HEARING_ISO_ARPA-H-SOL-26-154 Amendment 01.pdf

ARPA-H-SOL-26-154, HEARING

INNOVATIVE SOLUTIONS OPENING

FOR

Hearing Enhancement through ARtificially Intelligent

NeurotechnoloGy

HEARING

HEALTH SCIENCE FUTURES (HSF) OFFICE

ARPA-H-SOL-26-154

May 07, 2026 Amendment #01 May 08, 2026

INNOVATIVE SOLUTIONS OPENING (ISO) SUMMARY INFORMATION

1. INTRODUCTION TO HEARING

1.1 BACKGROUND

1.2 PROGRAM OVERVIEW

1.2.1 VISION AND CHALLENGE

1.2.2 GOALS AND DEFINITION OF SUCCESS

1.2.3 DESIRED IMPACT

1.2.4 PERFORMER EXPECTATIONS

1.2.5 OUT-OF-SCOPE EFFORTS

2. THE PROGRAM

2.1 TECHNICAL AREAS

2.2 PROGRAM STRUCTURE

2.2.1 PHASE 1: RESEARCH AND DEVELOPMENT (R&D), 18 MONTHS

2.2.2 PHASE 2: PRE-CLINICAL, 24 MONTHS

2.2.3 PHASE 3: CLINICAL, 12 MONTHS

2.3 METRICS

2.4 GENERAL REQUIREMENTS

3. ELIGIBILITY INFORMATION

3.1 ELIGIBLE PROPOSERS

3.1.1 PROHIBITION OF PERFORMER PARTICIPATION FROM FEDERALLY FUNDED

RESEARCH AND DEVELOPMENT CENTERS (FFRDCs) AND OTHER GOVERNMENT

ENTITIES

3.1.2 CURRENT PROFESSIONAL SUPPORT

3.1.3 NON-U.S. ENTITIES

3.2 SYSTEM FOR AWARD MANAGEMENT (SAM)

4 SUBMISSION PROCESS

4.1 SUBMISSION PROCESS OVERVIEW

4.2 SOLUTION SUMMARY SUBMISSIONS

4.3 FULL PROPOSAL SUBMISSIONS

4.4 SUBMISSION INFORMATION

4.5 PROPRIETARY INFORMATION

5 SUBMISSION REVIEW AND EVALUATION PROCESSES

5.1 CONFORMING SUBMISSIONS

5.2 SOLUTION SUMMARY REVIEW PROCESS

5.3 PROPOSAL REVIEW PROCESS

5.4 EVALUATION CRITERIA FOR PROPOSALS

5.4.1 CRITERIA 1: OVERALL SCIENTIFIC AND TECHNICAL MERIT

5.4.2 CRITERIA 2: PROPOSER’S CAPABILITIES AND/OR RELATED EXPERIENCE26

5.4.3 CRITERIA 3: POTENTIAL CONTRIBUTION TO RELEVANCE TO THE ARPA-H

MISSION AND USER EXPERIENCE

5.4.4 CRITERIA 4: ASSESSMENT OF PROPOSED COST/PRICE

5.5 HANDLING COMPETITION SENSITIVE INFORMATION

5.6 EVALUATION AND AWARD DISCLAIMERS

6. POLICY REQUIREMENTS AND MISCELLANEOUS OTHER INFORMATION

6.1 CONTROLLED UNCLASSIFIED INFORMATION (CUI) ON NON-FERERAL

INFORMATION SYSTEMS

6.2 ORGANIZATIONAL CONFLICTS OF INTEREST (OCI)

6.2.1 AGENCY SUPPLEMENTAL OCI POLICY

6.2.2 RESEARCH SECURITY DISCLOSURES

6.3 INTELLECTUAL PROPERTY

6.4 HUMAN SUBJECTS RESEARCH

6.5 ANIMAL SUBJECTS RESEARCH

6.6 ELECTRONIC INVOICING AND PAYMENTS

6.7 SOFTWARE COMPONENT STANDARDS

6.8 GENOMIC DATA SHARING

6.9 PROCUREMENT OF SYNTHETIC NUCLEIC ACIDS OR BENCHTOP SYNTHESIZERS 32

6.10 DANGEROUS GAIN-OF-FUNCTION RESEARCH

6.11 SAMPLE OT

6.12 ADDITIONAL LINKS

APPENDIX A: Solution Summary Format and Instructions

APPENDIX B: Proposal Format and Instructions

APPENDIX C: Technical and Management Proposal Model

APPENDIX D: Task Description Document Model

APPENDIX E: Cost Proposal Model

APPENDIX F: Administrative and National Policy Document for Other Transactions

APPENDIX G: Sample Other Transaction Agreement

INNOVATIVE SOLUTIONS OPENING (ISO) SUMMARY INFORMATION

FEDERAL AGENCY: Advanced Research Projects Agency for Health (ARPA-H)

PROGRAM TITLE: Hearing Enhancement through ARtificially Intelligent NeurotechnoloGy (HEARING)

ANNOUNCEMENT TYPE: Innovative Solutions Opening (ISO)

ISO SOLICITATION NUMBER: ARPA-H-SOL-26-154

ISO CONTACT: HEARING@arpa-h.gov

ANTICIPATED AWARDS: Multiple Other Transaction (OTs) Agreements

COST SHARING: Encouraged (Optional)

DATES: (All times listed are Eastern Time)

Proposers’ Day: Monday, June 8, 2026 Questions & Answers (Q&A) due date: Monday, June 15th, 2026, 5:00 PM ET Solution Summaries due date: Monday, June 29, 2026, 2:00 PM ET Full Proposals due date: Friday, August 14th, 2026, 2:00 PM ET

WHERE TO SUBMIT:

Proposers’ Day Registration: https://solutions.arpa-h.gov/events

Solution Summaries: https://solutions.arpa-h.gov/Submit-Solution/ Proposals: https://solutions.arpa-h.gov/proposal Questions: https://solutions.arpa-h.gov/Ask-A-Question/

PROPOSERS’ DAY

ARPA-H will host a Proposers’ Day in support of the HEARING Program as described in Special Notice ARPA-H-SN-26-154. The purpose is to provide potential proposers with information on the HEARING program, promote additional discussions, and encourage teaming and networking.

Interested proposers are not required to attend, and materials formally presented during Proposers’ Day will be posted to the ARPA-H Program Website. However, proposers are encouraged to attend to facilitate teaming discussions with other attendees. ARPA-H will not host additional events in-person or virtually for the HEARING program.

ARPA-H will not reimburse potential proposers for participation at Proposers’ Day (or time and effort related to submissions in response to this ISO).

mailto:HEARING@arpa-h.gov https://solutions.arpa-h.gov/events https://solutions.arpa-h.gov/Submit-Solution/ https://solutions.arpa-h.gov/proposal https://solutions.arpa-h.gov/Ask-A-Question/

1. INTRODUCTION TO HEARING

1.1 BACKGROUND

Age-related hearing loss is the second most common health problem in older adults. Yet, the current solutions are inadequate for many people, and there has not been a major innovation in hearing improvement in over 40 years. Today, the primary treatments for hearing loss continue to be hearing aids and cochlear implants, which succeed at amplifying sounds in the environment, but are not selective for the listener’s intent or attention in real time.

Most hearing loss, especially age-related hearing loss, is caused by degeneration of the delicate structures in the cochlea and inner ear that are responsible for converting sound waves into electrochemical signals for the brain. As these sensory cells and neural pathways deteriorate, simply making sounds louder is often not enough; the damaged pathway cannot faithfully transmit detailed sound information to the brain. Current technologies primarily work by amplifying or encoding sound at the ear and then pushing these signals up the compromised auditory pathway. While this approach can provide benefit for some, it often fails to restore natural, effortless hearing, especially in noisy environments or for those with more extensive neural degeneration. This limitation underscores the need for solutions that bypass or compensate for the damaged pathway and work directly with the brain’s own auditory centers to restore meaningful hearing.

To safely access the brain’s auditory centers, there is a critical need for a new class of minimally invasive, brain-connected hearing devices that can capture real-time brain signals reflecting listener intent and use these signals to guide algorithms that both interpret and help to reconstruct sound. Such devices would shift hearing care from ear-only devices to technologies that work directly with the brain’s own processing, enabling a step change beyond incremental improvements in acoustic signal processing.

The HEARING platform will be the first ever, minimally invasive brain‑connected hearing system that can adjust sound based on the listener’s focus, reduce background noise in real time, and fill in missing sound information to restore close to normal hearing for people with mild to severe hearing loss.

1.2 PROGRAM OVERVIEW

1.2.1 VISION AND CHALLENGE

The HEARING program aims to develop a new kind of hearing-assistive platform that directly engages the brain’s hearing centers to improve how people hear in complex, real‑world environments. Many people with hearing loss still struggle to understand speech in noisy places, even when using today’s best hearing aids or implants, because current devices only repair missing pieces of sound when the ear or auditory nerve is damaged or react to sound in the environment, rather than what the listener is trying to focus on. This limits everyday communication and contributes to social isolation, depression, and cognitive decline, particularly for older adults, veterans, and workers from high‑noise occupations. HEARING is designed to solve this problem by shifting hearing restoration from current care at the ear alone to a neurological condition by focusing on the auditory cortex, where the brain separates speech from noise to makes sense of what we hear.

To do this, HEARING will build a closed-loop, brain-connected system that can record (read out), neuromodulate, and stimulate (write in) information in key auditory cortical regions. The program is organized into three (3) Technical Areas (TAs) (Figure 1):

• TA1 Intracortical Device(s): Creation of minimally invasive devices that can interface with key auditory regions through recording and stimulation, without requiring surgical access through the skull, and are suitable for chronic use.

• TA2 Dynamic Sound Modulator: Development of an external, hearing-aid-like hub that wirelessly powers and bi-directionally communicates with the brain interface, adjusting sound in real time based on algorithmic control informed by neural signals.

• TA3 Auditory Read & Write Algorithms: Design and deployment of algorithms within a processing unit that decode what the listener is trying to hear to determine how to deliver or supplement sound so that speech becomes clearer, more complete and understanding speech is less strenuous. This will be accomplished either by enhancing the TA2 device with code to control audio input gain or by transmitting restorative code to stimulate the brain through the TA1 device.

Figure 1. Mechanisms for the HEARING Platform

Together, these components are intended to adjust sound based on the listener’s focus, reduce background noise in real time, and fill in missing sound information to restore hearing close to normal for people with mild to severe hearing loss.

This integrated approach is newly feasible because advances in minimally invasive brain interfaces, flexible electronics, low-power wireless communication and power transfer, real-time decoding of brain activity, and our understanding of auditory processing have matured to the point where they can be combined into practical systems rather than isolated laboratory demonstrations.

1.2.2 GOALS AND DEFINITION OF SUCCESS

Success for HEARING means delivering an integrated prototype that:

• In mild, moderate, and severe hearing loss patients, measurably improve speech understanding and listening comfort in realistic noisy environments.

• Restore hearing ability approaching the performance of age-matched listeners with normal hearing.

• Is positioned on a pathway for clinical translation, including alignment with regulatory, safety, usability, and manufacturability requirements.

1.2.3 DESIRED IMPACT

By funding HEARING, ARPA-H seeks to catalyze a shift from devices that simply make sounds louder to therapies that actively partner with the brain. The envisioned systems will reduce the effort required to follow conversations, provide benefit even when the ear or auditory nerve is severely damaged, and lay the foundation for a new class of cortical hearing therapies that improve communication, mood, and cognitive health for large populations affected by hearing loss. ARPA-H is uniquely positioned to convene multidisciplinary teams, set aggressive system-level milestones, and provide commercialization and regulatory support needed to transform a collection of promising but disconnected research efforts into a practical therapeutic option.

To meet this desired impact, proposers should adhere to the following key tenets:

• More effective than currently available therapies for speech-in-noise and related functional outcomes.

• Minimally invasive so that the relevant populations have broad access and are willing to adopt the technology.

• Sufficiently safe and durable to be used chronically, potentially for the lifetime of the patient.

By shifting the paradigm of hearing loss from an ear-only problem to one that explicitly engages the brain, HEARING will enable future technologies to address broader auditory processing conditions such as tinnitus and hyperacusis and ultimately support a more holistic approach to auditory health.

1.2.4 PERFORMER EXPECTATIONS

There is currently no solution that restores hearing in a fully personalized way that explicitly incorporates the brain’s auditory processing centers. HEARING emphasizes engagement with end users, clinicians, and regulatory experts from the outset. To ensure the perspectives necessary to achieve program milestones and deliverables:

• Performers are required to include representative end user, clinical, and regulatory expertise as part of their teams.

• Performers must form interdisciplinary teams capable of delivering innovative, integrated solutions spanning all three (3) TAs.

The end of program deliverable is a fully integrated platform with a bidirectional brain computer interface (BCI) system that can restore a patient’s hearing with bilateral mild, moderate, and severe hearing loss demonstrated in a first-in-human (FIH) trial (see Section 2.2).

The results of the FIH study should provide statistically significant and clinically meaningful improvement in one’s speech-in-noise perception and reduced cognitive burden compared to baseline state-of-the-art hearing aids.

1.2.5 OUT-OF-SCOPE EFFORTS

The following are considered out of scope and may be deemed non-responsive to the HEARING program:

• Proposals that do not address all TAs and all program phases, or that offer only incremental advances in the state of the art for BCIs, wireless power and data transmission, or decoder/encoder algorithms.

• Proposals that deliver the final TA1 device for intracortical interfacing via surgical access through the skull (e.g., craniotomy, craniectomy).

• Solutions that are designed with batteries implanted subdurally or deeper into the brain on the final TA1 device.

• Solutions that require form factors that would be cumbersome, limit normal daily living and locomotion, or that only function in highly controlled environments.

• Solutions that are not supported by sufficient preliminary evidence to justify the innovation claims and feasibility within the program timeframe.

Proposers are encouraged to carefully read the detailed description of the program’s requirements below in Section 2 before responding to this solicitation.

2. THE PROGRAM

2.1 TECHNICAL AREAS

The HEARING program takes a new approach to hearing restoration by focusing on how sound is processed in the auditory cortex, rather than at the ear.

HEARING will build a closed-loop system that:

1. Reads activity from the auditory cortex,

2. Uses it to control sound entering the ear in real time, and

3. Writes back missing information through targeted stimulation.

To develop such a platform, HEARING will occur over three-phases in a multi-disciplinary approach. Performer efforts will be organized around three (3) tightly integrated TAs whose outputs must be combined into a single, functioning platform:

• TA1-Intracortical Device(s): TA1 will deliver minimally invasive neural interfaces to key auditory cortical regions, enabling HEARING technologies to be accessible to patients at scale. Approaches may leverage any technology that allows final device delivery to target the auditory cortex without requiring surgical access through the skull (e.g., no craniotomy or craniectomy). Neural interfaces are not limited to electronic, biohybrid, ultrasound, or other specific modalities, but proposed technologies must be sufficiently mature at the time of award to meet program schedules and milestones. TA1 devices must support wireless bidirectional communication with TA2 devices. ARPA-H will assess record and stimulation, and safety in vivo performance capabilities as key end of phase deliverables.

• TA2-Dynamic Sound Modulator: TA2 will deliver an external wearable device that functions similarly to a hearing aid and serves as the hub between TA1 hardware and TA3 software. This device must: (1) Operate within a 16-hour total power budget under read-only conditions, (2) deliver power and support bi-directional wireless communication with TA1 devices, (3) be developed in a form factor that is not a burden to the daily living activities of the patient, and (4) interface with external devices (e.g., smartphones, clinical programming tools) to support optimization and adjustment of hearing performance. TA2 devices will be evaluated, in conjunction with TA1 devices, for sustained operational performance and safety in vivo as end-of-phase deliverables.

• TA3 Auditory Read & Write Algorithms: TA3 will deliver the first auditory BCI algorithms capable of both reading from and writing to the auditory cortex while controlling audio signals amplified through the ear (TA2). Algorithms are expected to:

(1) Personalize to individual users by leveraging sensory-neural profiles to determine how auditory information should be controlled in the TA2 device, (2) drive neuromodulation at key auditory regions and/or write in missing frequencies via TA1 devices at the auditory cortex, and (3) balance trade-offs among latency, decision-window length, decoding accuracy, model personalization time, and re-calibration frequency.

o Performers that collect data from neurosurgical populations (e.g., intracranial recording) and/or individuals with hearing loss (e.g., EEG or functional magnetic resonance imaging (fMRI)) are required to obtain Institutional Review Board (IRB) protocol approval following the time of award and justify data harmonization plans that align collected data to TA1 devices. TA3 is expected to drive both software behavior and clinical decision-making in assessing and demonstrating functional hearing recovery at the end of each phase.

o Software for real-time decoding/encoding is not required to be processed on the TA2 device (e.g., cloud computing via smart phone). However, the required metrics for latency still must be addressed and justified.

All proposals must address all three (3) TAs (TA1-3) across the program. TA1 and TA2 technologies may be combined into a single hardware platform if that better fits the proposed solution, provided all functional requirements can be met. Given the integration challenges inherent in HEARING, the program is best suited for teams that can address the full system to avoid development of orphaned technologies that are not compatible with a complete platform. Interim technologies in development do not need to be wireless or delivered minimally invasively, except to meet end-of-phase requirements (see Section 2.2).

Where appropriate, proposers are encouraged to identify components (e.g., interfaces, data formats, benchmarking tools, or non-core software utilities) that could be made available to other performers under open or permissive licenses to facilitate interoperability and broader impact, while preserving commercially valuable intellectual property (IP). Proposers are encouraged to outline plans for responsible data sharing and interoperability (e.g., contribution to common data standards or shared reference datasets) consistent with privacy, regulatory, and contractual requirements.

TA1 Intracortical Device(s) This sub-section outlines the broad scope of the TA1 objectives. Development of the TA1 device will require coordinated integration with TA2 wireless power and communication specifications and with TA3 requirements for precise targeting of specific auditory cortical regions to restore one’s hearing. A successful proposal should consider each of the following, and include strategies and information to achieve each goal:

• Delivery approach and target auditory regions:

o Describe the minimally invasive delivery approach (e.g., endovascular, sub-galeal, transnasal, etc.) for the intracortical device(s) in relevant animal models and human use including pre-operative planning, intra-procedural guidance, and post-procedural verification to ensure accurate placement and coverage of cortical regions.

o Specify target locations within the auditory cortex (e.g., Heschl’s gyrus, superior temporal gyrus, planar temporale, etc.) and justify their relevance to hearing restoration, including the level of spatial and temporal resolution to the approach.

• Device design, operation parameters, and interfaces:

o Specify device form factor(s), size, and configuration and how this supports minimally invasive delivery and fit within anatomical constraints.

o Read-out (recording) capabilities and data outputs:

Describe how the device will sense or record activity from target regions of the auditory cortex, as appropriate to the chosen modality.

Summarize key performance characteristics and explain how noise and artifacts will be managed.

Define types of data collected under TA1 to support algorithm development, personalization and drift correction in TA3.

Validate newly developed technology against state-of-the-art methods (e.g., intracortical electrodes, scalp EEG, electrocorticography (ECoG) or comparable). Proposers must define quantitative metrics and validation methods to demonstrate reliability and fidelity of the new recording technology (e.g., correlation). Proposers must justify their chosen metrics and thresholds, and provide a rationale for how their approach establishes equivalence or improvement over existing methods (see Section 2.3)

- For technologies that will not leverage electrophysiological outputs, teams must identify and justify state of the art technologies for validation of recording/stimulation output.

o Write-in (stimulation) capabilities:

Describe how stimulation and/or modulation will be delivered to targeted regions of the auditory cortex.

Define and justify quantitative metrics for stimulation accuracy and specificity, (e.g., behavioral response rates or comparable) to verify that stimulation reliably evokes intended auditory percepts and behaviors in animal models. Validation should include comparison to state-of-the-art methods (see Section 2.3).

• Power budget, communication, and latency:

o Describe the overall power budget and communication strategy for the TA1 device(s), including mechanism for power supplied from TA2, how data will be exchanged reliably, and plan for thermal management (see Section 2.3) and timing requirements to support real-time operation with TA2 and TA3.

• Biocompatibility, safety, and stability for chronic implantation:

o Outline a plan to validate the performance of the TA1 device(s) in vivo including in small and/or large animal models as relevant.

o Describe how you will ensure longevity (see Section 2.3) safety and stability of the implant, including material biocompatibility, monitoring and mitigation of tissue/device changes over time, and built-in safety features within the device and clinical procedures in case of device malfunction.

• Manufacturing and scaling for FIH trials:

o Summarize how fabrication and assembly plans for the intracortical device(s) from prototype through pilot scale, will be developed. Include identification of potential facilities/partners and describe how standardization and reproducibility for FIH studies will be ensured during the program.

• End of Phase Demonstrations:

o Proposals should include detailed strategies for demonstrating system capabilities in small and large animals, and humans (see Section 2.3).

• Integration and Management:

o Identify the integration lead responsible for TA1 sub-component integration and coordination with TA2 and TA3.

o Discuss anticipated obstacles that could require a revision in the scope of work or milestones and propose alternative approaches.

o A detailed schedule or timeline for each milestone, deliverable, and the overall goal (see Section 2.3).

TA1 program metrics and timeline are outlined in Tables 1, 2 and 4 of Section 2.3, and will increase in difficulty and complexity over the course of the HEARING program. Proposers must align technical requirements with program metrics and timeline for full consideration.

Proposers are encouraged to include additional metrics and key milestones as relevant to their respective solution.

TA2 Dynamic Sound Modulator This sub-section outlines the broad scope of the TA2 objectives. TA2 serves as the central hub between TA1 and TA3 and is critical to the success of HEARING. A successful proposal should consider each of the following, and include strategies and information to achieve each goal:

• Overall device role, form factor, and usability:

o Describe how the TA2 device functions as the primary interface between the user, TA1 device(s), and TA3 algorithms, including operation in real-world environments and support functional hearing restoration metrics (see Section 2.3).

o Specify the intended form factor(s) and justify why these are compatible with long-term daily use for the relevant hearing-loss populations, not limited to details such as size and weight.

o Outline user interface capabilities and strategies for engaging end-user feedback.

• Audio pathway and closed-loop performance:

o Describe how environmental sounds (i.e. speech, wind, traffic, music, etc.) will be captured, processed, and delivered to the ear, highlighting the main components and processing steps of this audio pathway.

o Explain the key performance goals for this pathway and how it will support closed-loop interaction with TA1 and TA3.

• Power, Communication, and TA1 Coupling:

o Provide an overall power budget for TA2, including different operating modes, and how the final system will achieve ≥16 hours of continuous operation under read-only conditions, as well as the expected operating time under read-and-stimulate conditions.

o Describe the power source, capacity, charging method, and include strategies for safe and reliable daily use for at least one year (see Section 2.3).

o Specify wireless communication modalities between TA2 and TA1, and between TA3 and external devices, including expected data rates, robustness to interference, security, and data formats or protocols that support interoperability and firmware updates.

o Explain how TA2 will deliver power wirelessly to TA1 devices, including the basic mechanism, expected efficiency, and alignment tolerances.

• Compute Architecture and External Interfaces:

o Describe the architecture that will be required for signal processing, including which elements of TA3 decoding/encoding will operate on the TA2 device and/or off-device, and justify how closed-loop timing requirements and power constraints will be satisfied.

o Detail capabilities and validation plan for external device interfacing such as setup and personalization for users and clinicians, and which parameters will be adjusted versus locked.

• Safety, Security and Reliability for Real-World Robustness:

o Identify key safety and security considerations and features for various scenarios, including scenarios such as hardware malfunction, loss of communication between TA1 and/or TA3, potential hacking or cybersecurity breaches. Describe how the system will fail safely in each scenario.

o Plan to manage thermal fluctuations and ensure comfortable and safe temperature during use (see Section 2.3).

o Highlight strategies for reliability and durability during daily life usage and methods to validate robustness.

• Manufacturing and scaling for FIH Trials:

o Summarize fabrication and assembly plans for the wearable device from, prototype through pilot scale, will be developed. Include identification of potential facilities/partners and describe how standardization and reproducibility will be ensured for FIH studies.

• End of Phase Demonstrations:

o Proposals should include detailed strategies for demonstrating devices in small and large animals, and humans (see Section 2.3).

• Clinical protocols:

o Describe the proposed clinical protocols and end-user usability evaluations for device application during Phase 3 FIH Trials (see Section 2.3).

• Integration and Management:

o Identify the integration lead responsible for TA2 sub-component integration and coordination with TA1 and TA3.

o Discuss anticipated obstacles that could require a revision in the scope of work or milestones and propose alternative approaches.

o A detailed schedule or timeline for each milestone, deliverable, and the overall goal (see Section 2.3).

TA2 metrics and timeline are outlined in Tables 1, 2 and 4 of Section 2.3, and will increase in difficulty and complexity over the course of the HEARING program. Proposers must align TA2 proposal elements with the program metrics and schedule for full consideration. Proposers are encouraged to include additional metrics and key milestones as relevant to their respective solution.

TA3 Auditory Read & Write Algorithms This sub-section outlines the broad scope of the TA3 objectives. TA3 responsibilities include the development of real-time decoder/encoder software of auditory signals, clinical expertise in neurology and audiology, and will facilitate patient engagement. TA3 will require integration with TA1 and TA2 hardware requirements relevant to controlling the physiology of auditory processing in Phase 3. A successful proposal should consider each of the following, and include strategies and information to achieve each goal:

• Overall device role and system context:

o Describe how TA3 will function as the “controller” for the HEARING system, linking signals received from TA1 and TA2 for closed-loop action to restore one’s hearing.

o Explain how TA3’s algorithms and clinical workflows will support program metrics for functional hearing restoration across all phases (see Section 2.3).

• Data strategy and model development:

o Describe and justify the types, sources, and size of data that will be used to develop and train TA3 algorithms, including plans for Institutional Review Board (IRB) protocol approval to conduct human research where applicable.

o Provide a data collection and quality-control plan adequate to train and validate models.

o Explain how data from different sources and TA1 hardware configurations will be aligned and harmonized to support model development, comparison across users and sites, and longitudinal analysis, as relevant.

o Outline approaches for building, training, validating, and updating models over time. Include a plan to achieve personalization including personalization training time of ≤30 minutes, and strategy to meet proposer set goals for model drift and recalibration by the end of Phase 3 (see Section 2.3).

o Describe how TA3 will manage data security, privacy, consent, and how common data formats/standards or shared tools will be implemented, as appropriate.

• Algorithmic strategies for read and write:

o Auditory reading algorithm capability:

Describe how TA3 will interpret signals from target regions of the auditory cortex and/or related neural or physiological signals to estimate user intent, listening state, or other relevant features of auditory perception.

Specify key performance goals and how these will be measured and reported.

o Auditory intent algorithm capability:

Describe how TA3 will control TA2 device auditory processing and delivery to the ear to improve speech-in-noise perception and listener intent.

Define initial safety and efficacy criteria for modulating auditory input.

o Auditory writing algorithm capability:

Describe how TA3 will control TA1 delivery of targeted stimulation or modulation that restores or enhances auditory perception at target cortical regions.

Explain how stimulation and encoding strategies will be personalized to an individual’s sensory-neural profile, including how targets, patterns, and intensities will be selected and adapted.

Define initial safety and efficacy criteria for stimulation paradigms and how these will be refined through clinical testing.

• Closed-loop control:

o Describe the closed-loop control architecture and framework that links TA1 and

TA2 devices with algorithmic processing and output, including information used as feedback.

o Explain how TA3 will balance key trade-offs such as latency, decision-window lengths, decoding accuracy, stimulation timing, and performance specifications that align with real-world use.

o Describe safety mechanisms in the software strategy and how they will integrate with TA1 and TA2 safety features.

o Outline plans for software verification and validation before use in Phase 3 studies.

• Clinical protocols, evaluation, and outcome measures:

o Describe the proposed clinical and behavioral testing protocols that will be used to evaluate TA3 performance during Phases 1-3 for audibility, lab-assessments for speech-in-noise (SIN) perception, and real-world world assessments for SIN perception (see Section 2.3).

Example gold-standard assessments include but are not limited to:

audiogram, real ear measurements, Quick speech-in-noise (QuickSIN), words in noise (WIN), and coordinate response measure (CRM).

Proposers must identify which tests will be used for assessing functional hearing assessments of auditory perception and justify any adaptations that will be made to meet the requirements of the HEARING Program.

o Explain how TA3 will work with clinical teams for surgical protocols, rehabilitation, and technology training to interpret study protocols over the course of the program. In addition, explain how technology training will be provided to both clinicians and patients to ensure effective use and understanding of the technology developed in the program.

• End of Phase Demonstrations:

o Proposals should include detailed strategies for demonstrating software capabilities in human populations, as relevant to each phase (see Section 2.3).

• Integration and management:

o Identify the TA3 lead responsible for coordinating algorithm development, data collection, and clinical activities, and for integration with TA1 and TA2 teams.

o Describe strategies for iterative integration and testing with TA1 and TA2, and key milestones for system-level demonstrations.

o Discuss anticipated technical, clinical, and regulatory risks and propose mitigation strategies.

o Provide a schedule or timeline for TA3 milestones and deliverables and explain how these align with overall HEARING program metrics and Phase 1–3 deliverables (see Section 2.3).

TA3 metrics and timeline are outlined in Tables 1, 3 and 4 of Section 2.3, and will increase in difficulty and complexity over the course of the HEARING program. Proposers must align TA3 proposal elements with the program metrics and schedule for full consideration. Proposers are encouraged to include additional metrics and key milestones as relevant to their respective solution.

Regulatory Requirements The HEARING program is intended to end with a completed FIH trial for mild, moderate, and severe hearing loss populations demonstrating the benefit of developed technologies.

Regulatory milestones will be considered key deliverables to inform the government’s decision to continue work into subsequent Phases after Phase 1, for all technologies developed as relevant. Proposals should lay out clear and concise regulatory plans for engagement with the U.S. Food and Drug Administration (FDA), not limited to details including proposed product classification, regulatory pathway, timeline to meet key regulatory milestones within each phase, security measures, locked prototype(s), and meeting frequency. The HEARING program is open to the integration of technologies and sub-components that have been previously approved for clinical use by the FDA.

2.2 PROGRAM STRUCTURE

The HEARING program is structured as a 4.5-year effort consisting of three (3) phases, as shown in Figure 2. HEARING Phase 1 (18 months) includes realistic and measurable goals for performers to ensure the success of the program, including checkpoints at the transition between phases (see Section 2.3). To progress to HEARING Phase 3 (FIH studies), performers must obtain relevant IND/IDE approvals or equivalent documentation from the U.S. Food and Drug Administration (FDA).

Figure 2. Program Structure and General Overview over five program years (PY).

2.2.1 PHASE 1: RESEARCH AND DEVELOPMENT (R&D), 18 MONTHS

Performing teams are expected to collaboratively refine their plans for developing and integrating their technologies. Performers in TA1 and TA2 will collaborate to develop the intracortical device(s) and wearable dynamic sound modulator, with the goal of establishing a prototype system design that can demonstrate proof-of-concept feasibility and preliminary safety in small and/or large animal trials before the end of Phase 1. Devices are not required to be in their final integrated state, only developed sufficiently to meet Phase 1 goals.

Furthermore, devices may be delivered through the skull of the selected animal model(s);

however, device powering and communication must be demonstrated wirelessly by the end of Phase 1.

Performers will initiate “read and write” algorithm (TA3) development and begin collecting data from relevant patient populations to enable progress toward the overall HEARING end-goal.

During this phase, performer teams are expected to collaborate and communicate frequently as they develop their respective technologies and to pivot rapidly to alternative approaches if an initial method proves infeasible.

Teams will be expected to complete an early formal feedback meeting with the US FDA.

Progress reviews will be conducted at the end of Phase 1 to assess the capabilities of performer technologies and readiness to transition to Phase 2.

2.2.2 PHASE 2: PRE-CLINICAL, 24 MONTHS

Performers will be required to obtain relevant IND/IDE approval or equivalent permissions from the U.S. FDA during Phase 2 to conduct FIH trials in Phase 3. During this period, performers are expected to lock in their final prototype designs for the intracortical device(s) (TA1) and dynamic sound-modulating wearable (TA2), and to demonstrate safety and performance in large animals, including closed-loop capabilities (see Section 2.3). Device delivery must occur through the developed minimally invasive procedure to target auditory cortical regions. Where possible, performers conducting pre-clinical studies to meet regulatory requirements are encouraged to leverage the same or similar animal populations.

Where appropriate and with all necessary regulatory approvals and IRB-protocol approvals in place, TA1 and TA2 performers may also conduct limited human feasibility studies during Phase 2 to further de-risk their technologies prior to FIH trials. TA3 performers may continue to collect only the data necessary to advance algorithm development and are expected to begin formal validation of their algorithms, including implementation of data harmonization plans with TA1 and TA2 devices.

Throughout Phase 2, TAs 1-3 will begin systematic integration of their systems and test them according to their proposed strategies. Before the end of Phase 2, it is expected that performers will initiate clinical trial patient recruitment and obtain informed consent in preparation for Phase 3.

2.2.3 PHASE 3: CLINICAL, 12 MONTHS

Performers will fully integrate TAs 1-3 to conduct FIH studies in a total of six (6) patients with mild, moderate, and severe hearing loss. Performing teams will work together to harmonize TA1 and TA2 outputs with TA3 read/write algorithms and to ensure the safe delivery and operation of cortical devices. Success will be measured based on the ability to restore hearing in alignment with the Phase 3 metrics, including user satisfaction (see Section 2.3).

2.3 METRICS

To evaluate the effectiveness of a proposed solution in achieving the stated program objectives, the following program metrics will serve as the basis for determination of satisfactory progress to warrant program continuation. Although the program metrics are specified below, proposers should note that the Government has identified these goals with the intention of bounding the scope of effort while affording maximum flexibility, creativity, and innovation or proposed solutions to the goals. Proposals should cite the quantitative and qualitative success criteria that the effort will achieve at each phase’s program milestone, as well as the measurement of intermediary metrics. If the metrics are not meaningful for a particular case, proposing teams are expected to provide their own metrics and describe the quantitative improvement that those metrics represent over the state-of-the-art.

Metrics and Deliverables

Table 1. Overall Program Metrics and Deliverables Overall Program Metrics and Deliverables

HEARING Restorative System

Development of (1) ≥1 device(s) capable of being delivered to the auditory cortex through minimally invasive procedures with recording and stimulation functionalities; and (2) ≥1 wearable, dynamic sound modulator with wireless power and data transmission to enable closed-loop system functionality.

Neuroprocessing Integrate a neuroprocessing algorithm capable

Algorithm of personalization to new patients in ≤30 minutes, interpreting neural signals from the auditory cortex to dynamically control the sound modulator in real time and deliver signals for precise stimulation for hearing restoration.

First-in-human (FIH) studies of HEARING device in hearing loss populations

Demonstrate closed-loop system efficacy in restoring hearing function across the outcomes below in ≥6 patients with mild, moderate, and severe hearing loss, compared to unaided hearing and patients' current hearing assistive device.

Speech-in-quiet Demonstrate restored audibility, validated by ≥2 gold-standard assessments (e.g., audiogram, real-ear measurements, or comparable).

Speech-in-noise Demonstrate significant improvement in ≥2 measurable speech-in-noise outcomes (e.g., SNR, intelligibility, or equivalent) using ≥4 relevant assessments (e.g., QuickSIN, WIN, HINT, CRM, or comparable).

Real-world user-feedback

Demonstrate significant improvement in patients’ ability to hear speech in real-world environments, validated by ≥2 assessments (e.g., rHHI, APHAB, SSQ, or comparable).

Definitions: QuickSIN: quick speech in noise; WIN: words in noise; HINT: hearing in noise test; CRM: coordinate response measure; rHHI: revised hearing handicap inventory; APHAB:

abbreviated profile of hearing aid benefit; SSQ: speech, spatial and qualities of hearing scale; SNR: signal-to-noise ratio

Table 2. TA1 & TA2 Phase 1-2 Metrics and Deliverables The expected metrics of HEARING TA1 & 2 Phases 1-2 are listed below.

TA1 & 2 Phase 1 Metrics and Deliverables Prototype Development Produce ≥1 prototypes for (1) External processing and communication device, (2) brain computer interface(s).

Proof-of-concept Demonstration

Demonstrate capabilities below in vivo using six small or large animal models.

Auditory Behavioral Task

Design and implement a behavioral task in animal models to demonstrate hardware capabilities to record and interpret auditory signals and evoke percepts via stimulation. Proposers must also establish and justify quantitative metrics that will verify the accuracy and specificity of the evoked percepts, ensuring that these percepts reliably result in the intended behavior (see Section 2.1 TA1 for additional details).

Spatial Resolution Ability to record and stimulate with effective spatial resolution of ≤1 mm3 per addressable volume at the level of neural tissue.

Sampling Rate Proposer set target for sampling rate to record relevant neural signals that will inform TA2 and TA3 technologies. Performer must meet ≥75% target by the end of Phase 1.

Accuracy Validate newly developed technology for the HEARING program against state-of-the-art methods for recording and stimulation capabilities (e.g., scalp EEG, ECoG, or comparable). Define quantitative metrics and thresholds that will provide evidence for the demonstration of equivalence or improvement over the state-of-the-art (see Section 2.1 TA1 for additional details).

Powering Demonstrate continuous operation for recording only, on a single charge for ≥8 hours on a single charge.

Latency The system must demonstrate ≤10 ms unidirectional latency for both recording (neural event to signal availability) and stimulation (command issuance to stimulus delivery), measured under standard operating conditions.

Longevity Maintain device performance and stability for ≥3 months.

Delivery procedure Demonstrate ≥1 surgical delivery procedures to implant a mock-prototype device to a target cortical region in small or large animals and assess relevant safety metrics as proposed.

Note: Only the form-factor of the prototype(s) and other relevant components need to be demonstrated, a fully integrated device is not required.

Regulatory Engagement Submit evidence of US FDA partner engagement, including relevant contacts/offices, and revised regulatory plans and milestones.

TA1 & 2 Phase 2 Program Metrics and Deliverables Regulatory Approval IND/IDE approved device(s) for (1) in ear listening device, (2) external to ear wireless powering and data transmission wearable, and (3) minimally invasive brain computer interface(s).

Functional Performance for Closed-Loop Capabilities in Large Animal Models

Using six large animal models (e.g., sheep, bovine, non-human primate models, etc.), demonstrate closed loop capabilities of the technology developed in the program below.

Auditory Behavioral Task

Design and implement a behavioral task in animal models to demonstrate closed-loop capabilities to record and interpret auditory signals and evoke percepts via stimulation. Proposers must also establish and justify quantitative metrics that will verify the accuracy and specificity of the evoked percepts, ensuring that these percepts reliably result in the intended behavior.

Sampling Rate Performer must meet ≥95% target rate by end of Phase 2.

Powering Demonstrate continuous operation for record only, on a single charge for ≥16 hours.

Closed-loop continuous operation for read and write should be demonstrated for ≥4 hours on a single charge.

Latency The closed-loop system should operate ≤50 ms from signal recording to response.

Longevity Maintain performance and safety for ≥6 months.

Delivery procedure Demonstrate delivery protocol(s) to implant fully integrated

TA1 and TA2 device(s) to target cortical regions.

Safety Proposer set metrics and validation to demonstrate safety, must include thermal safety ≤1oC rise in tissue being recorded or stimulated.

Table 3. TA3 Phase 1-2 Metrics and Deliverables The expected metrics for HEARING TA3 Phases 1-2 are below.

TA 3 Phase 1 Metrics and Deliverables Software proof-of-concept performance using human data

Performers will demonstrate software capabilities below using human data for read and write, as relevant. At minimum, data should be collected based on intracranial recordings (e.g., neurosurgery patient populations).

Personalization training time

Data collection to personalizing algorithms to new users should require ≤1.5 hours of training data.

Model drift Proposer set target and metrics to monitor drift and re-calibration frequency. By the end of Phase 1, performers should achieve ≥90% of target.

Neuroprocessing algorithm functionality in neurosurgical patient population

Assess algorithm functionality to improve hearing function across the outcomes below in ≥6 neurosurgical patients or comparable, compared to unaided baseline.

Speech-in-quiet Demonstrate patient audibility using ≥2 gold-assessments (e.g., audiogram, real-ear measurements, or equivalent).

Speech-in-noise Demonstrate improvement in ≥2 measurable speech-in-noise outcomes (e.g., SNR, intelligibility, or comparable) using ≥2 relevant assessments.

TA3 Phase 2 Metrics and Deliverables Software validation performance using human data

Performers will validate software capabilities below using human data for read and write, as relevant.

Personalization training time

Data collection to personalizing algorithms to new users should require ≤1 hour of training data.

Model drift Proposer set target and metrics to monitor drift and re-calibration frequency. By the end of Phase 2, performers should achieve ≥75% of target.

Closed-loop neuroprocessing algorithm capabilities in neurosurgical patient population

Verify closed-loop system efficacy to improve hearing function across the outcomes below in ≥6 neurosurgical patients or comparable, compared to unaided baseline.

Speech-in-quiet Demonstrate patient audibility using ≥2 gold-standard assessments (e.g., audiogram, real-ear measurements, or comparable).

Speech-in-noise Demonstrate improvement in ≥2 measurable speech-in-noise outcomes (e.g., SNR, intelligibility, or comparable) using ≥2 relevant assessments.

Table 4. TA 1-3 Phase 3 Metrics and Deliverables The expected metrics for HEARING TAs 1-3 Phase 3 are below.

TA1-3 Phase 3 Metrics and Deliverables First-in-human (FIH) studies of HEARING device in hearing loss populations.

Demonstrate closed-loop system efficacy to restore hearing function across the outcomes below in ≥6 patients with mild, moderate, and severe hearing loss, compared to unaided and standard hearing aids.

Speech-in-quiet Demonstrate restored audibility, validated by ≥2 gold-standard assessments (e.g., audiogram, real-ear measurements, or comparable).

Speech-in-noise Demonstrate significant improvement in improvement in ≥2 measurable speech-in-noise outcomes (e.g., SNR, intelligibility, or comparable) using ≥4 relevant assessments

Real-world user-feedback Demonstrate significant improvement in patients’ ability to hear speech in real-world environments, validated by ≥2 relevant assessments (e.g., rHHI, APHAB, SSQ, or comparable).

Hardware performance in FIH studies

Demonstrate the hardware capabilities below during the FIH studies.

Power Proposer set target in hours for continuous closed-loop operation of record and stimulate.

Longevity Maintain device performance and stability for ≥1 year.

Delivery procedure Deliver device(s) in single procedure to target auditory cortical regions (e.g., Heschl’s gyrus, superior temporal gyrus, planar temporale, etc.)

Software performance in FIH studies

Demonstrate the software capabilities below during the FIH studies.

Personalization training time Data collection to personalizing algorithms to new users should require ≤30 minutes of training data.

Model drift Proposer set target and metrics to monitor drift and re-calibration frequency. By the end of Phase 3, performers should achieve ≥100% of target.

2.4 GENERAL REQUIREMENTS

Project System Integration Lead

Due to the complexity and performance objectives of the HEARING platform, proposals must identify a lead program integrator with a proven track record of managing and integrating disparate technologies.