Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan

<b>Background:</b> Biological age (BA) is a better representative of health status than chronological age (CA), as it uses different biological markers to quantify cellular and systemic change status. However, BA can be difficult to accurately estimate using current methods. This study u...

Full description

Saved in:
Bibliographic Details
Main Authors: Chun-Feng Chang, Ta-Wei Chu, Chi-Hao Liu, Sheng-Tang Wu, Chung-Chi Yang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/9/1147
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850278786280456192
author Chun-Feng Chang
Ta-Wei Chu
Chi-Hao Liu
Sheng-Tang Wu
Chung-Chi Yang
author_facet Chun-Feng Chang
Ta-Wei Chu
Chi-Hao Liu
Sheng-Tang Wu
Chung-Chi Yang
author_sort Chun-Feng Chang
collection DOAJ
description <b>Background:</b> Biological age (BA) is a better representative of health status than chronological age (CA), as it uses different biological markers to quantify cellular and systemic change status. However, BA can be difficult to accurately estimate using current methods. This study uses multiple adaptive regression spline (MARS) to build an equation to estimate BA among healthy postmenopausal women, thereby potentially improving the efficiency and accuracy of BA assessment. <b>Methods:</b> A total of 11,837 healthy women were enrolled (≥51 years old), excluding participants with metabolic syndrome variable values outside two standard deviations. MARS was applied, with the results compared to traditional multiple linear regression (MLR). The method with the smaller degree of estimation error was considered to be more accurate. The lower prediction errors yielded by MARS compared to the MLR method suggest that MARS performs better than MLR. <b>Results:</b> The equation derived from MARS is depicted. It could be noted that BA could be determined by marriage, systolic blood pressure (SBP), diastolic blood pressure (DBP), waist–hip ratio (WHR), alkaline phosphatase (ALP), lactate dehydrogenase (LDH), creatinine (Cr), carcinoembryonic antigen (CEA), bone mineral density (BMD), education level, and income. The MARS equation is generated. <b>Conclusions:</b> Using MARS, an equation was built to estimate biological age among healthy postmenopausal women in Taiwan. This equation could be used as a reference for calculating BA in general. Our equation showed that the most important factor was BMD, followed by WHR, Cr, marital status, education level, income, CEA, blood pressure, ALP, and LDH.
format Article
id doaj-art-a1be01d421d14280824f460d78c467d9
institution OA Journals
issn 2075-4418
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-a1be01d421d14280824f460d78c467d92025-08-20T01:49:20ZengMDPI AGDiagnostics2075-44182025-04-01159114710.3390/diagnostics15091147Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in TaiwanChun-Feng Chang0Ta-Wei Chu1Chi-Hao Liu2Sheng-Tang Wu3Chung-Chi Yang4Division of Urology, Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei 114202, TaiwanDepartment of Obstetrics and Gynecology, Tri-Service General Hospital, National Defense Medical Center, Taipei 114202, TaiwanDivision of Nephrology, Department of Internal Medicine, Kaohsiung Armed Forces General Hospital, Kaohsiung 802301, TaiwanDivision of Urology, Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei 114202, TaiwanDivision of Cardiology, Department of Medicine, Taoyuan Armed Forces General Hospital, Taoyuan 325208, Taiwan<b>Background:</b> Biological age (BA) is a better representative of health status than chronological age (CA), as it uses different biological markers to quantify cellular and systemic change status. However, BA can be difficult to accurately estimate using current methods. This study uses multiple adaptive regression spline (MARS) to build an equation to estimate BA among healthy postmenopausal women, thereby potentially improving the efficiency and accuracy of BA assessment. <b>Methods:</b> A total of 11,837 healthy women were enrolled (≥51 years old), excluding participants with metabolic syndrome variable values outside two standard deviations. MARS was applied, with the results compared to traditional multiple linear regression (MLR). The method with the smaller degree of estimation error was considered to be more accurate. The lower prediction errors yielded by MARS compared to the MLR method suggest that MARS performs better than MLR. <b>Results:</b> The equation derived from MARS is depicted. It could be noted that BA could be determined by marriage, systolic blood pressure (SBP), diastolic blood pressure (DBP), waist–hip ratio (WHR), alkaline phosphatase (ALP), lactate dehydrogenase (LDH), creatinine (Cr), carcinoembryonic antigen (CEA), bone mineral density (BMD), education level, and income. The MARS equation is generated. <b>Conclusions:</b> Using MARS, an equation was built to estimate biological age among healthy postmenopausal women in Taiwan. This equation could be used as a reference for calculating BA in general. Our equation showed that the most important factor was BMD, followed by WHR, Cr, marital status, education level, income, CEA, blood pressure, ALP, and LDH.https://www.mdpi.com/2075-4418/15/9/1147biological agemachine learningpostmenopausalaging
spellingShingle Chun-Feng Chang
Ta-Wei Chu
Chi-Hao Liu
Sheng-Tang Wu
Chung-Chi Yang
Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
Diagnostics
biological age
machine learning
postmenopausal
aging
title Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
title_full Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
title_fullStr Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
title_full_unstemmed Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
title_short Equation Built by Multiple Adaptive Regression Spline to Estimate Biological Age in Healthy Postmenopausal Women in Taiwan
title_sort equation built by multiple adaptive regression spline to estimate biological age in healthy postmenopausal women in taiwan
topic biological age
machine learning
postmenopausal
aging
url https://www.mdpi.com/2075-4418/15/9/1147
work_keys_str_mv AT chunfengchang equationbuiltbymultipleadaptiveregressionsplinetoestimatebiologicalageinhealthypostmenopausalwomenintaiwan
AT taweichu equationbuiltbymultipleadaptiveregressionsplinetoestimatebiologicalageinhealthypostmenopausalwomenintaiwan
AT chihaoliu equationbuiltbymultipleadaptiveregressionsplinetoestimatebiologicalageinhealthypostmenopausalwomenintaiwan
AT shengtangwu equationbuiltbymultipleadaptiveregressionsplinetoestimatebiologicalageinhealthypostmenopausalwomenintaiwan
AT chungchiyang equationbuiltbymultipleadaptiveregressionsplinetoestimatebiologicalageinhealthypostmenopausalwomenintaiwan