The research, led by Justin Cottrell of Yale University together with colleagues at New York University Grossman School of Medicine, enrolled four patients aged between 5 months and 25 years who were undergoing preoperative evaluation for cochlear or auditory brainstem implantation between September 2022 and June 2024. Rather than adding a separate scanning session, the team modified the standard clinical MRI protocol, trimming conventional sequences to make room for a 1-millimeter isotropic three-dimensional FGATIR acquisition. The total scan time increased only from 32 minutes 28 seconds to 36 minutes 11 seconds, adding roughly 5 minutes of anesthesia time—a difference the investigators judged small enough to keep within institutional review board limits while preserving equivalent diagnostic capability for surgical planning.

FGATIR works by exploiting T1 relaxation properties of brain tissue. It is essentially a modified magnetization-prepared rapid gradient echo, or MPRAGE, sequence in which the inversion time is shortened from 900 to 410 milliseconds. This adjustment suppresses white matter signal and dramatically enhances gray–white matter contrast in subcortical structures such as the thalamus, basal ganglia, and brainstem—precisely where the ascending auditory pathways reside. Previous work had shown that FGATIR, especially when denoised with a convolutional neural network, can depict the size, orientation, and volume of auditory brainstem nuclei in healthy adults. What remained unknown was whether the technique could resolve these tiny structures in pediatric patients with complex hearing pathologies.

To find out, the team had two board-certified neuroradiologists with brainstem expertise review all FGATIR images. Critically, the readers were blinded to clinical and audiometric data, including the laterality of hearing loss, to conventional imaging findings, and to any implantation decisions. Their assessment was qualitative and reached by consensus, since no pediatric normative FGATIR dataset exists against which quantitative criteria could be defined. The series was explicitly designed to test technical feasibility, not diagnostic performance.

The results were striking. In three of the four patients, all aged 2 years or older with more adult-like brainstem myelination, FGATIR successfully resolved most of the relevant upper auditory pathway structures. The most consistent finding was a difference in the size or signal intensity of auditory nuclei, seen most often in the inferior colliculus. In a 4-year-old boy with auditory neuropathy spectrum disorder and choreoathetoid cerebral palsy, FGATIR revealed a reduced left medial geniculate nucleus, left inferior colliculus, and brachium of the inferior colliculus, along with a reduced right trapezoid body. In a 3-year-old with bilateral cochlear hypoplasia and absent cochlear nerves, the sequence showed subtly diminished myelin signal in the left inferior colliculus and its brachium. And in a 15-year-old with Wildervanck syndrome and no cochleae at all, both inferior colliculi appeared reduced in size, the right smaller than the left.

Notably, these structural differences were not apparent on the conventional imaging available in these cases. The findings were also anatomically concordant rather than random: because the principal ascending auditory projection crosses at the trapezoid body level, predominantly right-sided deprivation would be expected to affect the right trapezoid body together with the contralateral left inferior colliculus, brachium, and medial geniculate nucleus—exactly the pattern observed in the first case. Intriguingly, one patient who had also undergone positron-emission tomography showed asymmetric fluorodeoxyglucose uptake in the inferior colliculus on the same side as the FGATIR abnormality, a finding previously associated with hearing loss.

The fourth patient, a 5-month-old infant, proved the technique’s key limitation. FGATIR showed poor myelin contrast in this baby because the sequence’s inversion, echo, and repetition times are optimized for the mature subcortical T1 values found in patients older than about 2 years, whose central myelination is complete. The remaining sequences in the protocol still allowed conventional image analysis without degradation, and the infant went on to receive bilateral cochlear implants. The authors suggest that future work could tune FGATIR parameters to known T1 values for younger children—specific sequences calibrated for 6, 12, and 18 months of age—which would be an important advance because auditory brainstem implantation before age 2 may be ideal.

The technical advantages of FGATIR over diffusion-based imaging are substantial. Diffusion tensor imaging and tractography have previously revealed myelination changes in patients with sensorineural hearing loss, but the approach suffers from low spatial resolution—the lateral lemniscus is less than 1 millimeter thick, while state-of-the-art 3T diffusion MRI resolves only 2 millimeters—significant spatial misregistration and distortion near the skull base, and poor direct contrast between adjacent brainstem structures. There is also a risk of analytical circularity: using diffusion contrast or tractography to segment structures and then reporting diffusion changes associated with hearing loss invites confirmation bias. FGATIR, which more closely resembles the postmortem histology contrast of brainstem myelination, could provide complementary contrast or even an independent reference standard for extracting diffusion data, while its gradient-based three-dimensional volumetric acquisition offers the high resolution, accurate co-registration, and absence of skull base artifacts that make it suitable for surgical navigation.

Animal studies have long shown that sensorineural hearing loss reduces neuron soma size in the cochlear nucleus, the medial nucleus of the trapezoid body, the superior olivary complex, the lateral lemniscus, and the inferior colliculus. Prior diffusion imaging work in humans reported lower diffusion scalar values in several cortical and subcortical auditory structures when hearing loss was present, and positive correlations have been reported between cochlear implant Categories of Auditory Performance scores and fractional anisotropy values within the auditory radiation, superior olivary nucleus, inferior colliculus, medial geniculate body, and trapezoid body. What FGATIR adds is the possibility of seeing these structural differences directly, without the indirect inference required by diffusion contrast.

The authors are careful to emphasize the study’s limitations. With only four patients enrolled and three technically evaluable, no inference about diagnostic performance, prognostic value, or clinical utility is possible. FGATIR findings did not influence candidacy, implant type, implanted side, surgical planning, counseling, or rehabilitation for any patient in the series. Because the FGATIR acquisition and convolutional neural network postprocessing were used together in all cases, the contribution of denoising has not been independently assessed. The qualitative consensus readings lacked predefined quantitative criteria and a formal interobserver reliability metric, and some observed asymmetries could fall within normal developmental variation. Whether choosing the less affected side for implantation would yield better outcomes remains an open question, and because central auditory projections are extensively bilateral and predominantly crossed, an asymmetry in an isolated brainstem nucleus cannot be interpreted as evidence favoring one ear over the other.

Still, the demonstration that FGATIR can resolve upper auditory brainstem pathway anatomy in implant candidates older than approximately 2 years marks a genuine technical milestone. The sequence is currently investigational in this setting, pending prospective study after the development of age-matched pediatric reference data to enable quantitative assessment. If future research confirms that the structural differences observed with FGATIR carry diagnostic or prognostic significance, clinicians may one day peer directly into the brainstem of a deaf child and see, for the first time, whether the neural highway that an implant must engage is intact—turning outcome variability from a mystery into something measurable.

Subject of Research: Neuroscience

Article Title: Direct visualization of auditory brainstem pathways in auditory implant candidates

Article References: Cottrell, J., Breen, M., Santini, V., Stepanov, V., Friedmann, D. R., Jethanamest, D., McMenomey, S., Roland, J. T., Jr., & Shepherd, T. M. (2026). Direct visualization of auditory brainstem pathways in auditory implant candidates. Frontiers in Audiology and Otology, 4, Article 1912340. https://doi.org/10.3389/fauot.2026.1912340

Image Credits: AI Generated

DOI: 10.3389/fauot.2026.1912340

Keywords: FGATIR, MRI, cochlear implant, auditory brainstem implant, hearing loss, brainstem, inferior colliculus, neuroradiology, pediatric audiology, medical imaging, Direct, visualization