Machine Learning Models for Sarcopenia Identification Based on Radiomic Features of Muscles in Computed Tomography.

Young Jae Kim

doi:10.3390/ijerph18168710

Abstract

The diagnosis of sarcopenia requires accurate muscle quantification. As an alternative to manual muscle mass measurement through computed tomography (CT), artificial intelligence can be leveraged for the automation of these measurements. Although generally difficult to identify with the naked eye, the radiomic features in CT images are informative. In this study, the radiomic features were extracted from L3 CT images of the entire muscle area and partial areas of the erector spinae collected from non-small cell lung carcinoma (NSCLC) patients. The first-order statistics and gray-level co-occurrence, gray-level size zone, gray-level run length, neighboring gray-tone difference, and gray-level dependence matrices were the radiomic features analyzed. The identification performances of the following machine learning models were evaluated: logistic regression, support vector machine (SVM), random forest, and extreme gradient boosting (XGB). Sex, coarseness, skewness, and cluster prominence were selected as the relevant features effectively identifying sarcopenia. The XGB model demonstrated the best performance for the entire muscle, whereas the SVM was the worst-performing model. Overall, the models demonstrated improved performance for the entire muscle compared to the erector spinae. Although further validation is required, the radiomic features presented here could become reliable indicators for quantifying the phenomena observed in the muscles of NSCLC patients, thus facilitating the diagnosis of sarcopenia.

Highlights

Sarcopenia is an illness accompanied by the loss of muscle mass and muscle strength that becomes more prevalent with age
We aimed to find radiomic features that are useful for identifying sarcopenia in computed tomography (CT) images of non-small cell lung cancer (NSCLC) patients and to verify the performance of machine learning in identifying radiomic features for the diagnosis of sarcopenia
CT slices of 247 patients diagnosed with non-small cell lung carcinoma (NSCLC) and subsequently selected the relevant features for the effective identification of sarcopenia

Summary

Introduction

Sarcopenia is an illness accompanied by the loss of muscle mass and muscle strength that becomes more prevalent with age. Sarcopenia is closely associated with injury, decreased functioning, and death [1]. With the accelerated aging of the worldwide population in recent years, the prevalence of sarcopenia is increasing. Sarcopenia is recognized as one of the most serious threats to public health from clinical, social, and economic perspectives [4]. The accurate quantification of muscles is required for the accurate diagnosis of sarcopenia. The quantity and quality of muscles are the main factors to be measured. Muscle quantity is one of the most important indicators for diagnosing sarcopenia. Various methods have been proposed to measure muscle quantity accurately. The most accurate method is to measure the muscle mass manually via computed tomography (CT)

Objectives

Methods

Results

Discussion

Conclusion