Nikhil Prasad Fact checked by:Thailand Medical News Team Aug 18, 2026 1 hour, 13 minutes ago
Researchers have developed an unusual quantum-assisted method that may help identify subtle balance-control abnormalities in some people after COVID-19, including changes that could be difficult to detect during ordinary clinical examinations.
Quantum-assisted analysis identified a small subgroup of post-COVID participants showing distinctive abnormal-like
patterns in balance and postural control
The study analyzed 140 participants, comprising 114 people with documented previous SARS-CoV-2 infection and 26 controls. Importantly, the researchers stressed that their method is exploratory and cannot diagnose Long COVID.
Researchers And Institutions
The research was conducted by Alisa Kunapinun, Andrzej Nowak and Adar Pelah. The researchers are from the Harbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, Florida; Department of Psychology and Center for Complex Systems, Florida Atlantic University, Boca Raton, Florida; Department of Engineering and Computer Science, Florida Atlantic University, Boca Raton, Florida; and School of Physics, Engineering & Technology, University of York, York, United Kingdom.
Putting Balance Under Pressure
Participants performed seven established balance tasks, including Romberg, Tandem Romberg, CTSIB and Fukuda tests. Several involved standing with eyes open or closed. The tasks were then repeated while participants experienced an auditory Stroop task, adding mental distraction while they attempted to maintain balance.
A marker less depth-camera motion-capture system recorded body movements. Researchers examined 17 joint-angle features and converted movement signals into frequency patterns between 0 and 5 Hz.
They created 13 comparisons designed to reveal what happened when participants were mentally distracted, when visual information was removed, or when testing conditions changed.
Quantum-Assisted Method Finds a Smaller Abnormal Group
The most striking finding involved a mathematical approach called quadratic unconstrained binary optimization, or QUBO, solved through a quantum-annealing workflow.
Across all tests, QUBO classified only 1 of 307 control observations, or 0.3 percent, as abnormal-like. Among post-COVID observations, 118 of 1,366, or 8.6 percent, were classified as abnormal-like.
A conventional Gaussian mixture model produced a much broader result, labeling 63 of 307 control observations, or 20.5 percent, and 266 of 1,366 post-COVID observations, or 19.5 percent, as abnormal-like.
At participant level, 25 of 26 controls had no abnormal-like QUBO assignments. None were categorized as probable or strongly abnormal. The researchers argue that this selectivity suggests QUBO was better aligned with their goal of preserving the control group while finding unusual post-COVID patterns.
As highlighted in this Thailand
Medical News report, this does not mean only 8.6 percent of people recovering from COVID have balance proble
ms. These percentages represent individual test observations produced by an experimental clustering system, not disease prevalence.
What The Abnormal Patterns Looked Like
The researchers also examined four movement-frequency bands. The clearest differences appeared mainly in the first two bands, representing low-frequency postural sway and normal balance-control activity.
During several Stroop comparisons, QUBO-identified abnormal-like participants displayed stronger changes in these frequencies than the overall post-COVID group. This suggests that averaging everybody with previous COVID together could conceal smaller groups with distinctive neuromotor responses.
One participant with a physician-documented Long COVID evaluation was classified as probably abnormal by QUBO, with abnormal-like findings in four of ten available comparisons. However, this provides only limited supporting evidence because clinical information was incomplete.
Conclusions
The findings suggest that quantum-assisted QUBO analysis could become a useful research tool for identifying distinct post-COVID balance phenotypes that conventional group averages may overlook. Larger, independently validated studies with detailed clinical assessments are essential before this approach can be considered diagnostic or used for patient care.
The study findings were published as a non-peer-reviewed preprint on Preprints.org
https://www.preprints.org/manuscript/202608.0215
Read Also:
https://www.thailandmedical.news/articles/long-covid
Share
Tweet
Share
