Oculomics for sarcopenia prediction: a machine learning approach toward predictive, preventive, and personalized medicine

Abstract

Aims: Sarcopenia is characterized by a gradual loss of skeletal muscle mass and strength with increased adverse outcomes. Recently, large-scale epidemiological studies have demonstrated a relationship between several chronic disorders and ocular pathological conditions using an oculomics approach. We hypothesized that sarcopenia can be predicted through eye examinations, without invasive tests or radiologic evaluations in the context of predictive, preventive, and personalized medicine (PPPM/3PM).

Methods: We analyzed data from the Korean National Health and Nutrition Examination Survey (KNHANES). The training set (80%, randomly selected from 2008 to 2010) data were used to construct the machine learning models. Internal (20%, randomly selected from 2008 to 2010) and external (from the KNHANES 2011) validation sets were used to assess the ability to predict sarcopenia. We included 8092 participants in the final dataset. Machine learning models (XGBoost) were trained on ophthalmological examinations and demographic factors to detect sarcopenia.

Results: In the exploratory analysis, decreased levator function (odds ratio [OR], 1.41; P value <0.001), cataracts (OR, 1.31; P value = 0.013), and age-related macular degeneration (OR, 1.38; P value = 0.026) were associated with an increased risk of sarcopenia in men. In women, an increased risk of sarcopenia was associated with blepharoptosis (OR, 1.23; P value = 0.038) and cataracts (OR, 1.29; P value = 0.010). The XGBoost technique showed areas under the receiver operating characteristic curves (AUCs) of 0.746 and 0.762 in men and women, respectively. The external validation achieved AUCs of 0.751 and 0.785 for men and women, respectively. For practical and fast hands-on experience with the predictive model for practitioners who may be willing to test the whole idea of sarcopenia prediction based on oculomics data, we developed a simple web-based calculator application (https://knhanesoculomics.github.io/sarcopenia) to predict the risk of sarcopenia and facilitate screening, based on the model established in this study.

Conclusion: Sarcopenia is treatable before the vicious cycle of sarcopenia-related deterioration begins. Therefore, early identification of individuals at a high risk of sarcopenia is essential in the context of PPPM. Our oculomics-based approach provides an effective strategy for sarcopenia prediction. The proposed method shows promise in significantly increasing the number of patients diagnosed with sarcopenia, potentially facilitating earlier intervention. Through patient oculometric monitoring, various pathological factors related to sarcopenia can be simultaneously analyzed, and doctors can provide personalized medical services according to each cause. Further studies are needed to confirm whether such a prediction algorithm can be used in real-world clinical settings to improve the diagnosis of sarcopenia.

Supplementary information: The online version contains supplementary material available at 10.1007/s13167-022-00292-3.

Keywords: Age-related macular degeneration; Blepharoptosis; Cataract; Demographic factors; Machine learning; Marker patterns; Oculomics; Ophthalmologic examination; Predictive model for practitioners; Predictive preventive personalized medicine (PPPM / 3PM), Sarcopenia; Predictive algorithm; Risk assessment.

Fig. 1

Diagram of the oculomics-based system using ocular measurements to predict sarcopenia. We hypothesized that sarcopenia could be predicted through eye examinations, without invasive tests or radiological evaluations. An artificial intelligence-based approach can incorporate ophthalmologic measurements and clinical factors to analyze disease patterns effectively

Fig. 2

Flowchart of the study population inclusion and exclusion process from the Korean National Health and Nutrition Examination Survey (KNHANES), conducted in 2008–2011. The training set was used to construct the machine learning models. Internal and external validation sets were used to assess the ability of the prospective validation design to predict sarcopenia via chronological splitting. In the field of machine learning, validation is the process in which a trained model is evaluated with a testing dataset

Fig. 3

Prevalence of sarcopenia according to blepharoptosis test including marginal reflex distance 1 and levator muscle function test grading. Severe blepharoptosis and decreased levator function are significantly associated with sarcopenia. Eyelid examination could provide significant predictive markers for sarcopenia

Fig. 4

Distribution of the predicted sarcopenia-related indexes from linear regression with stepwise backward selection against the actual values. A Skeletal index in men. B Sarcopenia index in men. C Skeletal index in women. D Sarcopenia index in women

Fig. 5

SHAP plots representing the importance of the input variables in the XGBoost model to predict sarcopenia. Ocular measurements, including blepharoptosis, decreased levator function, pterygium, cataract, glaucoma, age-related macular degeneration, and diabetic retinopathy, accounted for 15% and 10% of the total SHAP importance in predicting sarcopenia in the male and female populations, respectively

Fig. 6

Prediction performance of machine learning models, age, and body mass index for detecting sarcopenia in the internal and external validation. The machine learning models improved the predictive performance by integrating the various risk factors related to sarcopenia and eye examinations

Fig. 7

Calculator webpage established for sarcopenia risk prediction based on demographic factors and ocular measurements. The application is publicly accessible via a web browser (https://knhanesoculomics.github.io/sarcopenia). This web-based application will facilitate the use of the calculator for clinicians to screen sarcopenia during ophthalmologic examinations

Fig. 8

Scheme of the role of artificial intelligence (AI) in screening sarcopenia via eye examinations for predictive, preventive, and personalized medical care. The proposed method will be a cost-effective framework for predicting sarcopenia at an early stage, as periodic ophthalmologic examinations are emphasized and recommended with the aging of the global population. The oculomics-based method shows promise in significantly increasing the number of patients diagnosed with sarcopenia, potentially facilitating earlier intervention