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

Kim, Young Jae

doi:10.3390/ijerph18168710

Cited by 17 publications

(22 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recently, a different AI approach has been tested to assess body composition; Kim used radiomic features extracted from CT images to identify sarcopenia in non-small cell lung carcinoma patients reporting interesting results [ 101 ]. Indeed, the concept of radiomics, understood as the conversion of images to a huge amount of data, could be the future of imaging assessment of sarcopenia.…”

Section: The Role Of Artificial Intelligence (Ai) In Sarcopeniamentioning

confidence: 99%

Sarcopenia: imaging assessment and clinical application

Chianca¹,

Albano

Messina

et al. 2021

Abdom Radiol

View full text Add to dashboard Cite

Sarcopenia is a progressive, generalized skeletal muscle disorder characterized by reduction of muscle mass and strength. It is associated with increased adverse outcomes including falls, fractures, physical disability, and mortality, particularly, in elderly patients. Nowadays, sarcopenia has become a specific imaging biomarker able to predict clinical outcomes of patients. Muscle fibre reduction has shown to be an unfavourable pre-operative predictive factor in patients with cancer, and is associated with worse clinical outcomes in terms of postoperative complications, morbidity, mortality, and lower tolerance of chemoradiation therapy. Several imaging modalities, including dual-energy X-ray absorptiometry, CT, MRI, and US can be used to estimate muscle mass and quality to reach the diagnosis of sarcopenia. This article reviews the clinical implications of sarcopenia, how this condition can be assessed through different imaging modalities, and future perspectives of imaging of sarcopenia.

show abstract

Section: The Role Of Artificial Intelligence (Ai) In Sarcopeniamentioning

confidence: 99%

Sarcopenia: imaging assessment and clinical application

Chianca¹,

Albano

Messina

et al. 2021

Abdom Radiol

View full text Add to dashboard Cite

show abstract

“…Young proposed a new diagnosis method for sarcopenia based on convolutional neural network and radiomics, which proved the feasibility of radiomics in the diagnosis of sarcopenia. However, he did not include muscle strength or physical performance in the sarcopenic auto-diagnosis model, which left their model of sarcopenia somewhat limited ( 14 ). In this study, we propose the concept of “radiomic sarcopenia” using sarcopenia features extracted from three-dimensional imaging.…”

Section: Introductionmentioning

confidence: 99%

Establish a New Diagnosis of Sarcopenia Based on Extracted Radiomic Features to Predict Prognosis of Patients With Gastric Cancer

Chen

Huang

et al. 2022

Front. Nutr.

View full text Add to dashboard Cite

BackgroundPreoperative sarcopenia is a prognostic risk factor for gastric cancer (GC). This study aimed to determine whether radiomic sarcopenia features on computed tomography (CT) could be used to diagnose sarcopenia preoperatively, and whether they could be used to accurately predict the postoperative survival and complication prognosis of patients with GC.MethodsWe retrospectively analyzed data of 550 patients with GC who underwent radical gastrectomy. The patients were divided into training (2014–2016) and validation (2017–2019) cohorts. We established a radiomics-based diagnosis tool for sarcopenia. Thereafter, univariate and multivariate analyses of diagnostic factors were carried out. Receiver operator characteristic (ROC) curves and area under the curve (AUC) were used to compare different diagnostic models. The Kaplan–Meier method was used to estimate the survival curve.ResultsRadiomic sarcopenia correlated with complications and long-term survival. Skeletal muscle index, grip strength, and walking speed were correlated with postoperative complications in both cohorts (AUCs: 0.632, 0.577, and 0.614, respectively in the training cohort; 0.570, 0.605, 0.546, respectively, in the validation cohort), and original sarcopenia was more accurate than any of these indicators. However, radiomic sarcopenia has a higher AUC in predicting short-term complications than original sarcopenia in both groups (AUCs: 0.646 vs. 0.635 in the training cohort; 0.641 vs. 0.625 in the validation cohort). In the training cohort, the overall survival time of patients with original sarcopenia was shorter than normal patients (hazard ratio, HR = 1.741; 95% confidence interval [CI], 1.044–2.903; p = 0.031). While radiomic sarcopenia had a greater prognostic significance, the overall survival time of patients with radiomic sarcopenia was significantly worse than normal patients (HR, 1.880; 95% CI, 1.225–2.885, p = 0.003).ConclusionExtracted sarcopenia features based on CT can predict long-term survival and short-term complications of GC patients after surgery, and its accuracy has been verified by training and validation groups. Compared with original sarcopenia, radiomic sarcopenia can effectively improve the accuracy of survival and complication prediction and also shorten the time and steps of traditional screening, thereby reducing the subjectivity effects of sarcopenia assessment.

show abstract

“…Kim, in contrast, expressed a different viewpoint. The radiomic features were reported to be reliable predictors of sarcopenia in patients with NSCLC by means of various machine-learning algorithms (40). Compared with the present study focused on FIGO stage IB1-IIA2 CC patients, differences in cancer types and substantial heterogeneities among patients might cause the inconsistency between the prior Dutch study (39) and the present analysis.…”

mentioning

confidence: 55%

Development and validation of a radiomics-based nomogram for the prediction of postoperative malnutrition in stage IB1-IIA2 cervical carcinoma

Yu¹,

Chen

et al. 2023

Front. Nutr.

View full text Add to dashboard Cite

ObjectiveIn individuals with stage IB1-IIA2 cervical cancer (CC) who received postoperative radiotherapy ± chemotherapy (PORT/CRT), the interaction between sarcopenia and malnutrition remains elusive, let alone employing a nomogram model based on radiomic features of psoas extracted at the level of the third lumbar vertebra (L3). This study was set to develop a radiomics-based nomogram model to predict malnutrition as per the Patient-Generated Subjective Global Assessment (PG-SGA) for individuals with CC.MethodsIn total, 120 individuals with CC underwent computed tomography (CT) scans before PORT/CRT. The radiomic features of psoas at L3 were obtained from non-enhanced CT images. Identification of the optimal features and construction of the rad-score formula were conducted utilizing the least absolute shrinkage and selection operator (LASSO) logistic regression to predict malnutrition in the training dataset (radiomic model). Identification of the major clinical factors in the clinical model was performed by means of binary logistic regression analysis. The radiomics-based nomogram was further developed by integrating radiomic signatures and clinical risk factors (combined model). The receiver operating characteristic (ROC) curves and decision curves analysis (DCA) were employed for the evaluation and comparison of the three models in terms of their predictive performance.ResultsTwelve radiomic features in total were chosen, and the rad-score was determined with the help of the non-zero coefficient from LASSO regression. Multivariate analysis revealed that besides rad-score, age and Eastern Cooperative Oncology Group performance status could independently predict malnutrition. As per the data of this analysis, a nomogram prediction model was constructed. The area under the ROC curves (AUC) values of the radiomic and clinical models were 0.778 and 0.847 for the training and 0.776 and 0.776 for the validation sets, respectively. An increase in the AUC was observed up to 0.972 and 0.805 in the training and validation sets, respectively, in the combined model. DCA also confirmed the clinical benefit of the combined model.ConclusionThis radiomics-based nomogram model depicted potential for use as a marker for predicting malnutrition in stage IB1-IIA2 CC patients who underwent PORT/CRT and required further investigation with a large sample size.

show abstract

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

Cited by 17 publications

References 29 publications

Sarcopenia: imaging assessment and clinical application

Sarcopenia: imaging assessment and clinical application

Establish a New Diagnosis of Sarcopenia Based on Extracted Radiomic Features to Predict Prognosis of Patients With Gastric Cancer

Development and validation of a radiomics-based nomogram for the prediction of postoperative malnutrition in stage IB1-IIA2 cervical carcinoma

Contact Info

Product

Resources

About