We introduce to you Morphii®, a patented “digital affect mirror” system to capture, measure, and monitor Affective Determinants of Health (ADoH™). In head-to-head construct validation studies this new methodology performs quite strongly against a variety of established traditional measures of stress, anxiety, loneliness, irritability, and depression. This new toolkit demonstrates promise for future development of engaging, efficient, and effective screening and progress monitoring of mental phenomena that represent risk factors for a wide variety of chronic diseases and non-adherence.

COVID-19 pandemic is a global calamity posing an unprecedented opportunity to study resilience. We developed a brief stress and resilience survey and applied it online during the acute COVID-19 outbreak (April 6-15, 2020), on a crowdsourcing research website (https://covid19resilience.org/) advertised through social media. Data collected from thousands of participants indicate that on general, people were (1) more stressed about others than about self and (2) stressed to a similar extent about contracting covid-19 compared to its economic impact. Our survey captured the buffering effect of resilience on the association between COVID-19 imposed stress and mental health outcomes, suggesting that emotion regulation may be a target for interventions aimed at the enhancement of resilience.

In this presentation I argue that it is an ethical imperative to design technologies, particularly in health,  that support wellbeing, human autonomy, fairness and do not harm. I discuss how everyday technology shapes our health and wellbeing, and how designing to support the satisfaction of psychological needs can lead to better health and wellbeing outcomes. But these intentions and methods should be supported to a commitment to assess the impact of technologies, just as we have done in Environmental Impact Assessment for the last 50 years.

Chandler Carter discusses her research into how individuals are influenced by communication that to greater or lesser degrees describe the struggle that may be required to meet weight loss objectives. She shares her thoughts on the broader implications of candid communication concerning struggle in other areas of health.

Social is currently used as a source of data about behavior and mood to predict high-risk and dangerous mental illness behaviors, such as suicidal ideation, disordered eating behaviors, and opioid abuse. In this talk, I’ll discuss the promises of AI systems to predict and understand these behaviors, as well as the perils and new challenges of applying these systems to assist in interventions.

We monitor the affective signals expressed in social media content and use these signals to infer social media user's depressive symptoms. We found the transitions from one affective state to another, temporal representation of affect, and even affective content not created by the social media users themselves, reflect users' depressive symptoms.

Mental and psychological well-being concerns can have wide-reaching effects on people’s daily activities, education, employment, occupational functioning, and relationships. There is a need for pro-active detection of mental illness, given prevailing stigma and people’s inability to access timely formal care; however, even with timely detection, management of mental illnesses remain difficult. As a way to tackle these challenges, Dr. De Choudhury, in this talk, will highlight her research examining if we can computationally employ social media data to proactively detect one’s risk to mental health challenges, and how they can facilitate technological interventions for improving one’s wellbeing, by adopting a multi-scale, multi-disciplinary, and multi-stakeholder perspective.

How has the COVID-19 Crisis affected our digital well-being? How have tech companies responded? In this presentation, I explore the merits of the digital well-being strategy provided by the Center for Humane Technology.

Effective labelling of datasets for supervised learning is a crucial step in any machine learning task, but how easy is it to label something as complex as mental health? In this talk we consider how this has been done in 66 papers that predicted mental health on Twitter, and consider the pitfalls of common data collection and annotation methods.

An account of how we came to conduct a systematic review of empirical studies that have compared the delivery of psychological therapies by telephone and face-to-face (i.e. in person). An overview of the aim and scope of the review, the variables reviewed, and our principal result, which was that there is very little evidence of comparative difference when considering such interactional features as alliance, disclosure, empathy, attentiveness or participation. There is therefore No empirical evidence to support prevailing professional narrative that absence of visual and physical co-presence is detrimental to alliance formation

Language is a window into our internal cognitive structures and mental states. By applying machine learning and natural language processing methods we can generate actionable inferences about patient mental states.

This talk focuses on collection and evaluation of privacy-based acoustic data for detection of Alzheimer's dementia Speech. It also presents emotional prosody information and acoustic feature transfer learning between monologues and dialogues of spontaneous speech for detection of Alzheimer's dementia Speech.

Human Values concerns such as privacy, safety, security expressed by users of COVID-19 contact tracing apps are poorly addressed or ignored by development teams. This talk explains how the mismanagement of users value concerns lead to the lack of contact tracing apps uptake and the mental health of its users.

This talk presents a brief introduction to the idea of ‘positive virtual ecosystems’. It specifically introduces how the wide variety of support offered by online counselling and support services might be conceptualised from a humanistic psychology perspective.

How might we measure therapeutic alliance between humans and mental health chatbots? If people could write their own survey, what questions would they want to be asked? In this talk, you will learn how to create your own Wiki Survey — a crowdsourcing tool for collecting and ranking data.

The majority of work in NLP on mental health prediction, even when using longitudinal social media data, involves distinguishing individuals with a condition from controls rather than assessing individuals’ mental health at different points in time. This talk will present the objectives addressed by Prof Liakata’s five-year Turing AI Fellowship which aims to establish a new area in natural language processing on personalised longitudinal language processing. Progress so far on developing sensors for capturing digital biomarkers from language and heterogeneous UGC to understand the evolution of an individual over time will also be discussed.

During Covid-19 related lockdown, the quality of the space we live in and the communication with our social community emerged as key factors impacting our mental well being. Digital technologies such as virtual reality can help us cope with these challenging factors and overcome social distancing. A meaningful support to not only general citizens but for hospitalized patients as well, undergoing a similar stressful experience of confinement and isolation.

This talk focuses on why it is important to pay attention to the construction of positive experience as part of an active intervention strategy to rebuild or maintain positive well-being. To achieve this outcome, practitioners would need to understand what mattering is, and how to embed mattering into our everyday lives. At the end of the session, practitioners will appreciate mattering as a refreshing idea and a unique strategy for formulating solutions to mental health issues and a unique framework to unpack and understand issues.

The developments of digital psychiatry have rapidly increased and cover the use of telehealth, electronic health records and prescription, remote patient monitoring through smart-phone apps and sensors, robots, virtual reality and artificial intelligence. This talk discusses the current clinical and research implications of using technology in mental health care, raising ethical questions that these advances may bring.

This presentation looks at how a person's ground truth can be discovered. It highlights how the principle of a 'centre of gravity' can be used by software to discover the core truth a person needs to make a priority to act on so as to make the best of their day and to evolve the best version of themselves. It also shows how cultivating this Personal Value can be expanded by helping others act on their ground truth so as to generate a shared Social Value in their lives.

AI and machine learning methods using human language have the potential to revolutionize mental healthcare, providing predictive models and insight into people's experiences and mental state. But there is a data crisis in language technology for mental health and healthcare in general. This talk lays out the problem and steps toward solving it.

In this talk I will describe the development of the #chatsafe guidelines, the world’s first guidelines designed to facilitate safe peer to peer communication about suicide online. I will also showcase some of the associated social media campaign. I will then talk briefly about the evaluation of the intervention that is currently underway, and highlight some of the next steps for the #chatsafe project.

In this talk, Daniel will be discussing the details of his Master’s thesis project, which focused on the presence of the Therapeutic Alliance in algorithm-based ‘therapeutic’ chatbots. Future research suggestions and aspirations will follow.

There is increasing concern that technology is harming youth and adolescent mental health. This talk will discuss the potentials of online spaces and new technologies to support youth and adolescent mental health. This includes considerations of current technologies like apps and potentials for future technologies.

I used neural networks combined with multiple different sources of wellbeing data to not only quantify the connection between beautiful places and our wellbeing, but also to understand what beautiful places are composed of. It turns out that beautiful places are not simply natural places - the story is more complex. And, I found that there is a connection between beautiful places and better reported health and greater happiness.

Electronic health records provide a rich resource for large-scale retrospective observational studies. Symptom onset information is important for improved understanding of disease trajectories and treatment outcomes, but this information is mainly documented as free text in clinical notes. In this presentation, work on using natural language processing approaches for extracting symptom onset information related to psychosis patients in mental health care is presented, with some initial results.

An overview of a research project using surveys and experiments to understand the UK public’s susceptibility to health-related misinformation during COVID-19, identifying factors associated with individuals who are likely to be vulnerable as well as factors associated with misleading content. This talks reports on the research motivation, hypotheses and design. Preliminary results are not yet available.

In this talk, I will provide an overview of interactive theatre and explore 'Digital Interactive Theatre' as an innovation assisting nurses to make informed clinical decisions.