By petr zhegin in neuro — Oct 29, 2024

(Empire) state of mind: preparing for NY BCI Symposium 2024

A selection of papers and links related to the agenda of the NY BCI Symposium

*Brooklyn Bridge from wework @ 199 Water St, Peter Zhegin*

Intro

This is a collection of links to papers published by speakers of the NY BCI Symposium and others. The links are not necessarily connected to the talks at the Symposium but provide context and reflect on what I find interesting as an investor/funder. These links are structured around the following themes:

Intro

Existing and emerging technologies
Technological challenges
Clinical outcomes and user preferences
Ethical challenges
Non-medical BCIs and enhancement
Neurotech translation
Funding
Asks and wants

References

Ping me at a conference or in NY (p@e184.com) and join podcast where I speak to @sab0rg, @PedroHenriched, and @NiklasAnzinger on BCIs and human augmentation - Thursday, Oct 31st at 11am GMT-6.

Ep #3: Human Augmentation, BCIs and E-Telekinesis
Hosted by @PedroHenriched & @NiklasAnzinger

(↓ Link of the Space in the next post )

🚀 Featuring as a startup: @sab0rg from Ucat

🎯 Special Investor Guest: @PeterZhegin from e184

→ Thursday, Oct 31st at 11am GMT-6 pic.twitter.com/ohafCdNKuR
— Vitalia City (@VitaliaCity) October 29, 2024

1. Existing and emerging technologies

Back to the table of contents

The brain-computer interface (BCI) ecosystem went a long way from the early 70es when the term BCI was coined, through the late 90es when ' the first long-term iBCI electrodes were implanted in 1998 at the Georgia Institute of Technology' [1]. Around a year ago, 'two brain-computer interfaces have been developed that bring unprecedented capabilities for translating brain signals into sentences — at speeds close to that of normal speech, and with vocabularies exceeding 1,000 words.' [28]

By 2023, 21 research groups had conducted 28 implanted BCIs (iBCIs) clinical trials with 67 participants [1]. Six companies across four electrode types provided systems for ongoing/recent trials: electrocorticography systems (ECoG) by Medtronic, Clinatec, and PMT Corp; microelectrode arrays (MEAs) by Blackrock; neurotrophic electrodes (NTE) by Neural Signals; and endovascular arrays (EVA) by Synchron.

Go to [1] and @iburkhart's talk for an overview of these and other systems and associated clinical trials.
For a detailed look at the Neuroport MEA, used by the BrainGate research group in 16 participants across two decades, go to [2] and @DBRubin, Florian Solzbacher. Other work on MEAs is from Vikash Gilja (MEAs on transparent substrate [9], [10]), @maxhodak_who also talk at the event.
More on ECoG from Benjamin Rapoport [13], @MJvanSteensel ([3], stability & longevity [4],[5],[8] epidural vs subdural [6]), @crone_nathan (speech synthesis [7], interfaces' longevity [8] and stability [29]).
Jamie Brannigan, @tomoxl, and @petereyoo on the machine learning side cover endovascular arrays (EVAs) and related work.

Chart 1. Anatomic location of selected sensors [18]

Leuthardt, E. C., Moran, D. W., & Mullen, T. R. (2021).

Implantable systems still have a long way to go. Challenges associated with scalability and portability, electrode stability and yield, information transfer rate, packaging, data processing, and surgical workflows should be solved. A great overview of these challenges by @tgconstandinou and @adyrapeaux in [17].

2. Technological challenges

Back to the table of contents

To put BCIs in the context, go to a paper by @Chad_Bouton, who speaks at the event. The paper covers historical perspectives, challenges, and future directions for implantable BCIs [43]. Here are more papers associated with these challenges from the event's speakers:

Surgical robotics @JoshBedersonMD [19], sonolucent cranioplasty implants @chriskellnerMD [20].
Material integrity and biological tissue encapsulation - @robert_gaunt1 [21].
Safety of the endovascular BCI - @JMoccoMD [22].

New players emerge with new takes on the existing modalities. For example, ECoG systems are being developed by Axoft [23], CorTec [25], and Neurosoft [24], while MEAs are investigated by Paradromics and Neuralink, to name a few.

Moreover,, new modalities emerge at the frontier. I'd highlight functional ultrasound (fUS)/Forest Neurotech ([14], [15]), the neural dust, its original creators [16], and Subsense, which is pushing the modality further.

It is probably also too early to discount non-implantable approaches. Progress was achieved in magnetoencephalography (MEG) through the use of optically pumped magnetometers (OPMs) [26]; the research on spintronics [27] and other modalities is ongoing.

3. Clinical outcomes and user preferences

Back to the table of contents

In addition to research and technology challenges, the clinical transition of BCIs is slowed down by '... an absence of consensus for clinically meaningful performance metrics that can be used when evaluating device efficacy in clinical trials' [30].

To dive deeper into BCI clinical outcomes assessments, go to [30] co-authored by Jamie Brannigan who speaks at the summit and to [31] - a recent FDA's workshop, where two other speakers, @DavidMcMullen and Anthony (TJ) Sutphin participated.

User preferences are a theme adjacent to clinical outcomes and are explored by:

@MJvanSteensel explores what applications would users like to control with a cBCI; (2) what mental strategies would users prefer to use for communications BCI control, and at what stage of their clinical trajectory would users like to be informed about assistive tech and cBCIs [41].
@Abbey_Sawyer1 and David Putrino/@PutrinoLab look into iBCIs as a way to address 'clear disparity ... for people who live with severe quadriplegia, who largely lack access to tools that would enable them to perform daily tasks digitally and communicate effectively with their environment.' [42]
@blaircaseynola/@TeamGleason shares a caregiver's perspective on ALS [51].
iburkhart shares his lived experience of the BCI, including the implant removal [52].
Jennifer French/@NeurotechNetwk will talk about patient advocacy. I enjoyed reading one of the first WSJ articles on BCIs from 2010, where Jennifer shared her thoughts on risks and development avenues [55].

4. Ethical challenges

Back to the table of contents

To be ready for BCIs going into clinics and beyond, the ecosystem must address societal challenges this tech will create or amplify. In addition to safety, agency, and privacy, @dougweberlab, one of the speakers, suggests adding access and equity to the priorities list. The paper suggests starting '... at the development stage, where developers can work to ensure that they follow best practices on how to secure data from BCI systems (such as privacy-by-design) or they critically assess how models are trained so as not to introduce inequities.' [32].

To explore other aspects of neurotech ethics, you could check the work of a few event speakers:

@MJvanSteensel - Neurotech in criminal justice [33].
@NitaFarahany - experimenting with the brain tissue [34] and guiding. principles for the NIH BRAIN initiative , an amazing read for anyone who works on field-building and investing in neurotech.

5. Non-medical BCIs and enhancement

Back to the table of contents

The non-medical or enhancement potential of neurotech does not seem to be addressed directly at the event. Some of its speakers, however, published on related topics:

David Putrino/@PutrinoLab on studying esports for a window into high-performance human-computer interaction [36], [37]; transcranial direct current stimulation (tDCS) and sports performance [38], [39].
A paper co-authored by @IgnacioSaezPhD explores, among other things, when '... such a device [BCI] will be safe enough for individuals without compelling medical needs, and what the limits of their applications should be. ... , can (or, indeed, should) we push beyond that [medical use-cases], into the realm of augmented cognition?' [40].

6. Neurotech translation

Back to the table of contents

A paper on neuronavigation by Isabelle Germano//@MountSinaiNeuro is an interesting case study and inspiration for neurotech translation. It highlights adjunct technology integration and academic and industry partnerships as factors advancing further applications [44].

A detailed view on the translation of neurotechnologies is in a paper by Gerwin Schalk, who does not speak at the event, but addresses the '... technical, clinical and commercial requirements to overcome translational barriers.' [45].

7. Funding

Back to the table of contents

Financing affects the pace of neurotech development and translation. A few event speakers published/spoke on this aspect:

@dougweberlab reviews some of DARPA's major advances in the field of BCI [46].
@genomicsdoc/@khoslaventures spoke about investing in neurotech [47].

To get a better grasp on the financing options for your neurotech ventures, one could read more about focused research organisations (FROs)/@Convergent_FROs [48], private advanced research projects agencies (ARPAs)/@Spec__Tech [49], and check out the neurotech investor directory I complied a year go [50], with 50+ active investors.

8. Asks and wants

Back to the table of contents

I'd love to explore several themes, but they are likely beyond the symposium's perimeter. If you would love to chat about these themes, please ping me.

BCI and national security - some context in @mekosal's work [54].
Solving the interdisciplinarity gap between BCI research and other fields like materials science, chemical engineering, etc., as '... some important disciplines have not been adequately engaged in the BCI research.' [55] I could hardly imagine a BCI breakthrough without new materials or insights in physics, etc.

Chart 2. BCI publications per discipline [55]

Maiseli B, Abdalla AT, Massawe LV, Mbise M, Mkocha K, Nassor NA, Ismail M, Michael J, Kimambo S. (2023)

Diversifying the funding landscape and offering a more comprehensive range of funding options to entrepreneurial scientists. I've mentioned above FROs and pARPAs, but crypto also looks like a promising source (see @FEhrsam of Coinbase launching fUS startup Nudge, Cryptocurrency company Tether announced an investment of $200 million into Blackrock Neurotech, etc.).
I'm curious about weird modalities not yet trialled in humans, anything outside the physics of the dominant implantable approach (ECoG, MEA, DBS, etc.).
BCI + AI, the impact of AI on the BCI research, and the interplay of human intelligence and AI when they are connected with the BCI.
Non-medical BCIs and their general-purpose potential. I've laid out my/e184's perspective here - please reach out if you want to research and build a non-medical BCI.

📬📬📬 Subscribe for notes like this and monthly updates on neurotech and neurotech investment/commercialisation activity.