Developing Advanced AI Ecosystems to Enhance Diagnosis and Care for Patients with Depression.

Major Depressive Disorder (MDD) has a significant impact on the daily lives of those affected. This concept paper presents a project that aims at addressing MDD challenges through innovative therapy systems. The project consists of two use cases: a multimodal neurofeedback (NFB) therapy and an AI-based virtual therapy assistant (VTA). The multimodal NFB integrates EEG and fNIRS to comprehensively assess brain function. The goal is to develop an open-source NFB toolbox for EEG-fNIRS integration, augmented by the VTA for optimized efficacy. The VTA will be able to collect behavioral data, provide personalized feedback and support MDD patients in their daily lives. This project aims to improve depression treatment by bringing together digital therapy, AI and mobile apps to potentially improve outcomes and accessibility for people living with depression.

Gaze-based attention refocusing training in virtual reality for adult attention-deficit/hyperactivity disorder.

BACKGROUND: Attention-deficit/hyperactivity disorder (ADHD) is characterized by substantial interindividual heterogeneity that challenges the systematic assessment and treatment. Considering mixed evidence from previous neurofeedback research, we present a novel feedback system that relies on gaze behavior to detect signs of inattention while performing a neuropsychological attention task in a virtual seminar room. More specifically, an audiovisual feedback was given whenever participants averted their gaze from the given task. METHODS: Eighteen adults with ADHD and 18 healthy controls performed a continuous performance task (CPT) in virtual reality under three counterbalanced conditions in which either gaze-based feedback, sham feedback, or no feedback was provided. In all conditions, phases of high and low virtual distraction alternated. CPT errors and reaction times, proportions of gaze dwell times (e.g., task focus or distraction focus), saccade characteristics, EEG theta/beta ratios, head movements, and an experience sampling of ADHD symptoms were analyzed. RESULTS: While patients can be discriminated well from healthy controls in that they showed more omission errors, higher reaction times, higher distraction-related dwell times, and more head movements, the feedback did not immediately improve task performance. It was also indicated that sham feedback was rather associated with an aggravation of symptoms in patients. CONCLUSIONS: Our findings demonstrate sufficient suitability and specificity for this holistic ADHD symptom assessment. Regarding the feedback, a single-session training was insufficient to achieve learning effects based on the proposed metacognitive strategies. Future longitudinal, multi-session trials should conclusively examine the therapeutic efficacy of gaze-based virtual reality attention training in ADHD. TRIAL REGISTRATION: drks.de (identifier: DRKS00022370).

Can biofeedback-based training alleviate fatigue and vigilance performance in fatigued MS patients?

MS related fatigue might be related to autonomous nervous system (ANS) dysfunctions or to inflammation related vagal (hyper-) activation. Consequently, influencing ANS status may lead to relieve of fatigue. We used two opposite biofeedback interventions to either increase sympathetic ("self-alert training", SAT) or parasympathetic activation ("progressive muscle relaxation", PMR). We recorded fatigue status of patients before and after a challenging vigilance task, their behavioural performance on this task, their skin conductance response (SCR), and parameters indicating parasympathetic activity concerning heart rate variability (HRV). We repeated these recordings after the biofeedback training sessions. Patients of the SAT group were able to learn to increase their SCR voluntarily. Patients of the PMR group showed increasing parameters indicating parasympathetic modulation of the HRV. The vigilance task increased their feeling of fatigue. However, there was no effect of biofeedback training on either fatigue status or performance on the vigilance task. Our results show that MS patients can learn to change voluntarily their ANS activity using biofeedback instructions based on SCR and this can be used in future studies to test the postulated link between ANS and fatigue. However, in this experimental intervention we were unable to document a relation between ANS activity and fatigue.Trial registration: ClinicalTrials.gov identifier: NCT03268187.

Exploring Self-Paced Embodiable Neurofeedback for Post-stroke Motor Rehabilitation.

Neurofeedback-guided motor-imagery training (NF-MIT) has been proposed as a promising intervention following upper limb motor impairment. In this intervention, paretic stroke patients receive online feedback about their brain activity while conducting a motor-imagery (MI) task with the paretic limb. Typically, the feedback provided in NF-MIT protocols is an abstract visual signal based on a fixed trial. Here we developed a self-paced NF-MIT paradigm with an embodiable feedback signal (EFS), which was designed to resemble the content of the mental act as closely as possible. To this end, the feedback was delivered via an embodiable, anthropomorphic robotic hand (RH), which was integrated into a closed-looped EEG-based brain-computer interface (BCI). Whenever the BCI identified a new instance of a hand-flexion or hand-extension imagination by the participant, the RH carried out the corresponding movement with minimum delay. Nine stroke patients and nine healthy participants were instructed to control RH movements as accurately as possible, using mental activity alone. We evaluated the general feasibility of our paradigm on electrophysiological, subjective and performance levels. Regarding electrophysiological measures, individuals showed the predicted event-related desynchronization (ERD) patterns over sensorimotor brain areas. On the subjective level, we found that most individuals integrated the RH into their body scheme. With respect to RH control, none of our participants achieved a high level of control, but most managed to control the RH actions to some degree. Importantly, patients and controls achieved similar performance levels. The results support the view that self-paced embodiable NF-MIT is feasible for stroke patients and can complement classical NF-MIT.

Motor Imagery Impairment in Postacute Stroke Patients.

Not much is known about how well stroke patients are able to perform motor imagery (MI) and which MI abilities are preserved after stroke. We therefore applied three different MI tasks (one mental chronometry task, one mental rotation task, and one EEG-based neurofeedback task) to a sample of postacute stroke patients (n = 20) and age-matched healthy controls (n = 20) for addressing the following questions: First, which of the MI tasks indicate impairment in stroke patients and are impairments restricted to the paretic side? Second, is there a relationship between MI impairment and sensory loss or paresis severity? And third, do the results of the different MI tasks converge? Significant differences between the stroke and control groups were found in all three MI tasks. However, only the mental chronometry task and EEG analysis revealed paresis side-specific effects. Moreover, sensitivity loss contributed to a performance drop in the mental rotation task. The findings indicate that although MI abilities may be impaired after stroke, most patients retain their ability for MI EEG-based neurofeedback. Interestingly, performance in the different MI measures did not strongly correlate, neither in stroke patients nor in healthy controls. We conclude that one MI measure is not sufficient to fully assess an individual's MI abilities.

Embodied neurofeedback with an anthropomorphic robotic hand.

Neurofeedback-guided motor imagery training (NF-MIT) has been suggested as a promising therapy for stroke-induced motor impairment. Whereas much NF-MIT research has aimed at signal processing optimization, the type of sensory feedback given to the participant has received less attention. Often the feedback signal is highly abstract and not inherently coupled to the mental act performed. In this study, we asked whether an embodied feedback signal is more efficient for neurofeedback operation than a non-embodiable feedback signal. Inspired by the rubber hand illusion, demonstrating that an artificial hand can be incorporated into one's own body scheme, we used an anthropomorphic robotic hand to visually guide the participants' motor imagery act and to deliver neurofeedback. Using two experimental manipulations, we investigated how a participant's neurofeedback performance and subjective experience were influenced by the embodiability of the robotic hand, and by the neurofeedback signal's validity. As pertains to embodiment, we found a promoting effect of robotic-hand embodiment in subjective, behavioral, electrophysiological and electrodermal measures. Regarding neurofeedback signal validity, we found some differences between real and sham neurofeedback in terms of subjective and electrodermal measures, but not in terms of behavioral and electrophysiological measures. This study motivates the further development of embodied feedback signals for NF-MIT.

Biofeedback-based self-alert training reduces alpha activity and stabilizes accuracy in the Sustained Attention to Response Task.

OBJECTIVE: Deficits in sustaining attention are common in various organic brain diseases. A recent study proposed self-alert training (SAT) as a technique to improve sustained attention. In the SAT, individuals learn to gain volitional control over their own state of arousal by means of electrodermal biofeedback. METHOD: In this study, we investigated the behavioral, electrodermal, and electroencephalogram correlates of the SAT with a blinded, randomized, and active-controlled pre-post study design. Sustained attention capacity was assessed with the Sustained Attention to Response Task (SART). RESULTS: The SAT resulted in strong phasic increases in skin conductance response (SCR), but endogenous control of SCR without feedback was problematic. Electroencephalogram analysis revealed stronger alpha reduction during SART for the SAT than for the control group. Behaviorally, the SAT group performed more accurately and more slowly after intervention than the control group. CONCLUSION: The study provides further evidence that SAT helps to maintain SART accuracy over prolonged periods of time. Whether this accuracy is more related to sustained attention or response inhibition is discussed.