A key problem in Mental Health is that up to one third of patients suffering from major mental disorders develop resistance against drug therapy. However, patients showing early signs of treatment resistance (TR) do not receive adequate early intensive pharmacological treatment but instead they undergo a stepwise trial-and-error treatment approach. This situation originates from three major knowledge and translation gaps: i.) we lack effective methods to identify individuals at risk for TR early in the disease process, ii.) we lack effective, personalized treatment strategies grounded in insights into the biological basis of TR, and iii.) we lack efficient processes to translate scientific insights about TR into clinical practice, primary care and treatment guidelines. It is the central goal of PSYCH-STRATA to bridge these gaps and pave the way for a shift towards a treatment decision-making process tailored for the individual at risk for TR. To that end, we aim to establish evidence-based criteria to make decisions of early intense treatment in individuals at risk for TR across the major psychiatric disorders of schizophrenia, bipolar disorder and major depression. PSYCH-STRATA will i.) dissect the biological basis of TR and establish criteria to enable early detection of individuals at risk for TR based on the integrated analysis of an unprecedented collection of genetic, biological, digital mental health, and clinical data. ii.) Moreover, we will determine effective treatment strategies of individuals at risk for TR early in the treatment process, based on pan-European clinical trials in SCZ, BD and MDD. These efforts will enable the establishment of novel multimodal machine learning models to predict TR risk and treatment response. Lastly, iii.) we will enable the translation of these findings into clinical practice by prototyping the integration of personalized treatment decision support and patient-oriented decision-making mental health boards.

Mobile communication systems are becoming more and more complex to design (by researchers), operate (by the operators) and used by the people in the street. Mobile users now wish to be always connected, irrespective of time and place, and have access to a range of new services to help him/her in everyday life, all at the lowest possible cost. Currently no one knows how to evaluate whether a system is efficient or not in such provision. The reason for this is the huge number of parameters involved which collectively influence system efficiency. So far the practice has been to use a subset of such parameters to define localised efficiency -- but this does not provide overall efficiency and it will not lead to low cost or optimum use of scare spectrum. There are three important criteria which need to be considered and designed together to achieve a highly efficient mobile system. These are: quality of offered service, capacity and the cost of the system. Each of these criteria are influenced by a large number of parameters individually, where each have different weightings. Optimum design needs to find a fine balance between the three different criteria and yet currently there is no technique available which enables them to be optimised together to provide the required low cost solution. What makes this difficult is that a mobile system is dynamic by nature in terms of: range of mobility of users, wide range of operational environments, wide range of services with different bit rates and expected qualities, etc. This all points to requirements for a system with a certain degree of adaptability so that the system can self-organise and adapt itself to changing conditions. Currently systems are designed and operated on more or less fixed technique and parameters. These include the design of air-interface, media access control, handover algorithms, cell sizes and fixed frequency band allocation which all lead to wastage of resources and expensive solutions. The mobile systems of the future, addressed herein, are continuously adaptable and reconfigurable and respond automatically to the conditions of environments and user demands. It is only by engaging with these factors that efficiency can be maximised and the required low cost new services can be delivered to users. The challenge of the research described herein is how to collectively design such very complex networks so that users, service providers and network operators will all consider it efficient and cost effective to participate in the mobile vision of the future.

Doctoral Training Partnerships: a range of postgraduate training is funded by the Research Councils. For information on current funding routes, see the common terminology at https://www.ukri.org/apply-for-funding/how-we-fund-studentships/. Training grants may be to one organisation or to a consortia of research organisations. This portal will show the lead organisation only.

Individuals at the early stages of psychosis already suffer from neuroanatomical, neurofunctional and neurocognitive alterations. However, it is not clear how these alterations interact with each other and, more importantly, how this interaction may help predict who will become unwell. The aim of the proposed study is to uncover previously hidden relationships between brain anatomy, function and cognition that can improve our understanding of the first signs of the illness and help predict illness trajectory at the individual level. This will be achieved by triangulating, for the first time, data fusion, machine learning and a unique longitudinal dataset of individuals at the prodromal stage or with a recent first episode of psychosis. This combination of methods will allow me to address three main objectives: 1) identify previously unknown relationships between anatomy, function and cognition in healthy individuals; 2) develop a model that captures the normative pattern of these hidden relationships and determine by how much each patient deviates from this pattern; 3) use these deviations to predict each patient’s longitudinal clinical outcomes. I will also develop an open access tool that allows non-expert researchers to perform data fusion on their own data. Keywords: psychosis, data fusion, machine learning, outcome prediction. Background: Individuals at the early stages of psychosis already suffer from alterations in their brain anatomy and function, and behaviour (e.g. cognitive skills such as memory and attention). However, it is not clear how these changes interact with each other and, more importantly, how this interaction may help predict who will become unwell. Approach: In this study, I propose to address this issue in three main steps. First, I will identify previously hidden brain-behaviour relationships in healthy individuals. Second, I will develop a model that captures the normative pattern of these brain-behaviour relationships and estimate by how much each individual patient deviates from this pattern. Third, I will then use these deviations to make predictions about each patient’s illness course. Impact: Findings from my study will help to better understand brain-behaviour changes in psychosis and how we can use this information to bring personalised care to psychosis patients.

My project centres on the concept of brain dysfunction as it features in psychiatry, clinical psychology and neuroscience. It can be placed broadly within the emerging field of philosophy of psychiatry, and in a naturalistic tradition. In scientific psychiatry, the emerging consensus position is that mental disorders can be conceptualised as brain disorders. In the philosophical literature, disorder is hypothesised as being constituted by dysfunction. Understanding the nature of brain dysfunction is thus of crucial importance to contemporary psychiatry and its philosophy. My proposal is premised upon a distinction between two related, but non-synonymous, questions at the heart of any satisfactory account thereof - an epistemological question and a metaphysical question. The epistemological question asks how brain dysfunction is identified. Can we look at the brain and know for certain that it is functioning incorrectly, or do we need information from a higher explanatory level, for instance the mental or behavioural, to make the relevant discrimination? Our answer here would seem to depend upon what brain dysfunction is. In other words, the epistemological question depends upon the metaphysical question. I argue that an evolutionary analysis of function - where function is construed as selected effects - implies that mental dysfunction is brain dysfunction. Further, if mental dysfunction is brain dysfunction, mental disorder is brain disorder (constituted by brain dysfunction). This follows from my analysis whether or not mental dysfunctions are identifiable as dysfunctions at lower levels; in other words, regardless of our answer to the epistemological question.