Reverse Engineering of Digital Measures: Inviting Patients to the Conversation.

Background: Digital measures offer an unparalleled opportunity to create a more holistic picture of how people who are patients behave in their real-world environments, thereby establishing a better connection between patients, caregivers, and the clinical evidence used to drive drug development and disease management. Reaching this vision will require achieving a new level of co-creation between the stakeholders who design, develop, use, and make decisions using evidence from digital measures. Summary: In September 2022, the second in a series of meetings hosted by the Swiss Federal Institute of Technology in Zürich, the Foundation for the National Institutes of Health Biomarkers Consortium, and sponsored by Wellcome Trust, entitled "Reverse Engineering of Digital Measures," was held in Zurich, Switzerland, with a broad range of stakeholders sharing their experience across four case studies to examine how patient centricity is essential in shaping development and validation of digital evidence generation tools. Key Messages: In this paper, we discuss progress and the remaining barriers to widespread use of digital measures for evidence generation in clinical development and care delivery. We also present key discussion points and takeaways in order to continue discourse and provide a basis for dissemination and outreach to the wider community and other stakeholders. The work presented here shows us a blueprint for how and why the patient voice can be thoughtfully integrated into digital measure development and that continued multistakeholder engagement is critical for further progress.

SIMON: A Digital Protocol to Monitor and Predict Suicidal Ideation.

Each year, more than 800,000 persons die by suicide, making it a leading cause of death worldwide. Recent innovations in information and communication technology may offer new opportunities in suicide prevention in individuals, hereby potentially reducing this number. In our project, we design digital indices based on both self-reports and passive mobile sensing and test their ability to predict suicidal ideation, a major predictor for suicide, and psychiatric hospital readmission in high-risk individuals: psychiatric patients after discharge who were admitted in the context of suicidal ideation or a suicidal attempt, or expressed suicidal ideations during their intake. Specifically, two smartphone applications -one for self-reports (SIMON-SELF) and one for passive mobile sensing (SIMON-SENSE)- are installed on participants' smartphones. SIMON-SELF uses a text-based chatbot, called Simon, to guide participants along the study protocol and to ask participants questions about suicidal ideation and relevant other psychological variables five times a day. These self-report data are collected for four consecutive weeks after study participants are discharged from the hospital. SIMON-SENSE collects behavioral variables -such as physical activity, location, and social connectedness- parallel to the first application. We aim to include 100 patients over 12 months to test whether (1) implementation of the digital protocol in such a high-risk population is feasible, and (2) if suicidal ideation and psychiatric hospital readmission can be predicted using a combination of psychological indices and passive sensor information. To this end, a predictive algorithm for suicidal ideation and psychiatric hospital readmission using various learning algorithms (e.g., random forest and support vector machines) and multilevel models will be constructed. Data collected on the basis of psychological theory and digital phenotyping may, in the future and based on our results, help reach vulnerable individuals early and provide links to just-in-time and cost-effective interventions or establish prompt mental health service contact. The current effort may thus lead to saving lives and significantly reduce economic impact by decreasing inpatient treatment and days lost to inability.

Validation of Visual and Auditory Digital Markers of Suicidality in Acutely Suicidal Psychiatric Inpatients: Proof-of-Concept Study.

BACKGROUND: Multiple symptoms of suicide risk have been assessed based on visual and auditory information, including flattened affect, reduced movement, and slowed speech. Objective quantification of such symptomatology from novel data sources can increase the sensitivity, scalability, and timeliness of suicide risk assessment. OBJECTIVE: We aimed to examine measurements extracted from video interviews using open-source deep learning algorithms to quantify facial, vocal, and movement behaviors in relation to suicide risk severity in recently admitted patients following a suicide attempt. METHODS: We utilized video to quantify facial, vocal, and movement markers associated with mood, emotion, and motor functioning from a structured clinical conversation in 20 patients admitted to a psychiatric hospital following a suicide risk attempt. Measures were calculated using open-source deep learning algorithms for processing facial expressivity, head movement, and vocal characteristics. Derived digital measures of flattened affect, reduced movement, and slowed speech were compared to suicide risk with the Beck Scale for Suicide Ideation controlling for age and sex, using multiple linear regression. RESULTS: Suicide severity was associated with multiple visual and auditory markers, including speech prevalence (ß=-0.68, P=.02, r2=0.40), overall expressivity (ß=-0.46, P=.10, r2=0.27), and head movement measured as head pitch variability (ß=-1.24, P=.006, r2=0.48) and head yaw variability (ß=-0.54, P=.06, r2=0.32). CONCLUSIONS: Digital measurements of facial affect, movement, and speech prevalence demonstrated strong effect sizes and linear associations with the severity of suicidal ideation.

Segregated Co-activation Patterns in the Emergence of Decision Confidence During Visual Perception.

Visual metacognition-the introspection and evaluation of one's own visual perceptual processes-is measured through both decision confidence and "metacognitive efficiency." Metacognitive efficiency refers to an individual's ability to accurately judge incorrect and correct decisions through confidence ratings given their task performance. Previous imaging studies in humans and nonhuman primates reported widely distributed brain regions being involved in decision confidence and metacognition. However, the neural correlates of metacognition are remarkably inconsistent across studies concerning spatial outline. Therefore, this study investigates the neural correlates of visual metacognition by examining co-activation across regions that scale with visual decision confidence. We hypothesized that interacting processes of perceptual and metacognitive performance contribute to the arising decision confidence in distributed, but segregable co-activating brain regions. To test this hypothesis, we performed task-fMRI in healthy humans during a visual backward masking task with four-scale, post-decision confidence ratings. We measured blood oxygenation covariation patterns, which served as a physiological proxy for co-activation across brain regions. Decision confidence ratings and an individual's metacognitive efficiency served as behavioral measures for metacognition. We found three distinct co-activation clusters involved in decision confidence: the first included right-centered fronto-temporal-parietal regions, the second included left temporal and parietal regions, and the left basal forebrain (BF), and the third included cerebellar regions. The right fronto-temporal-parietal cluster including the supplementary eye field and the right basal forebrain showed stronger co-activation in subjects with higher metacognitive efficiency. Our results provide novel evidence for co-activation of widely distributed fronto-parieto-temporal regions involved in visual confidence. The supplementary eye field was the only region that activated for both decision confidence and metacognitive efficiency, suggesting the supplementary eye field plays a key role in visual metacognition. Our results link findings in electrophysiology studies and human fMRI studies and provide evidence that confidence estimates arise from the integration of multiple information processing pathways.

The temporal evolution of pre-stimulus slow cortical potentials is associated with an upcoming stimulus' access to visual consciousness.

Slow cortical potentials (SCPs) have been proposed to be essential for the formation of conscious experience. To examine their temporal characteristics, we recorded electroencephalography during a visual backward-masking task, which required participants to localize the missing part of a target stimulus. A subsequent confidence rating was used as a proxy for the target's access to consciousness. Event-related potentials (ERPs) of all correct trials were determined relative to a brief period immediately before the target and then compared among consciousness levels. In an interval ranging from 2 s prior to target presentation up to this period, a negative relationship between slowly fluctuating ERP values and the level of consciousness became evident. After target presentation, high conscious awareness was characterized by an enhanced visual awareness negativity, an increased P3 component, and associated positive SCPs. Together, these findings add new evidence to the proposed role of SCPs in the emergence of visual consciousness.

Frequency-Dependent Spatial Distribution of Functional Hubs in the Human Brain and Alterations in Major Depressive Disorder.

Alterations in large-scale brain intrinsic functional connectivity (FC), i.e., coherence between fluctuations of ongoing activity, have been implicated in major depressive disorder (MDD). Yet, little is known about the frequency-dependent alterations of FC in MDD. We calculated frequency specific degree centrality (DC) - a measure of overall FC of a brain region - within 10 distinct frequency sub-bands accessible from the full range of resting-state fMRI BOLD fluctuations (i.e., 0.01-0.25 Hz) in 24 healthy controls and 24 MDD patients. In healthy controls, results reveal a frequency-specific spatial distribution of highly connected brain regions - i.e., hubs - which play a fundamental role in information integration in the brain. MDD patients exhibited significant deviations from the healthy DC patterns, with decreased overall connectedness of widespread regions, in a frequency-specific manner. Decreased DC in MDD patients was observed predominantly in the occipital cortex at low frequencies (0.01-0.1 Hz), in the middle cingulate cortex, sensorimotor cortex, lateral parietal cortex, and the precuneus at middle frequencies (0.1-0.175 Hz), and in the anterior cingulate cortex at high frequencies (0.175-0.25 Hz). Additionally, decreased DC of distinct parts of the insula was observed across low, middle, and high frequency bands. Frequency-specific alterations in the DC of the temporal, insular, and lateral parietal cortices correlated with symptom severity. Importantly, our results indicate that frequency-resolved analysis within the full range of frequencies accessible from the BOLD signal - also including higher frequencies (>0.1 Hz) - reveals unique information about brain organization and its changes, which can otherwise be overlooked.

Grading of Frequency Spectral Centroid Across Resting-State Networks.

Ongoing, slowly fluctuating brain activity is organized in resting-state networks (RSNs) of spatially coherent fluctuations. Beyond spatial coherence, RSN activity is governed in a frequency-specific manner. The more detailed architecture of frequency spectra across RSNs is, however, poorly understood. Here we propose a novel measure-the Spectral Centroid (SC)-which represents the center of gravity of the full power spectrum of RSN signal fluctuations. We examine whether spectral underpinnings of network fluctuations are distinct across RSNs. We hypothesize that spectral content differs across networks in a consistent way, thus, the aggregate representation-SC-systematically differs across RSNs. We therefore test for a significant grading (i.e., ordering) of SC across RSNs in healthy subjects. Moreover, we hypothesize that such grading is biologically significant by demonstrating its RSN-specific change through brain disease, namely major depressive disorder. Our results yield a highly organized grading of SC across RSNs in 820 healthy subjects. This ordering was largely replicated in an independent dataset of 25 healthy subjects, pointing toward the validity and consistency of found SC grading across RSNs. Furthermore, we demonstrated the biological relevance of SC grading, as the SC of the salience network-a RSN well known to be implicated in depression-was specifically increased in patients compared to healthy controls. In summary, results provide evidence for a distinct grading of spectra across RSNs, which is sensitive to major depression.