Clinical Scorecard: Modeling Contextual Recommendations in eCoaching Using Machine Learning, X-AI, and Semantic Ontology

At a Glance

Key Highlights

• Automated system uses weather data to recommend indoor or outdoor activities.
• Decision tree classifier achieved an accuracy of 99.1%.
• Integration of machine learning and semantic ontology for personalized recommendations.
• Addresses barriers to physical activity posed by adverse weather conditions.
• Promotes continuous physical activity regardless of external weather.

Guideline-Based Recommendations

Diagnosis

• Identify sedentary lifestyle and associated health risks.

Management

• Utilize eCoaching systems to provide personalized activity recommendations.

Monitoring & Follow-up

• Track weather conditions and user activity levels for tailored guidance.

Risks

• Inactivity linked to obesity, diabetes, hypertension, and cardiovascular diseases.

Patient & Prescribing Data

Adults, particularly those classified as overweight or leading sedentary lifestyles.

Personalized recommendations can mitigate risks associated with inactivity.

Clinical Best Practices

• Incorporate real-time weather data into physical activity recommendations.
• Utilize machine learning for activity type classification.
• Provide indoor alternatives during adverse weather to encourage year-round activity.

Related Resources & Content

• Source Article