Data-Driven Assessment of Lung Cancer Patients Using Performance Status and Wearable Device Metrics

Area of Science:

• Oncology
• Digital Health
• Biomedical Engineering

Background:

• Lung cancer, particularly non-small cell lung cancer (NSCLC), is a major global cause of cancer mortality.
• Traditional Performance Status (PS) assessments are subjective and lack inter-clinician consistency.
• Objective monitoring of physical activity and PS is vital for personalized lung cancer treatment.

Purpose of the Study:

• To evaluate the usability of wearable devices and a mobile app for monitoring lung cancer patients.
• To explore correlations between clinician-assessed PS and wearable-derived data.
• To investigate the potential of wearable technology in managing lung cancer patient care.

Main Methods:

• A two-phase, non-interventional, observational study (LUPA study) was conducted.
• Phase I utilized a mixed-methods approach to assess usability and data utility.
• Phase II involved 61 lung cancer patients in a clinical observation, correlating wearable data with ECOG-PS scores.

Main Results:

• Moderate correlations were observed between data from wearable devices and clinician-reported ECOG-PS scores.
• Challenges were identified in utilizing machine learning (ML) models for predicting patient condition fluctuations.
• Wearable technology shows promise for objective patient monitoring in lung cancer.

Conclusions:

• Wearable devices offer a potentially objective method for monitoring lung cancer patients' status.
• Refining ML models, increasing sample size, and integrating more wearable data features are crucial for enhanced predictive accuracy.
• Future research should focus on optimizing wearable data integration for improved clinical decision-making in oncology.