⚡ Quick Summary
This study explored the use of step count data from smartphones to predict clinical adverse events in cancer patients undergoing systemic treatment. The findings revealed that a decline in daily step counts could accurately forecast hospitalizations within the following week, with a high AUC of 0.88 achieved by the random forest model.
🔍 Key Details
- 📊 Dataset: 76 patients with various cancer types
- 🧩 Features used: Daily step count data from smartphones
- ⚙️ Technology: Machine learning models (Elastic Net, Random Forest, Neural Network)
- 🏆 Performance: Random Forest: AUC 0.88, Neural Network: AUC 0.84, Elastic Net: AUC 0.83
🔑 Key Takeaways
- 📉 Decline in step counts can serve as an early warning for potential hospitalizations.
- 🤖 Machine learning models demonstrated high accuracy in predicting adverse events.
- 🏥 14% of patients experienced unplanned hospitalizations during the study.
- 📅 Step count data was collected continuously over a period of 90 days.
- 🔍 Models struggled to predict treatment modifications or other adverse events.
- 🌟 Study highlights the potential of using everyday technology for health monitoring.
- 📈 AUC values indicate strong predictive capabilities for hospitalizations.
- 👩⚕️ Research conducted by a team of experts in oncology and machine learning.
📚 Background
Monitoring the health of cancer patients undergoing systemic treatment is crucial for timely interventions. Traditional methods of tracking patient well-being often rely on clinical assessments, which may not capture real-time changes in health status. The integration of smartphone technology and machine learning offers a promising avenue for enhancing patient monitoring through continuous data collection.
🗒️ Study
This prospective observational cohort study involved 76 patients receiving systemic anticancer treatment. Researchers monitored physical activity by measuring daily step counts using patients’ smartphones over a 90-day period. The aim was to determine if changes in step counts could predict clinical adverse events, specifically unplanned hospitalizations and treatment modifications, within the following week.
📈 Results
Among the patients analyzed, 14% experienced unplanned hospitalizations. The median step count during the first week of treatment was recorded at 4,303 steps. The machine learning models demonstrated impressive predictive capabilities, with the random forest model achieving an AUC of 0.88, indicating a high level of accuracy in forecasting hospitalizations. However, the models were less effective in predicting treatment modifications or other clinically relevant adverse events.
🌍 Impact and Implications
The findings from this study suggest that monitoring daily step counts can be a valuable tool in proactive patient management for cancer care. By identifying patients at risk of hospitalization early, healthcare providers can implement preventive strategies to mitigate treatment-related complications. This approach not only enhances patient safety but also optimizes healthcare resources, potentially leading to better outcomes in cancer treatment.
🔮 Conclusion
This research underscores the potential of using machine learning and smartphone technology to improve patient monitoring in oncology. The ability to predict hospitalizations based on step count data represents a significant advancement in proactive healthcare management. Future studies should explore the broader applications of this technology in various clinical settings to further enhance patient care.
💬 Your comments
What are your thoughts on using smartphone data for predicting health outcomes in cancer patients? We would love to hear your insights! 💬 Join the conversation in the comments below or connect with us on social media:
Machine Learning-Based Prediction of Clinical Outcomes in Patients With Cancer Receiving Systemic Treatment Using Step Count Data Measured With Smartphones.
Abstract
PURPOSE: This study aimed to investigate whether changes in step count, measured using patients’ own smartphones, could predict a clinical adverse event in the upcoming week in patients undergoing systemic anticancer treatments using machine learning models.
METHODS: This prospective observational cohort study included patients with various cancer types receiving systemic anticancer treatment. Physical activity was monitored continuously using patients’ own smartphones, measuring daily step count for 90 days during treatment. Clinical adverse events (ie, unplanned hospitalizations and treatment modifications) were extracted from medical records. Models predicting adverse events in the upcoming 7 days were created using physical activity data from the preceding 2 weeks. Machine learning models (elastic net [EN], random forest [RF], and neural network [NN]) were trained and validated on a 70:30 split cohort. Model performance was evaluated using the AUC.
RESULTS: Among the 76 patients analyzed (median age 61 [IQR, 53-69] years, 39 [51%] female), 11 (14%) were hospitalized during the study period. The median step count during the first week of systemic treatment was 4,303 [IQR, 1926-7,056]. Unplanned hospitalizations in the upcoming 7 days could be predicted with high accuracy using RF (AUC = 0.88), NN (AUC = 0.84), and EN (AUC = 0.83). The models could not predict treatment modifications (AUC = 0.28-0.51) or the occurrence of any clinically relevant adverse event (AUC = 0.32-0.50).
CONCLUSION: A decline in daily step counts can serve as an early predictor for hospitalizations in the upcoming 7 days, facilitating proactive and preventive toxicity management strategies.
Author: [‘Brouwer CG’, ‘Bartelet BM’, ‘Douma JAJ’, ‘van Doorn L’, ‘Kuip EJM’, ‘Verheul HMW’, ‘Buffart LM’]
Journal: JCO Clin Cancer Inform
Citation: Brouwer CG, et al. Machine Learning-Based Prediction of Clinical Outcomes in Patients With Cancer Receiving Systemic Treatment Using Step Count Data Measured With Smartphones. Machine Learning-Based Prediction of Clinical Outcomes in Patients With Cancer Receiving Systemic Treatment Using Step Count Data Measured With Smartphones. 2025; 9:e2500023. doi: 10.1200/CCI-25-00023