MedTech · Digital Health · AI Safety

Decision-certified control for future cortical visual prostheses.

AVCR is a history-, geometry-, intervention-, and uncertainty-aware software control layer designed to decide when to act, when to abstain, and what evidence must be preserved before stimulation.

Current stage: pre-clinical, non-human-use research and engineering prototype. No human stimulation, clinical efficacy, or patient-use claim.

Cortical vision is a closed-loop control problem.

Static camera-to-stimulation mapping can ignore person-specific history, gaze, cortical geometry, device state, task, and uncertainty. AVCR is designed to preserve those variables before a future stimulation decision is made.

ReconstructLink scene, gaze, device state, stimulation history, feedback, and task into a query-specific state.
MapRespect cortical geometry and support remapping when gaze, maps, or interfaces change.
CertifyUse uncertainty and response-diameter gates to determine act, limit, wait, or abstain.
VerifyProduce replayable hardware-in-the-loop logs, provenance, and safety evidence.

From scientific framework to a safe bench prototype.

Anahita is developing AVCR as a software and evidence layer for future cortical visual prosthesis platforms. The current P0-B concept is limited to non-human, non-stimulating hardware-in-the-loop verification.

Safety focus
E-stop, watchdog, fault injection, power recovery, timing, and fail-closed behavior.
Evidence focus
Immutable logs, software provenance, replay, and decision certificates.
Translation focus
Clinical-feasibility planning, regulatory readiness, and partner integration.

Abu Dhabi and Hub71 plan

Anahita intends to establish Abu Dhabi as AVCR's MENA engineering, validation, and commercialization base, with one founder relocating long-term if selected.

Month 1

Establish

Set up the local operating structure, relocate a founder, organize the IP and regulatory data room, and map partners.

Month 2

Verify

Complete independently witnessed hardware-in-the-loop verification and recruit clinical, regulatory, engineering, and quality advisors.

Month 3

Partner

Secure a scientific or clinical collaboration pathway and prepare the next financing and regulatory milestones.

About Anahita

Anahita develops multidisciplinary neurotechnology at the intersection of control theory, medical-device software, evidence architecture, and patient-specific adaptation. The AVCR programme is led by Esmaeil Farshi.

Partnerships and investment

We welcome discussions with life-sciences investors, hospitals, neurotechnology teams, implant-platform companies, and verification partners.

farshi@technologist.com

Website: fqxonline.com