A collaboration between therapist, client and Artificial Intelligence for NiceDay
The importance of AI is increasing in almost every industry. In the mental health care industry, AI has the potential to make care more personalized and need-driven. However, discussions in research and policy communities show that it is increasingly important that a developed decision-making AI system is understandable for both developers of AI systems, its users, and people that are subjected to an AI-enabled decision. In order to enable further adoption of AI in mental healthcare, it is inevitable that an AI system’s decisions need to become and remain understandable by humans. Deeploy supports this by its MLOps platform in which AI predictions are made explainable. Furthermore, its human-in-the-loop feedback system allows medical experts to provide feedback to AI decisions, overrule decisions when necessary and grow trust.
Deeploy supported Niceday, a leading (online) mental health care platform, to help both mental health therapists and their clients by making use of AI in an explainable way. Mental health therapists are supported by AI in their daily decision process to reach out to at which moment in time, based on the client’s personal needs. Deeploy provided the software to deploy the AI model and its explainers. Moreover, we developed a feedback loop to be able to keep improving the system over time, based on the therapists’ feedback .
NiceDay is a leading (online) mental healthcare platform in the Netherlands, providing a service to therapists to help treat clients going through mental difficulties more effectively, ultimately strengthening the mental wellbeing of as many people as possible. NiceDay aims to change the way how mental healthcare is provided. Current developments in AI make it possible to provide more personal and effective support from therapists in the recovery of people with mental health complaints.
Use case: Supporting therapists to provide need-driven care
Within the NiceDay app, therapists can check-in with their clients between the regular therapy sessions. In between sessions, clients can register information about their well-being, upload finished assignments, and write diary entries. Informed by these registrations, therapists may decide to check-in with a client via the chat function in NiceDay. This in-between session forms an opportunity to provide need-driven, personalized mental health care. However, for therapists to decide which clients to reach out to at what moment in time is a challenge, given their limited time and the large amount of information to process.
An explainable AI system
To support the therapists in this challenge, an explainable AI system was developed. This system helps them in deciding which clients to reach out to at what moment in time. Using the explainable AI system, mental health care becomes more need-driven and personalized. Ultimately, this explainable AI system shows that it is possible to use AI in mental health care in an easy, transparent and explainable manner.
Challenges for using AI within mental healthcare
During the conversations we had with the therapists, it became clear what their issue was: limited time and more specifically, how to use that time to provide their clients with the best care? This use case possessed three challenges for us that we had to take into account during this project:
- Dealing with sensitive client data
- Cooperation between the AI model and therapists
It’s vital to ensure the safety, security and confidentiality of client’s sensitive data. Client data is anonymised and confidential during the development phase, while the explainability methods provided sufficient insight to therapists and patients.
With multiple companies being involved, it was important to ensure seamless integration to create a valuable product to the end-user.
Besides that, designing the human-AI system interaction is challenging. It is important that therapists can easily interact with the platform and are able to understand and give feedback on given predictions. Therefore, the AI system is developed in a clear and intuitive manner. Finally, the suggestions with explanations provided by the AI system are presented to the therapists in their Niceday web environment, and feedback can be given in the same environment.
By using Deeploy, ultimately, usage of our explainable AI system provides the feasible usage of AI in mental health care in an easy, transparent and explainable manner.
Contact us to find out more about Deeploy, and how we could help you out.