Our objective was to test the hypothesis that AI-driven, Personalized Music-Based Intervention (PMBI) is efficacious in reducing anticipatory anxiety. Affective disorders, including anxiety, are more prevalent in neurodegenerative disorders including Alzheimer Disease.

Music interventions have been increasingly backed by high quality research, to the extent that Neurology recently published the “NIH Music-Based Intervention Toolkit” for designing studies on this topic. Music therapy has shown promise for reducing anxiety, though were are not aware of any AI-driven PMBI for one of the most common affective conditions in many developed nations. In this pilot study we employ a PMBI to attenuate anticipatory anxiety prior to examinations for interstitial lung disease (ILD). ILD has been associated with an increased probability of developing dementia.  

Our study involved ten ILD patients with comorbid anxiety, assessed using the Visual Analog Scale (VAS). Individual preferences were assessed using the AI-driven algorithm which then generated a 30-minute listening session for each patient. Anxiety levels were quantified with the VAS before and after each session. IRB approval was obtained.

Nine of the ten patients completed the 30-minute PMBI session. From pre- to post-intervention, there were statistically significant (p=0.02; paired t-test) reductions in anxiety by the VAS. Effect size was large (-0.77; Cohen’s d). No adverse events were observed with respect to heart rate, systolic and diastolic blood pressure, respiratory rate or oxygen saturation.

The results highlight the potential of AI-Driven PMBI as a non-pharmacological anxiolytic tool in this patient population. PMBI offers an innovative solution for alleviating anticipatory anxiety during medical examinations. Drawbacks of this study include small sample size, and poor power to detect adverse effects. Nonetheless, the primary outcome of anxiety reduction was achieved. Future directions include utilizing this foundational knowledge to use AI-based PMBI to study broader patient populations.