AI Revolutionizes Depression Diagnosis: KTU's Breakthrough Model

**Depression affects 280 million people worldwide**, and traditional diagnostic methods often rely on subjective assessments. To tackle this, researchers at Kaunas University of Technology (KTU) have developed an innovative AI model combining speech and brain activity data for diagnosing depression. This *pioneering method* achieved an impressive 97.53% accuracy, outperforming existing techniques by leveraging the complementary nature of voice and EEG data. KTU's model transforms these signals into spectrograms, which are then analyzed using a modified DenseNet-121 deep-learning model. This approach provides a more accurate depiction of a person's emotional state by detailing brain waveforms and vocal nuances such as speech pace and intonation. Despite these advancements, the AI still needs to learn to explain diagnostic results comprehensively to medical professionals, aligning with the growing emphasis on explainable AI. The groundbreaking work was published in the *Brain Sciences Journal*, paving the way for potentially remote and less subjective depression diagnostics in the future.