NIDCD’s 5-year strategic plan seeks innovations in assistive device technologies

1. NIDCD’s 5-year strategic plan

For more than 30 years, NIDCD-supported research has spurred critical discoveries leading to increasingly effective, evidence-based treatments for the millions of Americans impacted by conditions affecting taste, smell, hearing, balance, voice, speech, or language. To identify our next priorities, NIDCD developed the 2023–2027 NIDCD Strategic Plan: Advancing the Science of Communication to Improve Lives through iterative collaborations with scientific experts, the NIDCD Advisory Council, NIDCD staff, and the public. The result was a plan describing a unified vision organized around six main priorities: (1) basic research to better understand normal function and disordered processes; (2) model systems to inform research and transform findings into more effective treatments; (3) precision medicine approaches to prevention, diagnosis, and treatment; (4) translation of scientific advances into standard clinical care; (5) biomedical data sharing; and (6) advanced technologies to improve prevention, diagnosis, and treatment [1].

To accelerate discoveries in these priority areas, NIDCD encourages investigator-initiated applications to help us better understand the neurological processes underlying disordered communications. We also encourage research focused on improving assistive devices, such as cochlear implants, hearing aids and hearing devices, and brain–computer interfaces (BCIs), that are improving the lives of the millions of Americans impacted by hearing, speech, and other communications disorders [2, 3]. This article outlines current efforts and opportunities for innovations in these areas.

2. Focus on technological innovations

2.1. Cochlear and vestibular implants

Collaborations between neuroscientists, engineers, surgeons, and audiologists have significantly improved cochlear implant devices over the last several decades [4], and recent innovations are further advancing their effectiveness. For instance, electrode arrays are essential components of cochlear implants. NIDCD-supported research is investigating how different array design variables may affect high-quality hearing for people with cochlear implants. Recently, scientists have explored whether inserting a shortened electrode array into the portion of the cochlea that detects high frequencies can help individuals whose hearing loss is limited to the higher frequencies, while preserving their hearing at lower frequencies [5].

NIDCD hopes another tool will inspire innovation to improve cochlear implants. An NIDCD-supported research team developed CCi-MOBILE, an open-source sound processing platform that enables researchers to test and optimize cochlear implant signal processing in real world environments [6]. Researchers and cochlear implant users can collaborate to adjust CI settings to improve speech intelligibility and sound quality in both quiet and noisy settings. NIDCD is excited about the potential for novel cochlear implant studies that leverage this platform.

NIDCD-supported researchers are applying neural engineering technologies to restore balance in individuals with bilateral vestibular loss. A recent project developed a rationally designed implant that senses head motion and artificially stimulates the three semicircular canal branches of the vestibular nerve, thereby partially restoring steady posture and balance in individuals with ototoxic bilateral vestibular loss. This implant is being tested in clinical trials to determine whether its use may be expanded to help individuals with idiopathic adult-onset bilateral vestibular hypofunction, or adult-onset insufficient balance function in both ears [7].

NIDCD seeks continued innovation in the design and manufacture of neurotechnology tailored to address disordered function of the inner ear. Modern microfabrication techniques are not currently used to manufacture cochlear implant electrode arrays. They are assembled by hand which limits the pace of innovation for novel features such as varying mechanical properties across the electrode length to ensure optimal placement of the array without further damaging the inner ear structures. Microfabrication techniques would also enable the application of new biomedical technologies including perfusion of the cochlea with neurotrophins, fiber optics that support optical stimulation, and nanopatterning the surface of the array to discourage fibrosis.

2.2. BCIs

NIDCD is also investing in research that is helping us better understand speech processing in the brain, thereby informing potential approaches to address speech disorders. These findings are also assisting in the development of more effective BCIs that analyze brain signals [8], translating them into commands that are relayed to computers or assistive devices to carry out desired actions [9, 10].

For example, researchers funded by NIDCD and the National Institute of Neurological Disorders and Stroke have used high-density multi-electrode arrays to help us better understand brain processes underlying disorders of communication. In one study, scientists recorded activity from the cortical surface of three people just before epilepsy surgery, and matched the neural recordings with microphone recordings taken as the subjects read syllables aloud [8, 11]. The research team also recorded cortical surface activity in other pre-surgery subjects as they listened to natural, continuous speech [12, 13] to pinpoint the specific neurons that respond to distinctive speech sounds. These investigations help us understand the organization of the speech sensorimotor cortex, and provide insight into language-related disorders such as dyslexia and autism [8, 11, 14, 15].

Discoveries into the functional aspects of the human speech cortex are also being leveraged to develop increasingly effective BCIs that aid in the recognition or production of speech sounds. One such BCI technology has been used to help paralyzed persons with anarthria—which is the inability to articulate speech—regain their ability to communicate. Specifically, NIDCD-supported neural engineers used deep- learning algorithms to create computational models from patterns of recorded cortical activity that were taken from people with anarthria as they attempted to speak. The result was a ‘speech neuroprosthesis’ that decoded a person’s thoughts and turned them directly into words and sentences in real time [12, 13]. NIDCD continues to provide funding to expand the number of participants recruited for these clinical trials [16, 17].

3. Innovation through collaboration: the NIH brain initiative^®

Through partnerships such as the BRAIN Initiative [18], several NIH Institutes are working together to develop communication-specific BCIs [19–24]. By participating in this trans-NIH program, researchers can share information and expertise, coordinate research efforts, and pool resources, thereby gaining a better understanding of the neuronal processes underlying human communication and how BCIs may more effectively assist with disordered processes of communication [25].

For instance, one BRAIN Initiative project is focused on helping people with paralysis generate written communication. Researchers used an intracortical BCI (iBCI; implanted in the cortex) to measure motor cortex activity while a subject whose hand was paralyzed from spinal cord injury thought about typing letters. This information was then processed in real-time by a machine learning computer algorithm, which converted the data to generate words on a screen. The researchers found that typing is faster and more accurate than existing communication BCIs. Although demonstrated as a proof of concept in one patient so far, these findings could help people with paralysis rapidly type without needing to use their hands [21, 22]. This work is currently being extended to improve the performance of iBCI systems through the development of a fully-implantable wireless system [20].

In another NIDCD-supported BRAIN Initiative project, researchers are using a high-performance iBCI to decode motor cortical activity during complex movements. This technology enables people with severe speech and motor impairment to use pro-theses for computer cursor control, handwriting, and speech [23]. Researchers are using these findings help restore continuous motion of the entire body in virtual reality, as well as refining technologies that substantially increase on-screen text generation speed [24].

The iBCI technology is also being used to study how speech is regulated by neuronal ensembles within speech- related motor areas of people with amyotrophic lateral sclerosis, and to create a speech prosthesis to promote conversational speech (i.e. 120–150 spoken words per minute). These results will advance our understanding of human motor cortex function while simultaneously advancing the capabilities of iBCIs for assisted verbal communications [19].

NIDCD encourages the development of ethical AI technologies that facilitate the use of brain signals to transform the design of neuroprostheses that support communication for disabled users. Large repositories of speech signals from diverse speakers will be required to support the development of various machine-learning algorithms. In the near term, speech data could be used to create personalized voices for existing AAC technology [26, 27] and personalized signal processing algorithms that enhance one voice in a noisy background for a hearing aid [28]. Neural engineering experts must collaborate with machine learning experts to mine both the speech and brain activity data to create devices that transform brain signals directly into fluid, expressive speech. The proof of concept that this transformation of nonverbal signals into speech is possible was provided decades ago by the glove-talk system. Modern AI techniques, coupled with big data with speech waveforms and brain signals derived from emerging neurotechnologies, seem poised to provide a modern speech neuroprosthesis that goes far beyond the current state-of-the-art AAC system.

AI and ML may provide a similar opportunity to disrupt hearing aid designs by providing algorithms for brain- controlled hearing aids that discreetly select and then amplify one voice over others while reducing background noise. Modern hearing aids use directional microphones to perform this task with today’s technology. Rational development of a brain-controlled hearing aid would benefit from new forms of neurotechnology that obtain signals from the listening portion of the brain, the auditory cortex, to direct an AI algorithm to enhance one voice and then change the amplification to ‘glimpse’ across other audio signals in the environment before rapidly enhancing a different voice. The rapid exchange of information between the brain and audio amplification would enable hearing aids to operate in ways not currently imagined.

4. Opportunity to innovate: over-the-counter (OTC) hearing aids

Hearing loss is an enormous public health need. By 2050, the world health organization estimates that more than 700 million people—or one in every 10 people around the globe—will have disabling hearing loss. In the United States alone, hearing loss affects an estimated 30 million people [29]. Yet only 1 in 4 U.S. adults, ages 20 and over, who could benefit from hearing aids has used them.

One barrier to accessible hearing aids is the cost to the consumer. In 2013, the average retail cost of hearing aids was $4,700, when bundled with professional services [30]. As the lead federal agency funding research and initiatives to prevent, detect, and treat hearing loss, NIDCD supported decades of research to enhance the affordability and accessibility of hearing health care for adults who have trouble hearing. This research directly informed a landmark 2022 food and drug administration (FDA) announcement that created a new category of OTC hearing aids [31]. The new regulation enables millions of people with mild-to-moderate hearing loss to purchase hearing aids without a medical exam, prescription, or fitting by an audiologist.

NIDCD-supported research data also demonstrate a need for improved hearing aid technology, such as the ability to focus on a single speaker in a crowded room. NIDCD hopes the new FDA regulation will encourage innovation in OTC hearing aid models to transform the market.

Researchers need tools based on widely shared open hardware and software designs that are readily reconfigured to enhance collaborations between scientists, clinicians, and engineers seeking to improve hearing health care. Open designs lower barriers to innovation and provide research platforms that are easily reconfigured to track an evolving set of scientific questions. NIDCD has fostered development of these tools over the past decade with efforts that included a request for applications targeted towards both US- based small businesses [32] and academic groups [33].

Open-source research platforms provide the means for rapid discovery and validation of new hearing aid designs envisioned by a single researcher and validated through a large clinical study. Researchers may leverage open designs to enhance specific features of hearing aid algorithms while retaining other aspects of a shared reference design without modification. For example, clinicians may validate novel noise reduction algorithms uploaded to portable hardware that is commercially available. This approach will allow the study to proceed at scale with minimal engineering support and without requiring a partnership with a hearing aid company.

5. Looking forward

As the fields of neuroscience and engineering continue to converge in new and exciting ways, NIDCD is prioritizing support of research that leverages engineering solutions and seeks the development of improved engineered devices that advance our understanding of the brain. We encourage applications from multidisciplinary teams that propose novel approaches to solving important challenges in the field of communication science.

On behalf of NIDCD, i look forward to strengthening our valued partnerships with scientists, clinicians, and engineers to advance our mission to conduct and support research in the normal and disordered processes of hearing, balance, taste, smell, voice, speech, and language.

Data availability statement

The data that support the findings of this study are available upon reasonable request from the authors.