Real-Time Neurodiagnostics for Individualized Care

We aim to elevate mental health diagnostics to an unprecedented level of precision, collecting within-individual biometric data in real time with an ecosystem of wearables. Using novel analysis tools, achieve truly individualized mental healthcare by quantifying the efficacy of treatment options with confidence levels tailored to every patient's specific needs.

We are a team of technologists and researchers who are frustrated with the state of mental health care. Corey is a technical co-founder with first-hand experience with Post-Traumatic Stress Disorder: a survivor of childhood sexual assault, caretaker to his mother who was injured in the 2013 Boston Marathon Bombing, and friend to a variety of other survivors. Chase is a computational cognitive scientist who was previously an EMT and aid worker to refugees in the Balkans.

There is a critical need for improvements in mental healthcare hindered by a supply/demand mismatch and a trial/error approach to treatment. To combat this we are developing precision mental health analytics tools. These tools help augment existing healthcare infrastructure to scale evidence-based care.

Real-time data collection improves treatment efficacy by giving mental healthcare practitioners insight into the microstructure of mental health—the variance of a specific patient's symptoms over time in response to their genetics, environment, and personality. While treatment decisions are commonly informed by a single visit with a clinician, real-time data collection coupled with time-series analysis extends a clinician's view into a patient's lived experience. Real-time data collection captures many more data points over longer periods of time to better inform clinicians' treatment decisions.

Sophisticated methods of data analysis allow us to identify patterns in the time course of a patient's mental health, and match these patterns with "modules" of treatment options which address their specific needs best. For example, network analysis allows us to break down a patient's network of symptoms into clusters matched to treatment modules. Reducing the guesswork in psychiatric/psychological care is vital to promoting brain health.

We address the problem of non-ergodicity in research on mental health, which refers to the assumption that the variation of psychological processes between individuals mirrors how these processes vary within individuals - a significant limitation of standard research methods which obstructs the development of personalized mental healthcare (McNally, 2016). Person-specific assessment solves this problem by capturing patterns of variance within individuals as well as between them, allowing for clinicians to better tailor treatments to the specific needs of patients (Fisher & Boswell, 2016).

Our methods incorporate the latest data analysis techniques from psychopathology, psychophysiology, and clinical psychology. Initially, we will use network analysis methods (Haag et. al., 2017), real-time assessment (Fisher & Boswell, 2016), and Bayesian statistics (McNally, 2016).

Fisher, A. J., & Boswell, J. F. (2016). Enhancing the personalization of psychotherapy with dynamic assessment and modeling. Assessment, 23(4), 496-506.

Haag, C., Robinaugh, D. J., Ehlers, A., & Kleim, B. (2017). Understanding the Emergence of Chronic Posttraumatic Stress Disorder Through Acute Stress Symptom Networks. JAMA psychiatry, 74(6), 649-650.

McNally, R. J. (2016). Can network analysis transform psychopathology?. Behaviour research and therapy, 86, 95-104.

Patients will receive personalized, equitable mental health treatment. Clinicians have a more complete narrative of patient health and assessment. Researchers are empowered to turn their findings into translational diagnostic improvements. Insurance and governmental agencies will have better metrics to prioritize spending.

Run pilot study on 200+ PTSD patients - Reach 80% confidence level in our diagnostic analysis

Scale our electrophysiological collection capabilities to include more biometric identifiers such as EEG, galvanic skin response, and respiration rate - Achieve more comprehensive biometric data collection capabilities

Bring 50 external neuropsychiatrists, data scientists, and healthcare workers onto the platform to critique model architectures, collaborate on usability design, and spread the mission. - Community of collaborators on the data analysis platform

We are working with an entirely novel approach to data analysis in mental healthcare and diagnostics which offers an unprecedented sensitivity to the microstructure of the individual patient's mental health - the variance of their symptoms in real time, in light of the individual patient's specific history. While the research community has recognized these methods for their high impact potential, they have not yet been applied outside the lab. By developing hardware which hones in on biometric correlates of the symptoms of mental illness, we will remove what Professor Richard McNally has identified as the main obstacle to their application.

Our approach places the individual patient at the center of mental healthcare informed by participatory design. With our solution, we seek to answer the central question of individualized care: what treatment option, administered by which clinician, is best for this specific patient, given their unique circumstances?

In this way, we reconcile idiographic (or person-specific) statistical analyses with their nomothetic (or between-person) counterparts, combining information about the microstructure of an individual patient's mental health with standard between-person research results.

We will develop consumer-facing software with three consumer groups in mind - clinicians, researchers, and patients. Clinicians will access patient data through a user-friendly interface which operates at a high level of abstraction, making it easy to identify trends in patients' symptoms. In contrast, researchers will be able to test computational models on our data in a sufficiently technical framework which preserves patient privacy. Each patient will retain access to their own data and see, in a non-technical format, the results of analyses which their clinicians conduct.

We aim to release a set of products and services as we iterate. Two examples of products are a mass-casualty cognitive assessment tool and another is a dataset for neuropharmacodynamics to improve drug trial prioritization. We are optimistic that these will provide recurring revenue. We also seek a blend of traditional and nontraditional funding.

We will be applying to the NIH's Common Fund, NIMH's SBIR program, and other Federal grant programs. We have started communications with the military for potential partnerships/funding. We have begun to map out the Cambridge life sciences investor landscape. Previously we crowdfunded to cover a funding gap.

HIPAA/FHIR/Title 21 CFR Part 11 compliance & certification (Privacy).

NIST 800-160 standard adherence (Cybersecurity).

Access to larger permitted laboratory space.

Access to funding to cover materials expenses and compute time.

Proper marketing to improve patient buy-in after market approval.

https://www.generosity.com/medical-fundraising/help-build-the-first-medical-device-for-ptsd--3

Disruptive impact cannot be accomplished without sharing expertise, capabilities, and risk—we will leverage the Solve community to help accomplish our highest goal: equitable and individualized precision mental health care. We seek help to operationalize the 

We have been working with several groups in healthcare, biotech, and more in the Cambridge and Chicagoland areas, but do not yet have official written partnership agreements and so cannot publicly list.

Neurofusa - Dan Morden,

Neuroelectrics,

Aifred Health