This talk will review current research involving brain-computer interfaces to improve patients' health and well-being. In particular it will focus on the P300 speller, a device allowing severely motor disabled patients to use a keyboard based on their electroencephalogram. Ergonomic aspects of such a device are being studied in a partnership between Inria and Nice University Hospital.

An increasing number telehealth systems continuously collect selfreported data from patients. Objective and subjective collection of health data facilitates early detection and treatment of chronic conditions, but patient needs in these telehealth contexts are poorly understood. It is for example not clear how to support patients' reflection on their daily self-reported data. Inadequate support can result in fragmented daily health monitoring and poor adherence. This paper contributes 1) a synthesis of the related but hitherto disjunct personal informatics literature on self-tracking and 2) an indepth field study on how six people suffering chronic obstructive pulmonary disease (COPD) used a telehealth system as part of their health self-tracking. Our analysis showed that a telehealth solution which relegated patients to mere data suppliers missed out on opportunities to address user needs. We extended Li's 5-stage model to show where reflection manifested when interacting with the telehealth system.

Nowadays, it has become possible to measure different human activities using wearable devices. Besides measuring the number of daily steps or calories burned, these datasets have much more potential since different activity levels are also collected. Such data would be helpful in the field of psychology because it can relate to various mental health issues such as changes in mood and stress. In this paper, we present a machine learning approach to detect depression using a dataset with motor activity recordings of one group of people with depression and one group without, i.e., the condition group includes 23 unipolar and bipolar persons, and the control group includes 32 persons without depression. We use convolutional neural networks to classify the depressed and nondepressed patients. Moreover, different levels of depression were classified. Finally, we trained a model that predicts MontgomeryÅsberg Depression Rating Scale scores. We achieved an average F1-score of 0.70 for detecting the control and condition groups. The mean squared error for score prediction was approximately 4.0.

As the consequences of anxiety and depression have been compared to obesity and smoking as predictors of physical health, further findings, more advancements, and new technology are necessary to help those struggling with psychological disorders such as anxiety. This study investigates the potential relationships between Trait Anxiety or general anxiety scores and physiological and perceived reactions to a simulated virtual reality (VR) experience that induces mild anxiety as well as the ability to recover from the anxious event. The study additionally explores a potential relationship of a medical diagnosis on the physiological and perceived reactions to the simulated environment designed to induce mild anxiety and the potential effect on the ability to recover from such an event. Eighteen adults participated in the IRB (Institutional Review Board) approved study by completing a consent form, followed by the Trait Anxiety Questionnaire corresponding to the State Trait Anxiety Inventory form Y-2 to assess general anxiety levels. Participants additionally recorded a self-reflected Likert-scale interpretation of their perceived anxiety on a scale of one to ten after each phase of the study (Baseline, Introduction, Virtual Reality, Recovery). The experiment was designed to elicit mild anxiety with an ambiguous introduction and a shocking VR experience. The results showed no statistically significant difference between those with higher general anxiety with Trait Anxiety scores above 40 and those with lower Trait Anxiety in their percent increase of heart rate and increase of self-reflected anxiety score between baseline and VR phases as well as between baseline and recovery phases. Additionally, participants with medical diagnoses of anxiety showed no statistically significant difference in their percent increase of heart rate from baseline to VR phases as well as from baseline to recovery phases than their counterparts without any diagnoses of anxiety disorders. There is a potential indication, however, of a possible pattern of individuals with higher general anxiety (Trait Anxiety scores above 40) having a less-severe reaction, physiologically and perceptively, to an anxious situation than individuals with lower Trait Anxiety scores. This could indicate the possibility of desensitization to anxiety with frequent exposure. Conclusions of this study call for further investigation into this potential pattern and evaluation of future assistive technologies for individuals with anxiety.

Automated External Defibrillators (AEDs) are crucial for out-of-hospital cardiac arrests (OHCAs). Therefore, it is vital that AEDs are working and ready for use. However, knowledge of the locations of publicly accessible AEDs is limited, leading to difficulty in tracking, managing as well as obtaining one during an emergency situation. There is also often poor maintenance of AEDs (e.g., depleted batteries, expiration, or vandalism and theft). An AED locator mobile app was created to explore the feasibility of educating users on the location of AEDs and to motivate users to check the condition of the devices via gamification. An initial pilot study was conducted on the system's usability, effectiveness of gamification methods and users' motivation. Results suggest that gamification is effective in educating users on AED locations and motivates users to assess device condition

There is substantial evidence on the relevant factors that motivate participation in human subject studies and the expectations of participants when sharing their health data for research. However, most human subject studies focus on participant eligibility and data collection, omitting even a rudimentary use of the factors that motivate participation. We illustrate an approach to use motivation to construct personalized stories and exemplify it by using a chatbot under development towards monitoring, analyzing, and influencing health study participation, engagement, and retention. Additionally, we discuss the new advantages, challenges, and unexplored avenues for research stemming from our approach.