Microsoft collaborates with SilverCloud Health to develop AI for improved mental health

Published

By , Principal Research Manager , Principal Researcher

(opens in new tab)

One in four of us, at some point in our lives, will be affected by a mental health condition. Good mental health and well-being are fundamental to our general health and quality of life. It enables us to build resilience against everyday stresses, to work productively, to have fulfilling relationships, and to experience life as meaningful. In recent years, the topic of mental health has become a major concern to society due to increases in the occurrence of mental illness, particularly amongst young people, and the devastating effects it has on both the individual, their families, and others around them.

Current in-person clinical treatments are often labor-intensive, not scalable, and expensive for clinical organizations, with overly long waiting times for giving patients access to the treatments they need. This increasing global burden of mental illness has made the prevention and treatment of mental health disorders a public health priority. This need to increase access to treatment services has led to the successful development of computerized psychotherapy interventions. The majority of these interventions deliver cognitive behavioral therapy (CBT), an extensively validated, solution-based program frequently used in the treatment of depression, anxiety, and stress.

Spotlight: Blog post

Eureka: Evaluating and understanding progress in AI

How can we rigorously evaluate and understand state-of-the-art progress in AI? Eureka is an open-source framework for standardizing evaluations of large foundation models, beyond single-score reporting and rankings. Learn more about the extended findings. 

At Microsoft, we are collaborating with SilverCloud Health, the leading digital therapeutics platform for mental and behavioral health (opens in new tab), to improve mental health services using AI technology. This collaboration aims to jointly explore how AI can be used to enhance SilverCloud Health’s digital mental health platform and to deliver digital CBT-based programs, making treatment more accessible. In tandem with this collaboration, researchers at Microsoft Research Cambridge (opens in new tab) are investigating probabilistic machine learning frameworks to understand and meet people’s individual needs effectively with online CBT. We hope that this project will specifically lead to interventions that increase the quality of care for people living with mental health difficulties.

Bringing personalized cognitive behavioral therapy to more people who need it

CBT is a key approach used by clinicians to treat patients suffering from mental health issues. It is a therapy based on the concept that your thoughts, feelings, physical sensations, and actions are interconnected and that negative thoughts and feelings can trap you in a vicious cycle. CBT aims to help you deal with overwhelming problems in a more positive way by breaking them down into smaller parts. Unlike some other talking treatments, CBT deals with your current problems, rather than focusing on issues from your past.

Responding to the demand for effective and sustainable mental healthcare services, online-delivered psychotherapy interventions offer great potential for increased access to evidence-based care, cost reduction, and the provision of more tailored support. SilverCloud Health (opens in new tab) is the most widely used, online-supported, CBT-based platform employed by over 70% of NHS Improving Access to Psychological Therapies (IAPT) services in the UK. It is an evidence-based, digital mental health platform that delivers digital CBT-based programs in combination with limited but regular contact from a trained human supporter. It is one of very few mental health services that has been deployed at scale in routine clinical care. SilverCloud Health currently has the largest real-world patient user base of its kind globally. The platform offers more than 30 treatment programs to improve symptoms across the spectrum of well-being and mental health as well as to help determine the severity level of symptoms. The programs include sleep, resilience, depression, anxiety and chronic health conditions (such as diabetes).

“Through this research collaboration with Microsoft, SilverCloud Health will be able to leverage the latest in artificial intelligence and machine learning to further enhance our digital mental health platform. As a result, we hope to personalize treatment, intervene earlier, and deliver more effective treatment. This will ultimately help even more people improve their mental and behavioral health conditions.” – Ken Cahill, CEO of SilverCloud Health

Understanding patient behavior through probabilistic machine learning frameworks

Grounded in multi-disciplinary research, we are using machine learning and artificial intelligence to accelerate our understanding of personalizing mental health interventions. Leveraging several years of expertise in probabilistic machine learning and healthcare, we are already seeing promising results from one area of study, which is looking at whether people express different types of behavioral patterns in how they engage with CBT in an online environment. By developing a deeper understanding of different types of engagement behaviors, the view is that a probabilistic machine learning framework can help identify effective strategies. This would enable personalizing content or delivery of the CBT program to better meet each patients’ individual needs. For more information on our research, check out the Project Talia (opens in new tab) page.

October 10, 2019 marks World Mental Health Day (opens in new tab). This year, the campaign is raising awareness of the scale of suicide around the world and the role that each of us can play to help prevent it. Every 40 seconds someone loses their life to suicide. To gain more insight, check out the hashtags #WorldSuicidePreventionDay (opens in new tab) and #40seconds (opens in new tab) on social media.

Continue reading

See all blog posts