Assessment and risk prediction of frailty using texture-based muscle ultrasound image analysis and machine learning techniques

Aug 5, 2026

Upload by Aurora

Retrospective analysis of prospectively acquired data evaluated texture-based muscle ultrasound image analysis for frailty assessment and risk prediction of frailty phenotype. The study enrolled 101 participants undergoing ultrasound scanning of the anterior thigh and stratified them by frailty status, followed over two years. Forty-three texture features from rectus femoris and vastus intermedius were computed and modeled with machine learning. Performance used ROC AUC, while outcomes included death and comorbidity via regression. Models achieved moderate-to-good discrimination (AUC 0.67–0.79) and classified 70–87% of cases, with associations to comorbidity and mortality risk.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Background and need for standardized frailty definitions

Imaging and ultrasound texture features as biomarkers

Study rationale and objectives

Mechanisms of Ageing and Development 215 (2023) 111860 Contents lists available at ScienceDirect Mechanisms of Ageing and Development journal [homepage:](homepage: www.elsevier.com/locate/mechagedev)[ www.elsevier.com/locate/mechagedev](homepage: www.elsevier.com/locate/mechagedev) | Assessment and risk prediction of frailty using texture-based muscle ultrasound image analysis and machine learning techniques
Rebeca Mir´on-Mombielaa, b, c, 1, Silvia Ruiz-Espa˜na d, 1, David Moratal d, 2, *, Consuelo Borr´as a, e, f, 2, **
a Department of Physiology, Universitat de Val`encia/INCLIVA, Avda. Blasco Ib´˜anez, 15, 46010 Valencia, Spain b Hospital General Universitario de Valencia (HGUV), Valencia, Spain
c Herlev og Gentofte Hospital, Herlev, Denmark
d Center for Biomaterials and Tissue Engineering, Universitat Polit`ecnica de Val`encia, Camí de Vera s/n, 46022 Valencia, Spain e INCLIVA Health Research Institute, Av/ de Men´endez y Pelayo, 4, 46010 Valencia, Spain
f Center for Biomedical Network Research on Frailty and Healthy Aging (CIBERFES), CIBER-ISCIII, Valencia, Spain | | | | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | Keywords:
Frailty
Muscle Ultrasound Machine-learning Texture analysis Image biomarkers | | The purpose of this study was to evaluate texture-based muscle ultrasound image analysis for the assessment and risk prediction of frailty phenotype. This retrospective study of prospectively acquired data included 101 participants who underwent ultrasound scanning of the anterior thigh. Participants were subdivided according to frailty phenotype and were followed up for two years. Primary and secondary outcome measures were death and comorbidity, respectively. Forty-three texture features were computed from the rectus femoris and the vastus intermedius muscles using statistical methods. Model performance was evaluated by computing the area under the receiver operating characteristic curve (AUC) while outcome prediction was evaluated using regression analysis. Models developed achieved a moderate to good AUC (0.67 ≤ AUC ≤ 0.79) for categorizing frailty. The stepwise multiple logistic regression analysis demonstrated that they correctly classified 70–87% of the cases. The models were associated with increased comorbidity (0.01 ≤ p ≤ 0.18) and were predictive of death for pre-frail and frail participants (0.001 ≤ p ≤ 0.016). In conclusion, texture analysis can be useful to identify frailty and assess risk prediction (i.e. mortality) using texture features extracted from muscle ultrasound images in combination with a machine learning approach. | 1. Introduction Over the last few decades, geriatrics and gerontology researchers have devoted an increasing amount of effort to developing and implementing preventive interventions against frailty. The accomplishment of such a task has been hampered by the lack of standardized, and universally agreed definitions for frailty (Rodríguez-Ma˜nas et al., 2013). These definitional ambiguities are also reflected by the absence of reliable biomarkers that can identify frailty, track its progression, and monitor their response to interventions (Calvani et al., 2015). There is a lack of methodology that could be used across many populations to optimize frail patient care, and available diagnostic tools are not fully exploited. Modern imaging techniques have a high potential to help fill this gap and facilitate frailty assessment. Ultrasound echo intensity is an attractive candidate to explore further as a biomarker of frailty (Miron Mombiela et al., 2017), as it can be used to evaluate objectively muscle quality and is relatively cheap (Fukumoto et al., 2012; Watanabe et al., 2013; Akima et al., 2017; Miron Mombiela et al., Abbreviations: AUC, area under the receiver operating characteristic curve; BMI, body mass index; GLCM, gray-level co-occurrence matrix; GLRLM, gray-level runlength matrix; GLSZM, gray-level size-zone matrix; KN

FAQs

How many participants were included and what imaging area was scanned?

The study included 101 participants. Ultrasound scanning focused on the anterior thigh.

What outcomes were predicted over the two-year follow-up?

Primary and secondary outcomes were death and comorbidity, respectively.

How were texture features and model performance evaluated?

Forty-three texture features were computed from specific thigh muscles using statistical methods. Model discrimination used ROC AUC, and outcome prediction used regression analysis.

Aug 5, 2026

Upload by Aurora