Berlin, 24.10.2025 – At the ITG Conference on Speech Communication 2025 in Berlin, Jordan Behrendt presented his paper on using mixed-effects neural networks to improve speech-based predictions of cognitive decline.
Jordan Behrendt attended the ITG Conference on Speech Communication 2025 in Berlin, where he presented his paper “Mixed-Effects Models Neural Networks for Improved Speech-Based Predictions of Cognitive Decline.”
Using data from the Interdisciplinary Longitudinal Study on Adult Development and Aging (ILSE), the study compared standard and mixed-effects versions of linear models, random forests, and neural networks to predict future cognitive test scores over time.
The results show that mixed-effects models outperform standard approaches when current cognitive scores are unavailable, emphasizing the importance of modeling individual variation in longitudinal health data.
