MRI scanners produce noise loud enough to cause temporary hearing threshold shifts even when patients wear hearing protection, and in rare cases, permanent hearing damage including tinnitus has been documented. The acoustic output of a standard 3 T scanner routinely exceeds 95 decibels, which puts it in the same range as a chainsaw or a rock concert, and stronger 7 T research scanners can top 120 dB. That intensity, sustained for the 30 to 60 minutes a typical scan takes, is a genuine risk to hearing if protective measures fall short.
Why MRI Machines Are So Loud
The noise does not come from the powerful main magnet itself, which operates silently. It comes from the gradient coils, a set of smaller electromagnets inside the scanner that switch on and off rapidly to build the image. When electrical current pulses through these coils inside a strong magnetic field, the interaction produces what physicists call Lorentz forces. Those forces push and pull the coil structure thousands of times per second, making the hardware vibrate like a loudspeaker diaphragm and launching pressure waves into the air around the patient’s head.1PubMed Central. Overview of Methods for Noise and Heat Reduction in MRI Gradient Coils The vibrations also travel through the solid components of the scanner bore, including metallic parts that develop their own eddy currents and contribute additional noise sources.2Engineering Science and Technology, an International Journal. A systematic review of acoustic noise sources in Magnetic Resonance Imaging
The sound is not steady like a fan or a hum. Different imaging sequences switch the gradients at different rates and amplitudes, producing a percussive, rapidly changing pattern of knocking, buzzing, and beeping that can shift in pitch and intensity several times during a single scan. This variability matters because the ear responds differently to impulsive noise than to a constant drone; sharp, repeated impacts are generally more damaging at the same average decibel level.
How Loud the Numbers Actually Get
The noise depends heavily on the scanner’s field strength and the type of imaging sequence being run. A study measuring acoustic output across clinical scans at 3 T and 7 T found that the average sound level exceeded 95 dB for the majority of scans at both field strengths. At 3 T, the loudest sequence (a T1-weighted spin echo) peaked at about 105 dB. At 7 T, the loudest sequence reached roughly 122 dB, and the average across all sequences was about 106 dB compared to about 91 dB at 3 T.3PubMed Central. Acoustic Noise Levels in High‐field Magnetic Resonance Imaging Scanners
To put those numbers in perspective, occupational safety agencies typically set 85 dB as the threshold above which prolonged exposure starts to pose a hearing risk, and most recommend limiting unprotected exposure at 100 dB to about 15 minutes. A clinical MRI scan lasts far longer than that, and the loudest sequences blast past the 100 dB mark with room to spare. The trend in diagnostic imaging is toward stronger magnets, because higher field strengths produce sharper images, but the noise problem scales upward with them.2Engineering Science and Technology, an International Journal. A systematic review of acoustic noise sources in Magnetic Resonance Imaging
Temporary Hearing Shifts After a Scan
The most commonly documented auditory effect of MRI is a temporary threshold shift, meaning your hearing gets slightly worse right after the scan and then recovers. A study of patients undergoing 1.5 T head and neck MRI found that hearing thresholds shifted significantly at several frequencies immediately after the scan, particularly around 4 and 6 kHz. But when the same patients were tested again 24 hours later, no significant difference from their pre-scan hearing remained.4PubMed Central. Hearing Thresholds Changes after MRI 1.5T of Head and Neck
A separate study at 3 T went further by measuring brainstem responses, which are an objective way of testing hearing that does not depend on the patient’s self-report. Healthy volunteers who wore hearing protection during a roughly 51-minute neuroimaging session still showed a mean threshold shift of about 5 dB immediately after, despite the earplugs. By day 25, their hearing had returned to baseline.5PubMed. Temporary Hearing Threshold Shift in Healthy Volunteers with Hearing Protection Caused by Acoustic Noise Exposure during 3-T Multisequence MR Neuroimaging That 5 dB shift is subtle enough that most people would not notice it in daily life, and the researchers noted it was well below the threshold associated with nerve damage. Still, the fact that hearing protection did not fully prevent the shift is worth knowing.
The key question is what happens if you have multiple scans over a period of weeks or months, as some patients with chronic conditions do. Each temporary shift recovers, but whether repeated insults accumulate over time is harder to study. The ear’s recovery mechanisms are robust but not unlimited, and the research on cumulative MRI-related hearing exposure is thin. If you are someone who needs frequent MRI monitoring, this is a reasonable concern to raise with your care team.
Permanent Damage and Tinnitus
Reports of lasting hearing loss after MRI are rare but they exist. A case report described an otherwise healthy patient who developed bilateral sensorineural hearing loss accompanied by tinnitus after a brain MRI, with no improvement after three months of follow-up.6PubMed Central. Sensorineural hearing loss after magnetic resonance imaging A survey of MRI-related acoustic trauma found four cases of persistent hearing disturbance, with the authors stressing that permanent hearing problems are a real, if uncommon, consequence of MRI examinations.7Magnetic Resonance in Medical Sciences. A Survey Analysis of Acoustic Trauma Related to MR Scans
These numbers are small, and tens of millions of MRI scans are performed worldwide every year without incident. Permanent hearing loss from a single scan is an outlier event, not a common complication. But the cases that do occur tend to involve factors that reduce the effectiveness of hearing protection: the patient removing or repositioning their ear protection during the scan, protection that was improperly fitted to begin with, or unusually long scan protocols. In some documented cases, the patient was not given hearing protection at all, particularly in older reports before awareness of MRI acoustic risk became widespread.
Tinnitus specifically, the perception of ringing or buzzing without an external sound, shows up in several of these case reports alongside the hearing loss. It can also appear on its own. People who already have tinnitus sometimes report that it worsens temporarily after a scan, which may reflect the temporary threshold shift or may relate to the stress and anxiety of lying inside a noisy, confined space for an extended period. Stress and emotional states have been shown to share a similar probability of triggering tinnitus as occupational noise exposure, and chronic activation of the body’s stress response appears to be involved in tinnitus development and worsening.8PubMed Central. The association between stress, emotional states, and tinnitus: a mini-review Disentangling the acoustic damage from the stress-related flare-up is difficult in any individual case, but both pathways are real.
How Well Hearing Protection Actually Works
Every MRI facility is supposed to provide hearing protection. The standard options are foam earplugs, over-ear muffs, or both used together. How much protection these deliver depends heavily on fit and on which frequency range you are measuring. One detailed study found that in the frequency range where MRI noise is most intense (around 1 to 1.4 kHz), earmuffs alone attenuated perceived sound by about 30 to 37 dB, earplugs alone by about 25 to 28 dB, and the two combined by about 39 to 41 dB.9PubMed Central. Isolating the auditory system from acoustic noise during functional magnetic resonance imaging: examination of noise conduction through the ear canal, head, and body
Those numbers sound substantial, but consider them against the measured noise levels. If a 3 T scan peaks at 105 dB and your dual ear protection knocks off 40 dB, you are hearing about 65 dB, equivalent to normal conversation. That is fine. But if a 7 T scanner peaks at 122 dB and your earplugs only reduce 28 dB, you are still exposed to roughly 94 dB, which is well above safe limits for a sustained exposure. Even the combined earmuff-plus-earplug approach would leave you at about 81 dB, which is safer but still not trivial for an hour-long scan.
The gap between laboratory-rated protection and real-world performance makes things worse. Hearing protection ratings are measured under controlled conditions with a careful fit. In actual clinical settings, patients fidget, earplugs loosen, and muffs shift. A review of MRI acoustic noise and hearing protection performance concluded that when government derating recommendations are applied, which estimate what protection levels are actually achieved during practical use, it becomes unclear whether even the best passive hearing protection available is sufficient for the loudest equipment.10PubMed. A Review of MRI Acoustic Noise Outputs and Hearing Protection Device Performance That is a sobering finding, especially for high-field scanners.
Bone conduction adds another complication. Sound does not enter your inner ear only through the ear canal. It also travels through the bones of your skull, bypassing earplugs and earmuffs entirely. Researchers have measured this bone-conducted pathway and found that conventional hearing protection provides over 20 dB of insertion loss, with the strongest effects at certain frequencies, but the bone pathway remains open and contributes meaningfully to what the cochlea actually receives.11PubMed. Importance of bone-conducted sound transmission on patient hearing in the MR scanner Experimental “helmet” devices that encase the entire head in sound-dampening material have achieved dramatically better results, attenuating sound by 55 to 63 dB in the critical frequency range, but these are not standard clinical equipment.9PubMed Central. Isolating the auditory system from acoustic noise during functional magnetic resonance imaging: examination of noise conduction through the ear canal, head, and body
Software and Hardware Approaches to Quieter Scans
Scanner manufacturers have begun addressing the noise problem at its source. One approach, marketed under names like “Silent Scan,” modifies the gradient waveforms so the coils switch more gently, producing far less vibration. Testing of this technology showed it reduced mean noise from about 105 dB down to roughly 69 dB, which was indistinguishable from the background noise level in the scanner room.12PubMed Central. Acoustic noise reduction in MRI using Silent Scan: an initial experience That is an enormous reduction, on the order of a hundredfold decrease in sound pressure, though these quiet sequences are currently limited to certain image types and are not yet available for every scan a patient might need.
A similar technology called ComforTone works by reshaping the gradient waveforms to suppress the high-frequency noise components that human hearing is most sensitive to. Subjects in testing reported significantly less discomfort, and measured sound pressure levels dropped in the frequency ranges that matter most to the ear.13PubMed. Evaluation of magnetic resonance imaging acoustic noise reduction technology by magnetic gradient waveform control Radiographers can also achieve meaningful noise reduction simply by adjusting scan parameters, such as lowering the gradient slew rate or changing sequence timing, without any special hardware. Studies have demonstrated significant sound pressure level reductions across several standard sequence types using these manual adjustments.14Radiography. Reduction of MRI acoustic noise achieved by manipulation of scan parameters
Active noise cancellation, the same technology used in consumer headphones, has also been adapted for MRI. A microphone placed near the patient’s ear picks up the scanner noise and generates an opposing sound wave through the headphone speaker. Testing inside an MRI scanner showed that these systems could effectively reduce the perceived scanner noise, though the rapidly changing and broadband nature of MRI sound makes it a harder target than the steady low-frequency drone that consumer noise-cancelling headphones handle best.15Applied Acoustics. Developments in active noise control sound systems for magnetic resonance imaging
Protecting Newborns and Small Children
Neonates present a special challenge. Their cochleas are fully formed and vulnerable, but fitting standard earplugs into tiny ear canals is difficult, and babies cannot report discomfort or tell you their ears hurt. MRI systems installed directly in neonatal intensive care units have been designed with lower acoustic output in mind, and measured noise levels with appropriate hearing protection in place fell near or below the 65 dB target that neonatal guidelines recommend. In a pilot study using one such system, all 15 infants completed their scans without sedation and with minimal motion problems.16PubMed Central. Characterization of acoustic noise in a neonatal intensive care unit MRI system
Specialized hearing protectors designed specifically for neonatal MRI have also been developed and tested. Because they use entirely nonmetallic, nonconducting materials, they do not interact with the magnetic field and have been designated “MR Safe,” meaning they pose no known hazards in any MRI environment.17Journal of Radiology Nursing. Magnetic Resonance Imaging Evaluation of a Hearing Protector Device Designed for Neonatal Patients For older children who can cooperate, the same dual-protection strategy used for adults, earplugs plus muffs, is the standard recommendation, along with using the quietest imaging sequences that will still answer the clinical question.
Middle Ear Implants and MRI Compatibility
People with middle ear implants have a distinct set of concerns about MRI. The worry is not just about acoustic noise but about whether the magnetic field could physically move a metallic prosthesis implanted near the delicate structures of the inner ear. If a prosthesis shifts even slightly, it could damage the cochlea directly, causing immediate and possibly permanent hearing loss.
The reassuring news is that titanium, the material used in most modern middle ear prostheses, is effectively nonmagnetic. Safety testing of 24 titanium prostheses at 3 T showed that 21 exhibited no movement at all. The three that showed minimal deflection during an initial screening turned out to have essentially zero measurable force when tested more rigorously, and none of the tested prostheses showed any heating.18PubMed. Safety evaluation of titanium middle ear prostheses at 3.0 tesla Middle ear ossicular prostheses, including stapes prostheses, are generally classified as MR Conditional for 1.5 T scanners, and many are approved for 3 T as well.19PubMed. MRI information for commonly used otologic implants: review and update
The one significant exception involves a specific batch of McGee stapes pistons manufactured in 1987 that were accidentally made from a magnetic alloy instead of the intended nonmagnetic material. Patients with those particular lot numbers should not undergo MRI and should be directed to CT scanning instead, or have the prosthesis replaced if MRI is essential.20PubMed. MRI scanners and the stapes prosthesis This is an edge case that surgeons and radiologists are aware of, but it is worth knowing if you have a stapes prosthesis and are not sure when it was implanted. Your surgical records will specify the manufacturer and model.
What You Can Do Before Your Next Scan
If you are scheduled for an MRI and worried about your hearing, the single most effective thing you can do is make sure your hearing protection fits well and stays in place for the entire scan. Ask for both earplugs and over-ear muffs if they are available. Do not hesitate to tell the technologist if your protection shifts or falls out; every scanner has an intercom and a squeeze bulb you can use to pause the procedure. The few minutes it takes to refit earplugs is trivial compared to the scan time, and no technologist will object.
Ask whether quieter imaging sequences are available for your scan. Not every clinical question requires the loudest protocols, and many facilities now have access to noise-reduced sequence options. If you need frequent scans, mention that history so the team can plan accordingly, perhaps choosing lower-noise sequences where image quality allows. People who already have hearing loss, tinnitus, or heightened noise sensitivity should make that known upfront so extra precautions can be taken.
The overall picture is that MRI acoustic noise is a real occupational and patient safety concern, not a theoretical one. For the vast majority of people getting a scan with proper hearing protection in place, the risk of lasting damage is very small. Temporary shifts happen, recover within a day, and are clinically insignificant. Permanent injury is documented but rare, and almost always traceable to inadequate protection or unusually high-output equipment. The gap between the best available protection and the loudest scanners is narrowing as quiet-scan technologies mature, but it has not closed yet, and at the highest field strengths, current passive hearing protection alone may not be enough to fully shield the inner ear.