Ksana Health Wins $17.9M to Develop AI for Mental Health and Addiction Treatment

In an ambitious effort to advance the treatment and prevention of mental health disorders and substance use, Ksana Health, based in Eugene, Oregon, has been awarded a $17.9 million contract from the U.S. Department of Health and Human Services (HHS). The goal is to develop a Large Health Behavior Model (LHBM), an AI system that will analyze data from smartphones and wearables to provide continuous, data-driven insights into behavioral health.

This initiative represents a significant step towards transforming how mental health and substance use disorders are managed. Traditional approaches often rely on episodic, subjective assessments, which can be inconsistent and less effective. Ksana Health aims to shift the paradigm by using real-time data from everyday devices to offer continuous monitoring and personalized interventions.

Tony Scripa, Ksana Health’s COO and project co-investigator, emphasized the potential impact of this project: "This initiative augments Ksana’s current efforts to shift behavioral healthcare from episodic, subjective assessment toward continuous, data-driven health promotion. It has the potential to reduce healthcare spending, improve quality of life, and reach populations that currently lack access to effective behavioral health support."

Building a Comprehensive Model

The LHBM will be trained on a vast array of data points collected from smartphones and wearables, including sleep patterns, mobility, and language use. This data will be linked with large-scale electronic health records (EHRs) to create a comprehensive model that can identify early signs of mental health issues and substance use disorders.

Ksana Health is not working alone on this ambitious project. The research effort involves multiple institutions, including Providence Health & Services, MedStar Health, and the University of Washington. Providence and MedStar will lead participant recruitment, while the University of Washington will handle computational aspects.

The project will be executed in phases, starting with a proof-of-concept study and pilot data collection. Once these initial stages are successful, the model will scale to include tens of thousands of participants across multiple health systems. This phased approach ensures that the model is robust and can be effectively applied in real-world settings.

In April, HHS announced the first set of research teams as part of its Advanced Research Projects Agency for Health’s (ARPA-H) Evidence-Based Validation & Innovation for Rapid Therapeutics in Behavioral Health (EVIDENT) initiative. Other participants include Duke University, the Johns Hopkins University School of Medicine, and the Yale Stress Center. These collaborations underscore the interdisciplinary nature of the project and the commitment to rigorous scientific validation.

What Comes Next

The development of the LHBM holds significant promise for improving mental health care, but it also raises important ethical and practical considerations. Privacy is a critical issue, as the model will rely on collecting and analyzing personal data from participants. Ksana Health and its partners must ensure that robust privacy protections are in place to safeguard sensitive information.

The effectiveness of AI models in healthcare depends on the quality and diversity of the data used for training. Ensuring that the LHBM is inclusive and representative of diverse populations will be crucial to avoid biases and ensure equitable access to the benefits of this technology.

As the project progresses, ongoing evaluation and feedback from participants and healthcare providers will be essential to refine the model and address any emerging challenges. The success of the LHBM could pave the way for a new era of mental health care, where continuous, data-driven insights lead to more effective and accessible treatments.

The potential impact of this initiative is profound. By leveraging AI and real-time data, Ksana Health aims to transform behavioral healthcare, making it more responsive, personalized, and accessible. This could not only improve the lives of individuals struggling with mental health issues but also reduce the overall burden on healthcare systems, leading to better outcomes for all.