The EU-funded Mindspeak project starts
Digital mental healthcare interventions (DMHIs) powered by GenAI are a solution to provide accessible treatment for patients suffering from common mental health problems (i.e., depression and anxiety disorders).
NiceDay‘s DMHIs platform is an academically verified effective tool for treating depression or anxiety disorders. Within the Mindspeak project this platform will be enhanced by state of the art NLP technologies from Kairntech. The new language-based capabilities enhance productivity while improving treatment efficiency.
More efficient and higher quality healthcare
Mental health problems are growing in incidence and severity. The most prominent, depression and anxiety disorders, together amount to a significant loss of economic productivity. They also result in a reduced quality of life and life expectancy.
At the same time mental healthcare is under-treated. In the United States, fewer than half of individuals in need of mental health treatment receive care. Many people have difficulty accessing appropriate mental health care and this has been exacerbated by the COVID-19 pandemic.
When mental healthcare professionals can spend more time on treatment and these treatments become even more effective, mental health problems can be treated more decisively.
Automated natural language processing
In the MindSpeak project (Enhancing data-supported psychotherapy through Advanced AI-Powered Language Technologies and Personalization) NiceDay and Kairntech will develop a unique and revolutionary feature for Digital Mental Health Care Interventions (DMHIs). Natural Language Processing (NLP) will automate therapist tasks. Data is created and leveraged to radically improve the quality of treatments.
At the core of the process is the automated generation and smart summarization of meeting notes after each treatment. With the use of state-of-the-art NLP, the spoken text during the (online) treatment session will be transcribed and cleaned. And also reworked in order to be able to distinguish between speakers.
Through complex NLP pipelines the solution generates smart summaries. These are not only generated during the sesssion with the therapist, but also enriched with metadata such as historical minutes and therapist input. Automating this task saves mental healthcare professionals up to an hour a day and allows them to treat more patients within the same amount of time.
Meeting notes for GenAI mental healthcare
Meeting notes as a data source improve treatment quality and tailor the treatment to the individual patient needs. The analysis of meeting notes, their history and relevant metadata are all used to track treatment progress.
Such analysis gives the therapist and the patient more meaningful and data-driven insights. In doing so, NiceDay can recommend new exercises and therapies based on the patient’s previous treatments and progress. This refers to both sessions with the therapist and online exercises that the patient performs independently.
The MindSpeak project will enhance the NiceDay platform by incorporating microlearning capabilities. Microlearning involves delivering small, digestible content chunks, such as videos, podcasts, quizzes, and downloadable materials, which take 3-5 minutes to consume. Each content piece focuses on a specific learning goal, providing quick answers to relevant questions.
The NiceDay platform will use NLP to create tailored, engaging content from its open-source database, which will then be reviewed by therapists before being sent to patients. This personalized content, delivered in various formats, aims to improve treatment effectiveness through microlearning.
The project will support multiple languages (French, Dutch, and English) and will use specialized NLP models designed for mental health care. Additionally, metadata collected from generated summaries and other content will be analyzed to provide insights into treatment effectiveness, helping therapists improve outcomes. This aligns with NiceDay’s goal of ensuring all psychologists perform at the highest level.