Inertial Measurement Unit-Based Romberg Test for Assessing Adults With Vestibular Hypofunction

Kuan-Chung Ting, Yu-Chieh Lin, Chia-Tai Chan, Tzong‐Yang Tu, Chun-Che Shih, Kai-Chun Liu, Yu TsaoIEEE Journal of Translational Engineering in Health and Medicine 2023 · DOI

TL;DR

This study demonstrates that wearable Inertial Measurement Units (IMUs) can transform the traditional, subjective Romberg balance test into a precise, objective diagnostic tool. By quantifying kinematic metrics like head and pelvic acceleration, researchers were able to clearly differentiate adults with bilateral vestibular hypofunction from healthy individuals, providing a scalable path for clinical screening and rehabilitation tracking.

What Problem Does This Paper Address?

Vestibular hypofunction (VH)—a deficit in the inner ear’s balance system—can lead to debilitating issues with vision, gait, and stability. For decades, clinicians have relied on the Romberg test to assess this. The test is deceptively simple: stand still with your feet together, first with eyes open, then with eyes closed. If you sway excessively or fall, the test is "positive."

However, the traditional Romberg test has a major flaw: it is subjective. It relies on a clinician’s visual observation, which is prone to observer bias and lacks the resolution to detect subtle kinematic differences. Furthermore, standard clinical tools like force plates are expensive, bulky, and often unavailable in primary care settings. There is a pressing need for a portable, low-cost, and objective way to quantify balance deficits to improve diagnosis and monitor patient recovery.

Methods

The researchers conducted a cross-sectional study involving 62 participants (31 with VH and 31 healthy controls). They "instrumented" the Romberg and tandem Romberg (heel-to-toe) tests using APDM Opal wearable sensors.

  • Sensor Placement: Four IMUs were strapped to the head (occipital bone), pelvis (L4-L5 level), and both ankles.
  • Data Capture: The sensors (triaxial accelerometers, gyroscopes, and magnetometers) recorded data at 128 Hz.
  • Kinematic Metrics: The team focused on four primary metrics: Maximum Acceleration (ACC), Mean ACC, Root Mean Square (RMS) of ACC, and Mean Sway Velocity (MV).
  • Axes of Movement: Data was analyzed across three anatomical axes: medio-lateral (ML - side-to-side), cranio-caudal (CC - up-and-down), and antero-posterior (AP - front-to-back).
  • Signal Processing: To ensure clean data, they applied a moving average filter to denoise the signal and removed baseline offsets caused by sensor tilting or individual posture differences.

Datasets

The study utilized a primary clinical dataset collected at an academic tertiary medical center between 2021 and 2022.

  • VH Group (n=31): Included 20 individuals with Bilateral Vestibular Hypofunction (BVH) and 11 with Unilateral Vestibular Hypofunction (UVH).
  • Healthy Control Group (n=31): Age- and sex-matched individuals with no history of dizziness or neurological disorders.
  • Age Range: 20 to 49 years (selected to avoid age-related balance declines that might confound the data).

Key Insights and Ideas

  • The ML Axis is the "Smoking Gun": The most significant differences between healthy and vestibular-impaired participants occurred in the medio-lateral (side-to-side) axis. While we often think of balance as "not falling over," the specific direction of the sway is a key biomarker for vestibular health.
  • Head vs. Ankle: Interestingly, head and pelvic movements were much more sensitive indicators of VH than ankle movements. This suggests that the body’s "top-down" stabilization strategy is more heavily impacted by vestibular loss.
  • The Compensation Effect: Participants with unilateral (one-sided) loss were significantly harder to distinguish from healthy controls than those with bilateral (two-sided) loss. This highlights the brain's remarkable ability to compensate for a single failing sensor, but also underscores the difficulty in screening for UVH using static balance tests alone.
  • Sensory Integration: The study confirms that balance is a tripod of vision, proprioception, and vestibular input. When vision is removed (eyes closed), people with VH rely solely on proprioception. The IMUs captured the exact moment this "tripod" became unstable.

Key Results

  • Eyes-Closed Romberg: VH participants showed significantly higher Maximum ACC and RMS of ACC in head movement along the ML axis compared to controls (p < 0.05).
  • Bilateral vs. Healthy: Individuals with BVH exhibited markedly higher sway metrics in both the head and pelvis. For example, the head's Max ACC in the ML axis was a pivotal indicator for identifying BVH.
  • Tandem Romberg: In the eyes-open tandem condition (a more "stressful" posture), BVH participants showed significantly increased head acceleration in the ML axis.
  • Effect Sizes: The study reported strong effect sizes (Cohen’s d ranging from 0.54 to 0.89), suggesting that these IMU metrics are robust differentiators.

Limitations and Future Work

The authors acknowledge several limitations:

  • Sample Size: The subgroups (especially UVH) were relatively small, which may have limited the statistical power to detect subtle differences in one-sided vestibular loss.
  • Age Cutoff: By excluding participants over 50, the study doesn't yet account for how age-related sensory decline interacts with VH.
  • Specificity: Balance tests are indirect. Issues like foot deformities or neuropathy could potentially mimic VH results, requiring careful clinical correlation.

Future Directions: The team plans to incorporate angular velocity (from gyroscopes) into their analysis and explore more dynamic tests like the "Timed Up and Go" (TUG) to create a more comprehensive ambulatory profile for patients.

Summary updated

Abstract

This work aims to explore the utility of wearable inertial measurement units (IMUs) for quantifying movement in Romberg tests and investigate the extent of movement in adults with vestibular hypofunction (VH). A cross-sectional study was conducted at an academic tertiary medical center between March 2021 and April 2022. Adults diagnosed with unilateral vestibular hypofunction (UVH) or bilateral vestibular hypofunction (BVH) were enrolled in the VH group. Healthy controls (HCs) were recruited from community or outpatient clinics. The IMU-based instrumented Romberg and tandem Romberg tests on the floor were applied to both groups. The primary outcomes were kinematic body metrics (maximum acceleration [ACC], mean ACC, root mean square [RMS] of ACC, and mean sway velocity [MV]) along the medio-lateral (ML), cranio-caudal (CC), and antero-posterior (AP) axes. A total of 31 VH participants (mean age, 33.48 [SD 7.68] years; 19 [61%] female) and 31 HCs (mean age, 30.65 [SD 5.89] years; 18 [58%] female) were recruited. During the eyes-closed portion of the Romberg test, VH participants demonstrated significantly higher maximum ACC and increased RMS of ACC in head movement, as well as higher maximum ACC in pelvic movement along the ML axis. In the same test condition, individuals with BVH exhibited notably higher maximum ACC and RMS of ACC along the ML axis in head and pelvic movements compared with HCs. Additionally, BVH participants exhibited markedly increased maximum ACC along the ML axis in head movement during the eyes-open portion of the tandem Romberg test. Conversely, no significant differences were found between UVH participants and HCs in the assessed parameters. The instrumented Romberg and tandem Romberg tests characterized the kinematic differences in head, pelvis, and ankle movement between VH and healthy adults. The findings suggest that these kinematic body metrics can be useful for screening BVH and can provide goals for vestibular rehabilitation.
BibTeX
@article{ting2023inertial,
  author = {Kuan-Chung Ting and Yu-Chieh Lin and Chia-Tai Chan and Tzong‐Yang Tu and Chun-Che Shih and Kai-Chun Liu and Yu Tsao},
  title = {{Inertial Measurement Unit-Based Romberg Test for Assessing Adults With Vestibular Hypofunction}},
  journal = {IEEE Journal of Translational Engineering in Health and Medicine},
  year = {2023},
  doi = {10.1109/jtehm.2023.3334238},
}
Figure