Knowledge Vault 3/28 - G.TEC BCI & Neurotechnology Spring School 2024 - Day 3
EEG in action extracting fMRI guided EEG biomarkers for therapy
Oded Kraft, Graymatters (IL)
<Resume Image >

Concept Graph & Resume using Claude 3 Opus | Chat GPT4 | Llama 3:

graph LR classDef main fill:#f9d4d4, font-weight:bold, font-size:14px; classDef technology fill:#d4f9d4, font-weight:bold, font-size:14px; classDef mentalHealth fill:#d4d4f9, font-weight:bold, font-size:14px; classDef product fill:#f9f9d4, font-weight:bold, font-size:14px; classDef development fill:#f9d4f9, font-weight:bold, font-size:14px; A[Oded Kraft] --> B[CEO shares Brain Matters journey 1] A --> C[Technology treats mental disorders
using brain biomarkers 2] C --> D[Prism for PTSD, depression
uses EFP biomarker 3] A --> E[Mental disorders: prevalent, undertreated 4] E --> F[Brain subsystem imbalances linked
to mental disorders 5] F --> G[Brain Matters targets emotion
regulation, reward systems 6] C --> H[fMRI neurofeedback trains amygdala
regulation, but inaccessible 7] H --> I[EEG-fMRI bridge enables EFP
biomarker for neurofeedback 8] I --> J[EFP: patient-generic, computationally
efficient for real-time EEG 9] D --> K[Prism: 8-electrode EEG cap,
15 sessions over 4-8 weeks 10] K --> L[Patients find strategies to
calm avatars, regulate EFP 11] K --> M[Data collected for personalized
treatment, quality control 12] D --> N[Prism: user-friendly, empowering, avoids
exposure therapy 13] D --> O[Prism benefits providers: revenue,
augments standard care 14] D --> P[Prism for depression: reward
system, dog-walking avatar 15] C --> Q[Multidisciplinary expertise in
product development 16] D --> R[Clinical evidence: target engagement,
outcomes, safety 17] R --> S[FDA study: 67% improvement,
remission, few side effects 18] D --> T[Prism improves sleep, biomarker
regulation in responders 19] D --> U[Prism now commercial in
US, future plans 20] B --> V[Four years: fundraising to
team growth 21] D --> W[Home use challenges: dry
electrodes, insurance, environment 22] C --> X[EFP biomarker currently patient-generic,
tailoring explored 23] C --> Y[Virtual crowd responds to
EFP regulation, no stimulation 24] R --> Z[Various control groups used
in studies 25] I --> AA[EFP development: complex EEG-fMRI
analysis, 7 years 26] C --> AB[Data proprietary, some published,
collaborations case-by-case 27] C --> AC[Quantifying PTSD, depression severity
possible with large datasets 28] K --> AD[Patients use varied mental
strategies, commonalities analyzed 29] C --> AE[Promising for resilience in
healthy, challenges in consumerization 30] class A,B main; class C,H,I,J,Q,W,X,Y,AA,AB,AC,AE technology; class D,E,F,G,K,L,M,N,O,P,R,S,T,U,V,Z,AD product;


1.-Oded Kraft, CEO and co-founder of Brain Matters Health, shares their journey of developing technology for treating mental disorders.

2.-Brain Matters uses brain-specific biomarkers from machine learning on EEG and fMRI data to provide non-invasive treatments for mental disorders.

3.-Their first commercial product, Prism for PTSD, uses an amygdala-based EFP biomarker. Prism for depression is in clinical development.

4.-Mental disorders affect many lives globally. Existing treatments have limited efficacy, side effects, and accessibility issues.

5.-Different mental disorders are associated with imbalances in various brain subsystems, such as the emotion regulation and reward systems.

6.-Brain Matters started with the negative valence system (amygdala) as it impacts many lives and lacks targeted treatments.

7.-Real-time fMRI neurofeedback can train individuals to regulate the amygdala, but it's costly and not widely accessible.

8.-Professor Handler hypothesized finding a statistical bridge between EEG and fMRI of the amygdala, leading to the EFP biomarker.

9.-The EFP biomarker is patient-generic and computationally lightweight, allowing real-time EEG neurofeedback inspired by fMRI data.

10.-The Prism product uses an 8-electrode EEG cap and software. Treatment involves 15 sessions over 4-8 weeks.

11.-During sessions, patients find mental strategies to calm avatars, representing their ability to downregulate the EFP biomarker.

12.-The system collects data during sessions for quality control and predictive analytics to personalize treatment.

13.-Prism aims to be easy to use, empower patients, avoid exposure therapy, and help patients gain emotion regulation skills.

14.-For providers, Prism offers a new revenue stream and augments standard care. It's designed to be simple to operate.

15.-Prism for depression uses the reward system biomarker. The interface involves helping an avatar take a dog for a walk.

16.-The product development involves expertise from neuroscience, psychiatry, psychology, neuroengineering, data science, software, and more.

17.-Clinical evidence shows target engagement (amygdala downregulation), improved outcomes over time, and a high safety profile.

18.-The FDA study showed 67% clinical improvement 3 months post-treatment, high remission rates, and few side effects.

19.-Prism improved sleep quality, a common lingering issue in PTSD. Clinical responders showed better biomarker downregulation over time.

20.-Prism is now commercial in the US. Future plans include more indications, simplified products, and data-driven services.

21.-The last four years involved fundraising, product development, clinical studies, FDA clearance, and team growth.

22.-Home use would require dry electrodes and addressing insurance coverage. Creating a controlled environment at home is challenging.

23.-The EFP biomarker is currently the same for all patients, with a short calibration phase. Tailoring is being explored.

24.-The virtual crowd responds to the patient's ability to downregulate the EFP biomarker. Brain stimulation is not used.

25.-Different control groups were used in studies, such as yoked sham neurofeedback and enhanced dialectical behavior therapy.

26.-The EFP development involved complex analysis of simultaneous EEG-fMRI recordings to find correlations, taking about seven years.

27.-Data is currently proprietary, but some is published. Collaborations are considered case-by-case.

28.-Quantifying PTSD and depression severity using this technology is possible in the future with large datasets and evolving standards.

29.-Patients use varied mental strategies based on individual experiences. Common threads exist and are being analyzed.

30.-The technology shows promise for building resilience to stress in healthy populations, with some challenges in consumerization.

Knowledge Vault built byDavid Vivancos 2024