
Researchers aimed to develop a machine learning algorithm that could provide individualized predictions for asthma exacerbations using data from an electronic multi-dose dry powder inhaler (eMDPI) with integrated sensors. In the record, published in the Journal of Asthma and Allergy, the authors reported that their machine learning model was successful, and could reverse the paradigm of reactive treatment in patients with asthma.
The study included 360 adult patients with asthma who used a sensor-equipped eMPDI containing albuterol for rescue therapy for 12 weeks. The eMPDI recorded inhaler use, peak inspiratory flow (PIF), inhalation volume, inhalation duration, and time to PIF.
Machine Learning Model Predicts Asthma Exacerbations
Data from the inhalers were combined with clinical and demographic information to construct the predictive model for impending exacerbations, which was then evaluated by area under the receiver operating characteristic (ROC) curve (AUC).
Among the cohort, 64 patients had a total of 78 exacerbations over the 12-week period. The authors noted increased use of albuterol preceded exacerbations, with a mean of 7.3 (standard deviation, 17.3) inhalations in the 24 hours prior to an exacerbation. Researchers used gradient-boosting trees with data from eMDPIs and baseline patient characteristics to construct the machine learning model.
According to the authors, the model predicted an impending exacerbation over the next 5 days with an AUC of 0.83 (95% CI, 0.77-0.90). In addition, the variable with the highest weight in the model was mean number of daily inhalations over the 4 days prior to the predicted date of exacerbation.
Overall, the authors suggested “this approach may support a shift from reactive care to proactive, preventative, and personalized management of chronic respiratory diseases.”
Read More Recent Research on Asthma