Public Funding for Cristal Health Ltd

A billion people worldwide live with mental disorders, with global costs projected to surpass £5tn by 2030\. In the UK, mental illness is the leading cause of disability, affecting one in six adults annually. Despite this, current treatments for mental disorders are effective for fewer than 50% of patients, reflecting the fact that we still understand very little about their underlying causes. One challenge is that diagnostic labels (like, 'major depression', 'schizophrenia', etc) do not necessarily correspond with biological differences between patients. Consequently, researchers are increasingly focusing on features that bridge diagnostic categories, like symptoms, blood tests, brain scans, or life circumstances (trauma, inequality etc). This trend is part of the so-called 'precision medicine' model - finding the right treatment, for the right patient at the right time. Electronic health records (EHRs) can be extremely useful in 'precision' research. EHRs capture patients' full history, including rich information on symptoms, life circumstances etc. Unfortunately, EHRs can be difficult to use for research. This is partly due to a lack of secure, privacy-preserving access infrastructure, and partly because EHRs are not designed for research, meaning their data can be hard to analyse, especially at scale. To address these issues, the Akrivia Health team have established a 15-year collaboration with the NHS, developing a secure, strictly access-controlled platform for research access to anonymised EHRs. Akrivia has developed a system of 'natural language processing' (NLP) algorithms to translate sensitive clinical notes into anonymised, structured tables that preserve only research-relevant information. Akrivia's platform now includes anonymised data from ~4.6 million patients', and is used to support precision research into mental health across the NHS (for free), academia (at a large discount) and industry. Since spinning out from Oxford University in 2019, demand for Akrivia's services has grown dramatically. In response, Akrivia is seeking Innovate support to transform our current manual, labour-intensive service into a highly-automated, scalable platform. We will develop materials to help users understand our data, standardise processes for contracting and project review by our oversight committee (including patient representatives), and build a more scalable version of our current secure data access environment. We will guide this work by involving patient groups, legal experts and potential users, and will continue expanding our dataset to increase its research impact. Ultimately, we aim to improve patients' lives by enabling research to understand the causes of mental illness develop more effective treatments.

Akrivia curates a database of 4.6 million+ patients' anonymised psychiatric electronic health records (EHRs) on behalf of 19 NHS healthcare organisations (HCOs). Using this privileged access, Akrivia aims to transform mental health and dementia research, where current treatments are often ineffective, or simply do not exist. Akrivia faces a challenge that ~85% of our EHRs' research-relevant data is stored in 'free-text' fields (e.g., clinical notes), which are difficult to analyse at scale and inaccessible outside the NHS due to their sensitivity. 'Natural language processing' (NLP) can help solve this. Akrivia has developed an AI-based NLP system to extract structured data on medications, symptoms, diagnoses etc from clinical notes. Akrivia's NLP is now enabling numerous NHS research projects, and driving revenue via our commercial research services. Thanks to this success, Akrivia has an opportunity to apply our NLP to new datasets, including NHS GP and US psychiatric data. This could dramatically improve our research impact, providing more complete patient journeys and increasing the diversity of our population. However, to support this, Akrivia's NLP solution needs to become more scalable. Some of our NLP models are based on older technology which takes a long time to update/expand, and our deployment processes rely heavily on manual steps. In this project, Akrivia will industrialise a prototype NLP solution we have developed, leveraging recent advances in AI research to improve scalability, accuracy, and human-interpretability. We will redevelop our technical infrastructure for increased automation, and expand our inhouse NLP tools so those without technical AI expertise (e.g., clinicians) can apply their specialist domain knowledge directly in model development. These innovations will enable us port our NLP approach to new datasets, helping provide researchers in mental health and dementia with the best possible data to address these complex, costly, and historically unaddressed conditions.

**Summary** Akrivia is using a new innovation from the field of artificial intelligence (AI) research to unlock the potential of electronic health record (EHR) data. Akrivia curates the world's largest database of psychiatric EHRs, with 4 million+ patients' deidentified data managed securely on behalf of 16 NHS healthcare organisations (HCOs). Akrivia's goal is to use this unique resource to transform mental illness and dementias research, driving treatment discovery and reducing trials costs. However, ~85% of UK psychiatric EHRs' actionable data is stored in unstructured, 'free-text' notes, which are difficult to analyse at scale and completely inaccessible to non-NHS researchers due to their level of personal information. 'Natural language processing' (NLP) can provide a solution to this inaccessibility, and Akrivia has developed an AI-based NLP system to extract new structured data on medications, symptoms, diagnoses etc. NLP models like Akrivia's can achieve high accuracy, but are traditionally limited by the need for large amounts of human-annotated training data. Creating this training data takes a long time, meaning that Akrivia's current NLP solution is not scalable enough meet the demands of their user base. In response, Akrivia has developed an alternative, prototype NLP solution using a novel 'few-shot' training method published in 2021\. This few-shot model requires very little training data to achieve high single task performance. Akrivia has used their prototype model to replace human annotators in their standard NLP development pipeline, achieving equal (or better) task accuracy with far shorter time-to-production (~4 weeks per concept versus ~6-9 months). **Vision** With this project, Akrivia will create the tools to scale their NLP development far faster than previously possible, with significantly less exposure of sensitive patient data. The toolkit will allow clinicians without technical AI expertise to develop models directly, creating a 'researcher-in-the-loop' solution to ensure Akrivia's NLP library embeds expert domain knowledge. The toolkit will also open a potential new service line in bespoke NLP solutions. Akrivia wants to provide researchers and clinicians working on mental illness and dementias with as broad and deep a dataset as possible. These diseases are complex, costly, and historically lacking in funding and treatment options. Large scale patient data with deep, broad descriptions of disease states has been critical to drug development and effective therapy provision in other areas like oncology. Through this project, Akrivia will develop the tools to make comparable data for mental illness and dementias a reality within a matter of months.

**Summary** Akrivia Health (Akrivia) is pioneering the development of a world-leading digital health platform, utilising the Electronic Health Records (EHR) data-sets held within "CRIS" (Clinical Record Interactive Search); an established platform that has the potential to transform mental health diagnosis and treatment. Akrivia is aiming to support researchers, scientists and industry to develop new treatments, better understand disease and manage healthcare services, whilst protecting the data privacy of patients and the integrity of NHS data. Akrivia's database is potentially the largest, deepest and most comprehensive dataset specifically for mental health worldwide. No other dataset includes narrative text completed by clinicians together with communications between healthcare providers alongside extensive structured data. Akrivia provides access to ~3 million de-identified patient records from 12 NHS Mental Health Trusts, ~1/3 of all secondary healthcare in NHS England and with plans for greater coverage and extension into devolved nations. The data governance model provides a safe and secure environment to engage both the public and private sectors. The de-identified patient-level data includes all clinical/patient interactions and interventions in both structured and unstructured data formats and with linkage to a wide range of other de-identified datasets, including UKBioBank. **Vision** Despite the spiralling unmet need, increasing costs to society and families, mental healthcare and research has historically been under-funded. The use of observational healthcare data has been limited due to this data being fragmented, unstructured and limited in size/scope. Clinical studies/trials are notoriously problematic in psychiatry as participants are hard to identify and keep engaged. In other therapeutic areas, real-world observational data is increasingly important to drug development and accelerating the development of effective therapies in cancer, cardiovascular medicine, and more recently with COVID-19\. Akrivia fundamentally believes that this should also be true in mental health. Akrivia intends to provide this missing data, giving all stakeholders the ability to access and derive information from mental health records, whilst having data security and privacy foremost. Akrivia's vision is that researchers armed with the right information can rapidly accelerate the development of new treatments. Akrivia itself intends to become the largest and most comprehensive mental health data bank globally. In this project, Akrivia is seeking to translate one of the world's most significant mental health datasets with cutting-edge AI to derive deep clinically relevant information from unstructured data. This will allow Akrivia to provide insights for clinical trial design, drug efficacy improvement, post-market surveillance and treatment pathway optimisation.