connect with us
Enabling early detection of psychosis with speech and AI
Feb 16, 2023
New start-up team at Psyrin are disrupting psychosis care with speech and AI
A London-based startup looking to transform psychosis care through the power of speech has been developing machine learning algorithms that will detect impending psychosis from speech samples collected using just a smartphone. With Psyrin, practitioners along the psychosis care pathway will save time, improve triage accuracy, and increase capacity.
Julianna, one of the co-founders of Psyrin has been inspired by the suffering of her loved ones to work on solving mental health problems. By studying psychology, she realized the importance of prevention in mental health care.
"In case of severe mental disorders, especially in psychosis, the effect of prevention and early intervention can be really difference-making in the patient's life. However, preventative treatments need tools and technology to upscale and be accessible for as many people as possible."
At Psyrin, they want to unlock the potential of objective, scalable, AI-powered analysis of speech to help clinicians screen, triage and monitor and understand patients better, in a more effective way and eventually, use their technology to transform psychosis care pathways from reactive to preventative.
Supported by the
King's Entrepreneurship Institute.
Share
DRIVE-Health has been awarded £7.9 million from The Engineering & Physical Sciences Research Council (EPSRC) for student intake from 2024 onwards. DRIVE-Health is one of 65 CDTs which received funding, totalling more than £1 billion. Using seed funding from King’s Centre for Doctoral Studies awarded in 2020, DRIVE-Health has trained 30 students to date. Building on this, the new award will support five additional cohorts at King’s, totalling at least 85 talented PhD students. The CDT is expecting to welcome its fourth intake of at least 15 students in October 2024. DRIVE-Health is the first health data science training centre in the UK to harness cross-sector collaboration across the NHS, industry, enterprise, policy makers, and academia. Working with diverse partners, DRIVE-Health PhD students develop cutting-edge models which leverage healthcare data to improve patient outcomes, streamline operations, and enhance clinical decision-making processes. EPSRC CDT DRIVE-Health’s vision is informed by three core goals: To provide world-class training in health data science research to the next generation of health data scientists, who will have the multidisciplinary skills needed to enable transformations in public health and breakthrough treatments. To solve the most challenging problems in data-driven health research through a diverse community of the brightest minds in health data science and an open, collaborative culture which fosters exchange and champions innovation. To co-create a translational cross-sector collaboration with the NHS, industry, enterprise, policy makers and academia. Professor Richard Dobson, Co-Director of DRIVE-Health and Professor of Medical Informatics at King’s IoPPN, says "As more data from biological, social, genomic, imaging, smart devices, and electronic health records becomes available, there are significant opportunities to revolutionise the way healthcare is delivered. Through DRIVE-Health, we will train some of the brightest minds in health data science to develop cutting-edge tools which utilise data to improve healthcare systems and patient outcomes." "This is an exciting time for medicine, with new data paradigms creating a novel research and implementation landscape covering the full span from cell to society. Over the next nine years, DRIVE-Health will nurture world-class researchers that will chart that landscape and drive the UK’s health data agenda." Professor Vasa Curcin, Co-Director of DRIVE-Health and Professor of Health Informatics at King’s FoLSM. The DRIVE-Health PhD Programme (2024-2032) focuses on five key scientific research themes: Sustainable health data systems engineering: Investigates methods to develop secure and scalable software systems for healthcare. Theme lead: Dr Zina Ibrahim. Multimodal patient data streams: Integrates diverse patient data types for analysis, including wearables and electronic health records. Theme lead: Dr Jorge Cardoso. Complex simulations and digital twins: Builds simulated environments to train AI models for healthcare applications. Theme lead: Dr Steffen Zschaler. Next-generation clinical user interfaces: Ensures healthcare data science applications are usable in clinical settings. Theme lead: Professor Nick Holliman. Co-designing impactful patient-centric healthcare solutions: Co-producing and co-designing healthcare solutions to maximise impact across all themes. Theme lead: Professor Claire Steves. On top of the £7.9m provided by the EPSRC, DRIVE-Health has received over £5.1m from partners, as well as in-kind contributions worth nearly £4m.
We are delighted to announce that our first Seminar Series of 2024 will be hosted by the esteemed Dr Yves-Alexandre de Mountjoye from Imperial College, London. Companies and governments are increasingly relying on privacy-preserving techniques to collect and process sensitive data. In this talk, Yves-Alexandre will discuss their efforts to red team deployed systems and argue that red teaming is essential to protect privacy in practice. He will describe how traditional de-identification techniques mostly fail in the age of big data and then show how implementation choices and trade-offs have enabled attacks against real-world systems, from query-based systems to differential privacy mechanisms and synthetic data. He plans to conclude by describing recent successes in using AI to automatically discover vulnerabilities.
