Key result
A Decision Tree model optimized with the Honey Badger Algorithm correctly classified 90% and 94% of samples for Chronic Venous Insufficiency severity before and one month after yoga sessions.
Why the study?
Non-pharmacological approaches like yoga show promise for conditions like chronic venous disease, and proficient machine learning classifiers were needed to evaluate and classify severity levels.
Can hybrid machine learning algorithms accurately classify Chronic Venous Insufficiency severity based on VCSS before and after yoga sessions?
Can hybrid machine learning algorithms accurately classify Chronic Venous Insufficiency severity based on VCSS before and after yoga sessions?
Hybrid machine learning models, specifically Decision Trees optimized with the Honey Badger Algorithm, can accurately classify the severity of Chronic Venous Insufficiency based on clinical scores before and after yoga interventions.
Supports hybrid ML feasibility for CVI grading in yoga studies; leaves open prospective validation before clinical use.
• Intersection of Healthcare and Technology : The study explores the potential of non-pharmacological interventions like yoga in reducing the urgency of Chronic Venous Insufficiency (CVI), especially during periods of increased stress and sedentary lifestyles, such as the global lockdown due to the COVID-19 pandemic. • Application of Machine Learning in Healthcare : The research demonstrates the effectiveness of machine learning algorithms in healthcare by using predictive modeling and precision medicine to enhance patient care based on data from 100 samples. • Optimized Decision Tree Models : The study introduces Decision Tree (DT) models optimized with the Harris Hawks Optimization (HBA) and Aquila Optimizer (AOA), providing valuable insights into classifying CVI severity levels. • High Accuracy and Precision of DTHB Model : The identified optimal model, DTHB, shows exceptional accuracy and precision, particularly in moderate and severe CVI categories where timely medical intervention is critical. The model correctly classified 90% and 94% of samples according to VCSS-PRE and VCSS-1, respectively. • Foundation for Future Research : The comparative analysis of single and optimized machine learning models in this study sets a foundation for future research, promoting the development of more robust models and a deeper understanding of the relationship between lifestyle practices, mental health, and physiological well-being. Advancements in technology have increased work demands, neglecting individual well-being and causing mental pressure and decreased fitness. The COVID-19 pandemic has worsened this, leading to a surge in psychological stress. Non-pharmacological approaches like yoga are gaining popularity for stress management, showing positive effects on the autonomic nervous system and offering benefits such as improved cardio-respiratory health and metabolic efficiency, as well as positive effects on conditions like Type-2 diabetes, Chronic Venous Disease (CVD), and obesity. This investigation aims to introduce proficiently functioning machine learning classifiers, such as single and hybrid forms of Decision Trees (DT), into the domain of studies within this particular category. The study utilized 2 optimization algorithms, namely the Honey Badger Algorithm (HBA) and the Arithmetic Optimization Algorithm (AOA), to develop hybrid models. Venous Clinical Severity Score (VCSS) levels prior to and one month following yoga sessions were among the variables in the questionnaire that were considered as inputs: effective influences on CVD. The developed prediction models were trained, and their operational capability was tested. The extracted results classified the samples into 4 classes: Absent, Mild, Moderate, and Severe Chronic Venous Insufficiency (CVI). Comparative analysis revealed that DTHB (Decision Tree optimized with Honey Badger Algorithm) with maximum Accuracy and Precision of higher than 90% was the optimal model, especially for classifying patients with Moderate and Severe levels of CVI needing emergency medical action.
No takes yet. Share an insight, caveat, or question.
Pan et al. (2025) studied Chronic Venous Insufficiency (CVI) (n=100). Yoga practices and DTHB machine learning model was evaluated on Classification accuracy of CVI severity levels (VCSS-PRE and VCSS-1). A Decision Tree model optimized with the Honey Badger Algorithm correctly classified 90% and 94% of samples for Chronic Venous Insufficiency severity before and one month after yoga sessions.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: