California-based startup Kintsugi has spent the last seven years developing AI technology aimed at detecting signs of depression and anxiety through analyzing speech patterns. However, the company has encountered a setback as it failed to obtain FDA clearance in time, leading to its decision to shut down. Despite this challenge, Kintsugi is releasing most of its technology as open-source, potentially paving the way for new innovations in AI-driven mental health solutions.
Traditional mental health assessments heavily rely on patient questionnaires and clinical interviews, lacking the objective data that lab tests provide in physical medicine. Kintsugi’s approach focuses on analyzing how speech is delivered rather than just the content, leveraging indicators like pauses, sentence structure, and speed to identify potential mental health issues. While the exact features driving the AI’s predictions remain undisclosed, peer-reviewed research has shown promising results comparable to established self-report screening tools.
By positioning its technology as a complementary or alternative solution to existing self-reported screening tools, Kintsugi aimed to offer a more objective and potentially more accessible means of identifying mental health concerns. This approach could help address the limitations of current screening methods, which rely on patient accuracy and may not capture the full spectrum of symptoms associated with mental health disorders.
Source: The Verge