A Machine-Learning Approach to Target Clinical and Biological Features Associated with Sarcopenia: Findings from Northern and Southern Italian Aging Populations

Machine learning-based prediction of sarcopenia in community-dwelling middle-aged and older adults: findings from the CHARLS

BACKGROUND

Sarcopenia is a prominent issue among aging populations and associated with poor health outcomes. This study aimed to examine the predictive value of questionnaire and biomarker data for sarcopenia, and to further develop a user-friendly calculator for community-dwelling middle-aged and older adults.

METHODS

We used two waves (2011 and 2013) of the China Health and Retirement Longitudinal Study (CHARLS) to predict sarcopenia, defined by the Asian Working Group for Sarcopenia 2019 criteria. We restricted the analytical sample to adults aged 45 or above (N = 2934). Five machine learning models were used to construct Q-based (only questionnaire variables), Bio-based (only biomarker variables), and combined (questionnaire plus biomarker variables) models. Area under the receiver operating characteristic curve (AUROC) was used for performance assessment. Temporal external validation was performed based on two datasets from CHARLS. Important predictors were identified by Shapley values and coefficients.

RESULTS

Extreme gradient boosting (XGBoost), considering both questionnaire and biomarker characteristics, emerged as the optimal model, and its AUROC was 0.759 (95% CI: 0.747-0.771) at a decision threshold of 0.20 on the test set. Models also performed well on the external datasets. We found that cognitive function was the most important predictor in both Q-based and combined models, and blood urea nitrogen was the most important predictor in the Bio-based model. Other key predictors included education, haematocrit, total cholesterol, drinking, number of chronic diseases, and instrumental activities of daily living score.

CONCLUSIONS

Our findings offer a potential for early screening and targeted prevention of sarcopenia among middle-aged and older adults in the community setting.

A Patented Dietary Supplement (Hydroxy-Methyl-Butyrate, Carnosine, Magnesium, Butyrate, Lactoferrin) Is a Promising Therapeutic Target for Age-Related Sarcopenia through the Regulation of Gut Permeability: A Randomized Controlled Trial

Adequate diet, physical activity, and dietary supplementation with muscle-targeted food for special medical purposes (FSMP) or dietary supplement (DS) are currently considered fundamental pillars in sarcopenia treatment. The aim of this study is to evaluate the effectiveness of a DS (containing hydroxy-methyl-butyrate, carnosine, and magnesium, for its action on muscle function and protein synthesis and butyrate and lactoferrin for their contribution to the regulation of gut permeability and antioxidant/anti-inflammation activity) on muscle mass (assessed by dual X-ray absorptiometry (DXA)), muscle function (by handgrip test, chair test, short physical performance battery (SPPB) test, and walking speed test), inflammation (tumor necrosis factor-alpha (TNF-a), C-reactive protein (CRP), and visceral adipose tissue (VAT)) and gut axis (by zonulin). A total of 59 participants (age 79.7 ± 4.8 years, body mass index 20.99 ± 2.12 kg/m2) were enrolled and randomly assigned to intervention (n = 30) or placebo (n = 28). The skeletal muscle index (SMI) significantly improved in the supplemented group compared to the placebo one, +1.02 (CI 95%: -0.77; 1.26), p = 0.001; a significant reduction in VAT was observed in the intervention group, -70.91 g (-13.13; -4.70), p = 0.036. Regarding muscle function, all the tests significantly improved (p = 0.001) in the supplemented group compared to the placebo one. CRP, zonulin, and TNF-alpha significantly decreased (p = 0.001) in intervention, compared to placebo, -0.74 mg/dL (CI 95%: -1.30; -0.18), -0.30 ng/mL (CI 95%: -0.37; -0.23), -6.45 pg/mL (CI 95%: -8.71; -4.18), respectively. This DS improves muscle mass and function, and the gut muscle has emerged as a new intervention target for sarcopenia.

Metaheuristic-Based Feature Selection Methods for Diagnosing Sarcopenia with Machine Learning Algorithms

This study explores the efficacy of metaheuristic-based feature selection in improving machine learning performance for diagnosing sarcopenia. Extraction and utilization of features significantly impacting diagnosis efficacy emerge as a critical facet when applying machine learning for sarcopenia diagnosis. Using data from the 8th Korean Longitudinal Study on Aging (KLoSA), this study examines harmony search (HS) and the genetic algorithm (GA) for feature selection. Evaluation of the resulting feature set involves a decision tree, a random forest, a support vector machine, and naïve bayes algorithms. As a result, the HS-derived feature set trained with a support vector machine yielded an accuracy of 0.785 and a weighted F1 score of 0.782, which outperformed traditional methods. These findings underscore the competitive edge of metaheuristic-based selection, demonstrating its potential in advancing sarcopenia diagnosis. This study advocates for further exploration of metaheuristic-based feature selection's pivotal role in future sarcopenia research.

Discovering the Individualized Factors Associated with Sarcopenia and Sarcopenic Obesity Phenotypes-A Machine Learning Approach

The literature shows how sarcopenia often occurs along with different phenotypes based either on the concomitant presence of adipose tissue excess (i.e., sarcopenic obesity, SO), or osteopenia/osteoporosis (osteosarcopenia, OS), or the combination of the two conditions, so-called osteosarcopenic obesity (OSO). This research aimed to assess the prevalence of sarcopenia phenotypes (SO, OS, OSO), their associated risk factors and their health impact in a population of out- and inpatients living in the North of Italy. Male and female subjects aged ≥18 years were enrolled for the study. A blood sample was collected to measure targeted blood makers. A comprehensive anthropometric clinical assessment (height, weight, Body Mass Index, BMI and Dual Energy X-ray Absorptiometry, DXA) was performed to measure ponderal, bone, fat, and muscle status. A total of 1510 individuals participated to the study (females, n = 1100; 72.85%). Sarcopenia was the most prevalent phenotype (17%), followed by osteosarcopenia (14.7%) and sarcopenic obesity. Only 1.9% of the sample was affected by OSO. According to logistic regression analysis, sarcopenia was associated with age, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) (positively) and BMI, Iron (Fe), Total Cholesterol, albumin (%), albumin (g), and gamma proteins (negatively). Sarcopenic obesity was associated with age, ferritin, ESR, CRP (positively) and BMI, Fe, and albumin (%) (negatively). Osteosarcopenia was associated with age, ESR (positively) and BMI, Total Cholesterol, albumin (%), albumin (g), and Ca (negatively). Osteosarcopenic obesity was associated with glycemia and gamma-glutamyl transferase (gGT) (positively). According to random forest analysis, a higher BMI was the most important protective factor for sarcopenia, for sarcopenic obesity (along with Iron) and for osteosarcopenia (along with albumin). Moreover, osteosarcopenic obesity was positively associated with GgT and glycaemia. The possibility of gaining such information, especially in the younger population, could help to prevent the onset of such diseases and best fit the patient's needs, according to a precision-medicine approach.

Multidisciplinary approach to sarcopenia: a narrative review

Sarcopenia is a condition in which muscle mass and strength are decreased and muscle function is impaired. It is an indicator of frailty and loss of independence in older adults. It is also associated with increased physical disability, which increases the risk of falls. As a multifactorial disease, sarcopenia is caused by a combination of factors including aging, hormonal changes, nutritional deficiencies, and physical inactivity. Understanding the underlying pathophysiology of sarcopenia and identifying its different causes is critical to developing effective prevention and treatment strategies. This review summarizes the pathophysiology, consequences, diagnostic methods, and multidisciplinary approaches to sarcopenia.

Plasma pentosidine as a useful biomarker of sarcopenia, low gait speed, and mortality in patients with cirrhosis

Purpose

The accumulation of advanced glycation end products (AGEs) is associated with various diseases and age-related impairments, including loss of muscle mass and function. We investigated the association between plasma pentosidine, which is one of the AGEs, and sarcopenia, low gait speed, and mortality in patients with cirrhosis.

Methods

This retrospective study divided 128 patients with cirrhosis into three groups by 25th and 75th quartiles of baseline plasma pentosidine levels: low (L)-, intermediate (I)-, and high (H)-pentosidine (Pen) groups. Sarcopenia was diagnosed following the Japan Society of Hepatology criteria. Low gait speed was defined as <0.8 m/s. The cumulative survival rates were compared between the three groups. Cox proportional hazards regression analysis was performed to identify independent factors associated with mortality.

Results

Of the 128 patients, 40 (31.3%) and 34 (26.6%) had sarcopenia and low gait speed, respectively. The prevalence of sarcopenia and low gait speed significantly increased stepwise with increasing plasma pentosidine levels, with the highest in the H-Pen group (59.4% [19/32] and 56.3% [18/32], respectively) and lowest in the L-Pen group (18.8% [6/32] and 6.3% [2/32], respectively). Multivariate analysis identified plasma pentosidine levels as a significant and independent factor associated with sarcopenia (odds ratio [OR], 1.07; p = 0.036) and low gait speed (OR, 1.06; p = 0.036), with the cutoff levels of 0.0792 μg/mL (sensitivity/specificity, 0.600/0.773) and 0.0745 μg/mL (sensitivity/specificity, 0.735/0.691), respectively. The cumulative survival rates were significantly lower in the H-Pen group than in the L-Pen (hazard ratio [HR], 11.7; p = 0.001) and I-Pen (HR, 4.03; p < 0.001) groups. Plasma pentosidine levels were identified as a significant and independent prognostic factor (HR, 1.07; p < 0.001).

Conclusion

Plasma pentosidine levels are associated with sarcopenia, low gait speed, and mortality and may serve as a useful surrogate biomarker for these clinical events in patients with cirrhosis.

Prevalence of associations among sarcopenia, obesity, and metabolic syndrome in Brazilian older adults

Background

Although aging is a process associated with the development of obesity, metabolic syndrome (MetS), and sarcopenia, the prevalence of these conditions in older adults from São Paulo, Brazil, is unclear.

Methods

Therefore, the current study aimed to investigate the prevalence of obesity, sarcopenia, and MetS, both separately and together, in a community-based sample of older adults from São Paulo, Brazil. Data from the medical records of 418 older adults of both genders, aged 60  years or older (mean age 69.3 ± 6.5  years), who were not physically active, were used to conduct this retrospective cross-sectional study. Anthropometric variables were used to determine both body mass index (BMI) and Conicity index (C index). Sarcopenia and MetS were defined according to the criteria of the European Working Group on Sarcopenia in Older People and by the Brazilian Society of Endocrinology and Metabolism, respectively.

Results

Based on BMI, the group of older men (n = 91) showed a predominance of adequate weight (n = 49) and the group of older women (n = 327) showed a predominance of obesity (n = 181). In association with obesity, while only the group of older women presented with sarcopenia (n = 5), 52 older women and 9 older men presented with MetS, and two older women presented with sarcopenia + MetS [prevalence ratio = 0.0385, 95% CI (0.007;0.1924)]. Based on the C index, 58 older women and 11 older men presented with MetS, while the occurrence of sarcopenia or MetS + sarcopenia was found in 32 and 5 older women, respectively [prevalence ratio = 0.0910, 95% CI (0.037;0.2241)].

Discussion

Our results suggest that obesity, as measured by BMI or the C Index, was more closely associated with the occurrence of MetS than sarcopenia, regardless of gender, and also that sarcopenic obesity was only found in the group of older women. Additionally, the prevalence ratio of obesity, sarcopenia, and MetS evidenced using the C index was 2.3 times higher than the values found using the BMI classification.

Machine Learning Applications in Sarcopenia Detection and Management: A Comprehensive Survey

This extensive review examines sarcopenia, a condition characterized by a loss of muscle mass, stamina, and physical performance, with a particular emphasis on its detection and management using contemporary technologies. It highlights the lack of global agreement or standardization regarding the definition of sarcopenia and the various techniques used to measure muscle mass, stamina, and physical performance. The distinctive criteria employed by the European Working Group on Sarcopenia in Older People (EWGSOP) and the Asian Working Group for Sarcopenia (AWGSOP) for diagnosing sarcopenia are examined, emphasizing potential obstacles in comparing research results across studies. The paper delves into the use of machine learning techniques in sarcopenia detection and diagnosis, noting challenges such as data accessibility, data imbalance, and feature selection. It suggests that wearable devices, like activity trackers and smartwatches, could offer valuable insights into sarcopenia progression and aid individuals in monitoring and managing their condition. Additionally, the paper investigates the potential of blockchain technology and edge computing in healthcare data storage, discussing models and systems that leverage these technologies to secure patient data privacy and enhance personal health information management. However, it acknowledges the limitations of these models and systems, including inefficiencies in handling large volumes of medical data and the lack of dynamic selection capability. In conclusion, the paper provides a comprehensive summary of current sarcopenia research, emphasizing the potential of modern technologies in enhancing the detection and management of the condition while also highlighting the need for further research to address challenges in standardization, data management, and effective technology use.

Dietary diversity and possible sarcopenia among older people in China: a nationwide population-based study

Background

Sarcopenia is a common geriatric disease. Many dietary factors may contribute to the development of sarcopenia. Few studies have been conducted on dietary diversity and sarcopenia in Chinese older adults. Among a nationwide sample, the objective of this study is to assess the association between the dietary diversity score (DDS) and the prevalence of possible sarcopenia. We considered the different patterns of dietary diversity in relation to possible sarcopenia.

Methods

We conducted this analysis utilizing the cross-sectional data from the 2012, 2014, and 2018 waves of the Chinese longitudinal healthy longevity survey (CLHLS). A standard developed by the Asian Working Group for Sarcopenia 2019 (AWGS2019) was used to assess the possibility of sarcopenia. On the basis of the DDS generated by previous studies, we have constructed four new indicators as follows: total diet, animal-based diet, plant-based diet, and plant-based diet without the consumption of legume products and nuts. We used the generalized estimation equation (GEE) model to evaluate the associations between the DDS of the total diet, animal-based diet, plant-based diet, and plant-based diet without the intake of legume products and nuts and possible sarcopenia. These associations were statistically adjusted for a variety of potential confounders. Sensitivity analysis was performed by excluding some participants who were long-term bedridden, had Alzheimer's disease, or were terminally ill.

Results

The analysis included 6,624 participants (mean age 83.4 years at baseline). In our study, we found that participants with a higher DDS of the total diet (OR = 0.62; 95% CI: 0.51-0.77), animal-based diet (OR = 0.62; 95% CI: 0.49-0.79), and plant-based diet (OR = 0.64;95% CI: 0.51-0.80) were at a lower risk of developing sarcopenia. In sensitivity analyses, the associations remained unchanged.

Conclusion

Taking a diversified diet, including animal foods, may reduce the risk of developing sarcopenia. According to the findings of this study, adopting a diversified diet might reduce the risk of sarcopenia for older adults.