 2016, EVIDENCE, V8, pe1000
- Awad A, 2017, INT J MED INFORM, V108, P185, DOI 10.1016/j.ijmedinf.2017.10.002
- Awad A, 2017, HEALTH SERV MANAG RE, V30, P105, DOI 10.1177/0951484817696212
- Bader-El-Den M, 2016, IEEE IJCNN, P3584, DOI 10.1109/IJCNN.2016.7727660
- BENJAMIN E.J., 2017, HEART DIS STROKE STA
- Blagus R, 2015, BMC BIOINFORMATICS, V16, DOI 10.1186/s12859-015-0723-9
- Browne MW, 2000, J MATH PSYCHOL, V44, P108, DOI 10.1006/jmps.1999.1279
- Carmona F, 2017, BRAZ J CARDIOVA SURG, V32, P260, DOI 10.21470/1678-9741-2016-0074
- Chawla NV, 2002, J ARTIF INTELL RES, V16, P321, DOI 10.1613/jair.953
- CHURPEK M. M., 2017, CRIT CARE MED, V44, P298
- DAVIS J., 2006, RELATIONSHIP PRECISI
- Fawcett T., 2004, MACH LEARN, V31, P1, DOI 10.1016/J.PATREC.2005.10.010
- Hajian-Tilaki K, 2013, CASP J INTERN MED, V4, P627
- Hastie T., 2009, ELEMENTS STAT LEARNI
- Hawkins DM, 2004, J CHEM INF COMP SCI, V44, P1, DOI 10.1021/ci0342472
- Haykin S, 2001, ADAPT LEARN SYST SIG, P1
- HoRER J., 2016, WORLD J PEDIAT CONGE
- Jacobs JP, 2019, ANN THORAC SURG, V108, P558, DOI 10.1016/j.athoracsur.2019.01.069
- Jacobs JP, 2018, WORLD J PEDIATR CONG, V9, P392, DOI 10.1177/2150135118775962
- JENKINS K. J., 2002, J THORACIC CARDIOVAS
- Jenkins Kathy J, 2004, Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu, V7, P180, DOI 10.1053/j.pcsu.2004.02.009
- Jenkins KJ, 2002, J THORAC CARDIOV SUR, V124, P97, DOI 10.1067/mtc.2002.122311
- Kohavi R., 1995, INT JOINT C ART INT
- Lacour-Gayet F, 2004, EUR J CARDIO-THORAC, V25, P911, DOI 10.1016/j.ejcts.2004.03.027
- LI H., 2017, MED US
- Lin K, 2019, INT J MED INFORM, V125, P55, DOI 10.1016/j.ijmedinf.2019.02.002
- McParlin C, 2010, BMC PREGNANCY CHILDB, V10, DOI 10.1186/1471-2393-10-76
- Mitchell T., 1997, MACHINE LEARNING
- Nina Rachel Vilela de Abreu Haickel, 2007, Braz. J. Cardiovasc. Surg., V22, P425, DOI 10.1590/S0102-76382007000400008
- O M, 2015, J INTENSIVE CARE, V3, P42
- Provost F, 2001, MACH LEARN, V42, P203, DOI 10.1023/A:1007601015854
- RUIZ-FERNaNDEZ D., 2016, COMPUTER METHODS PRO
- Saito T, 2015, PLOS ONE, V10, DOI 10.1371/journal.pone.0118432
- SILVA I. N. DA, 2010, REDES NEURAIS ARTIFI
- Wen SP, 2019, IEEE T SYST MAN CY-S, V49, P1787, DOI 10.1109/TSMC.2018.2825021
- XU J., DATA MINING ICU MORT, V16