Neural network prediction of 30-day mortality following primary total hip arthroplasty

Safa C. Fassihi; Abhay Mathur; Matthew J. Best; Aaron Z. Chen; Alex Gu; Theodore Quan; Kevin Y. Wang; Chapman Wei; Joshua C. Campbell; Savyasachi C. Thakkar

doi:10.1016/j.jor.2021.11.013

Neural network prediction of 30-day mortality following primary total hip arthroplasty

Safa C. Fassihi, Abhay Mathur, Matthew J. Best, Aaron Z. Chen, Alex Gu, Theodore Quan, Kevin Y. Wang, Chapman Wei, Joshua C. Campbell, Savyasachi C. Thakkar

School of Medicine

Research output: Contribution to journal › Article › peer-review

Abstract

Purpose: The purpose is to utilize an artificial neural network (ANN) model to determine the most important variables in predicting mortality following total hip arthroplasty (THA). Methods: Patients that underwent primary THA were included from a national database. Demographic, preoperative, and intraoperative variables were analyzed based on their contribution to 30-day mortality with the use of an ANN model. Results: The five most important factors in predicting mortality following THA were preoperative international normalized ratio, age, body mass index, operative time, and preoperative hematocrit. Conclusion: ANN modeling represents a novel approach to determining perioperative factors that predict mortality following THA.

Original language	English (US)
Pages (from-to)	91-95
Number of pages	5
Journal	Journal of Orthopaedics
Volume	28
DOIs	https://doi.org/10.1016/j.jor.2021.11.013
State	Published - Nov 1 2021

Keywords

Mortality
Neural network
Risk factors
THA
Total hip arthroplasty

ASJC Scopus subject areas

Orthopedics and Sports Medicine

Access to Document

10.1016/j.jor.2021.11.013

Cite this

@article{bd35e8de01494f28b946e232fcf38b9f,

title = "Neural network prediction of 30-day mortality following primary total hip arthroplasty",

abstract = "Purpose: The purpose is to utilize an artificial neural network (ANN) model to determine the most important variables in predicting mortality following total hip arthroplasty (THA). Methods: Patients that underwent primary THA were included from a national database. Demographic, preoperative, and intraoperative variables were analyzed based on their contribution to 30-day mortality with the use of an ANN model. Results: The five most important factors in predicting mortality following THA were preoperative international normalized ratio, age, body mass index, operative time, and preoperative hematocrit. Conclusion: ANN modeling represents a novel approach to determining perioperative factors that predict mortality following THA.",

keywords = "Mortality, Neural network, Risk factors, THA, Total hip arthroplasty",

author = "Fassihi, {Safa C.} and Abhay Mathur and Best, {Matthew J.} and Chen, {Aaron Z.} and Alex Gu and Theodore Quan and Wang, {Kevin Y.} and Chapman Wei and Campbell, {Joshua C.} and Thakkar, {Savyasachi C.}",

note = "Publisher Copyright: {\textcopyright} 2021 Professor P K Surendran Memorial Education Foundation",

year = "2021",

month = nov,

day = "1",

doi = "10.1016/j.jor.2021.11.013",

language = "English (US)",

volume = "28",

pages = "91--95",

journal = "Journal of Orthopaedics",

issn = "0972-978X",

publisher = "Elsevier BV",

}

TY - JOUR

T1 - Neural network prediction of 30-day mortality following primary total hip arthroplasty

AU - Fassihi, Safa C.

AU - Mathur, Abhay

AU - Best, Matthew J.

AU - Chen, Aaron Z.

AU - Gu, Alex

AU - Quan, Theodore

AU - Wang, Kevin Y.

AU - Wei, Chapman

AU - Campbell, Joshua C.

AU - Thakkar, Savyasachi C.

PY - 2021/11/1

Y1 - 2021/11/1

N2 - Purpose: The purpose is to utilize an artificial neural network (ANN) model to determine the most important variables in predicting mortality following total hip arthroplasty (THA). Methods: Patients that underwent primary THA were included from a national database. Demographic, preoperative, and intraoperative variables were analyzed based on their contribution to 30-day mortality with the use of an ANN model. Results: The five most important factors in predicting mortality following THA were preoperative international normalized ratio, age, body mass index, operative time, and preoperative hematocrit. Conclusion: ANN modeling represents a novel approach to determining perioperative factors that predict mortality following THA.

AB - Purpose: The purpose is to utilize an artificial neural network (ANN) model to determine the most important variables in predicting mortality following total hip arthroplasty (THA). Methods: Patients that underwent primary THA were included from a national database. Demographic, preoperative, and intraoperative variables were analyzed based on their contribution to 30-day mortality with the use of an ANN model. Results: The five most important factors in predicting mortality following THA were preoperative international normalized ratio, age, body mass index, operative time, and preoperative hematocrit. Conclusion: ANN modeling represents a novel approach to determining perioperative factors that predict mortality following THA.

KW - Mortality

KW - Neural network

KW - Risk factors

KW - THA

KW - Total hip arthroplasty

UR - http://www.scopus.com/inward/record.url?scp=85120001386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85120001386&partnerID=8YFLogxK

U2 - 10.1016/j.jor.2021.11.013

DO - 10.1016/j.jor.2021.11.013

M3 - Article

C2 - 34898926

AN - SCOPUS:85120001386

SN - 0972-978X

VL - 28

SP - 91

EP - 95

JO - Journal of Orthopaedics

JF - Journal of Orthopaedics

ER -

Neural network prediction of 30-day mortality following primary total hip arthroplasty

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this