The research team, led by Susanna Gagliardi, Nicole Stark, Michael Madigan, and Steve Rowson, tackled a deceptively simple question with a sophisticated multi-step analytical framework: when a worker falls from a given height, how fast is the head really moving when it strikes the ground? The intuitive answer, drawn from basic physics, assumes a rigid body in free fall, striking the ground head-first with no interference. But real human beings brace with their arms, rotate their torsos, strike objects on the way down, and land on body parts other than their heads. The researchers call these behaviors fall impact mitigating mechanisms, or FIMMs, and their study is the first to attach hard numbers to how much these mechanisms collectively reduce head impact velocity in falls from industrial heights.

The methodology stitched together three very different sources of evidence. First, the team performed laboratory drop tests using a medium NOCSAE headform mounted on a twin-wire test system, striking it bare and while wearing a 3M SecureFit safety hard hat at the front, side, and rear locations. Drop heights ranged from 12 to 72 inches, producing measured head impact velocities between 2.25 and 6.00 meters per second. Each impact was instrumented with a three-degree-of-freedom accelerometer package sampled at 20 kilohertz, filtered according to SAE J211 standards, and used to calculate two classic injury metrics: peak linear acceleration (PLA) and the head injury criterion (HIC), computed over a 15-millisecond window. The bare head data followed a clean linear relationship with velocity, yielding R-squared values of 0.95 for HIC and 0.87 for PLA, while the helmeted data required an exponential fit because the helmet’s protective foam saturates at higher impact energies.

Next came the bridge from laboratory measurements to injury probability. The team generated skull fracture risk curves from a previously published dataset of post-mortem human surrogate forehead impacts published by Mertz and colleagues, applying Nusholtz’s Consistent Threshold method, a non-parametric estimator that handles doubly censored data without assuming an underlying distribution. The resulting risk point estimates were then fitted with log-normal cumulative distribution functions to produce smooth, continuous risk curves. Notably, the new curves outperformed the previously published Mertz risk curves in predicting skull fracture incidence within the source dataset, a meaningful improvement given that the older Mertz/Weber method has been criticized for its sensitivity to threshold values and its inability to accommodate censored observations.

The third and perhaps most novel ingredient was real-world accident data. The researchers mined the Occupational Safety and Health Administration’s Fatality and Catastrophe Investigation Summaries, searching for falls involving concussion, brain injury, or skull injury in construction-related industries between January 2009 and September 2024. They analyzed 591 accident reports, of which 260 documented skull fractures, 94 documented falls without skull fracture, and 237 contained unspecified head injuries and were excluded. Among the reports that recorded impact location, the back of the head was struck in 55.9 percent of cases, the side in 32.4 percent, and the front in only 11.8 percent. Falls resulting in skull fracture averaged 16.6 feet in height, compared with 12.7 feet for falls that did not fracture the skull, though the substantial overlap between the two groups underscored that fall height alone does not determine injury outcome.

With skull fracture risk curves built from both laboratory metrics and real-world fall heights, the team could perform what they call risk matching. For any laboratory head impact velocity, the drop tests yielded a PLA or HIC value; the PMHS-derived risk curves translated that value into a skull fracture probability; and the OSHA-derived fall height risk curve identified the fall height associated with that same probability. Because the OSHA data inherently include whatever bracing, rotation, or object contact occurred during real falls, the resulting velocity-to-height relationship already embeds the effects of FIMMs. The final step computed a theoretical worst-case velocity for each fall height using conservation of energy for a falling particle, adding 1.75 meters to account for the height of a 50th percentile male, since the head of a standing worker starts well above their feet. The difference between this worst-case velocity and the laboratory-estimated velocity quantified the aggregate FIMM effect.

The headline result is striking: fall impact mitigating mechanisms appear to reduce head impact velocity by 44.5 to 64.4 percent compared with a direct, head-first free fall from the same height. For bare head impacts, the average reduction was 55.3 percent when HIC was the injury predictor and 64.4 percent when PLA was used. For helmeted impacts, the reductions were somewhat smaller, averaging 44.5 percent for HIC and 51.5 percent for PLA, a difference the authors attribute to the helmet’s energy absorption shifting the velocity at which a given injury metric is reached. The effect also grew with fall height in the PLA analysis, and the gap between bare head and helmeted estimates narrowed from more than 22 percent at low heights to under 11 percent at the tallest falls of roughly 6 meters, as the helmet approached its protective limit and injury metrics climbed exponentially.

The findings align reasonably well with the only human subjects study to capture fall kinematics all the way to ground impact. That work, by Ferro and colleagues, subjected volunteers to laboratory-induced ladder falls and recorded vertical head impact velocities between 0.42 and 3.88 meters per second from heights of 0.792 to 1.798 meters. Recalculating FIMM effects from those data produced an average reduction of 73 percent, somewhat higher than the Virginia Tech estimates of 59.6 percent for bare head conditions at comparable heights. The discrepancy is plausibly explained by the fact that Ferro’s participants knew a fall was coming, allowing faster bracing reactions, and landed on a padded surface that absorbed energy independently of any body movement. The comparison suggests the new population-level estimates are conservative and that individual variability in fall response is substantial, which is why the authors recommend treating FIMM effects as a range rather than a single number.

The practical implication is direct and potentially transformative for helmet standards. If laboratory tests are meant to replicate the impact conditions of real occupational falls, the head impact velocities used in those tests should be reduced by roughly 44.5 to 64.4 percent from the worst-case free-fall value. Testing helmets at unrealistic, worst-case velocities risks both over-designing helmets for scenarios that rarely occur and, more importantly, mischaracterizing performance across the range of impacts workers actually experience. The team even provides linear regression equations linking real-world fall heights to recommended laboratory impact velocities for both bare and helmeted conditions, with R-squared values as high as 0.984, giving manufacturers and researchers a ready-made tool for selecting test conditions.

The authors are careful to frame their results as approximate, population-level estimates rather than precise predictions. OSHA reports contain no video or kinematic reconstructions, so FIMMs could not be directly observed, only inferred as an aggregate effect. The PMHS risk curves derive from frontal impacts, while most real-world falls strike the back of the head, and skull fracture tolerance varies with location. The parametric risk curves underpredicted observed skull fracture counts, and testing used a single helmet model on a single anvil surface. The researchers also caution that a fall carrying low skull fracture risk can still cause concussion through linear and rotational accelerations, and they emphasize that the measured reductions reflect the natural dynamics of falls, not trainable behaviors workers can reliably deploy. Even so, by anchoring laboratory test conditions to the epidemiology of actual workplace falls, this study gives helmet designers, standards bodies, and safety engineers something they have never had before: a defensible, data-driven answer to the question of how hard a falling worker’s head really hits the ground.

Subject of Research: Biomechanical estimation of head impact velocity in occupational falls to inform safety helmet testing standards

Article Title: Relating Real-World Falls to Laboratory Test Conditions

Article References: Gagliardi, S. M., Stark, N. E.-P., Madigan, M. L., & Rowson, S. (2026). Relating Real-World Falls to Laboratory Test Conditions. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04353-w

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04353-w

Keywords: falls, traumatic brain injury, safety helmets, helmet testing, biomechanics, skull fracture risk, OSHA, construction safety, head impact velocity, injury risk curves, HIC, fall protection

The new method, described in the journal Medical & Biological Engineering & Computing by Wanchao Yao, Tianshu Gu, and Fuwang Wang, combines three technologies into a single pipeline called MVMD-FCASNN: multivariate variational mode decomposition, a frequency and channel attention spiking neural network, and a carefully designed signal-processing front end. The system reads multichannel electroencephalogram, or EEG, signals collected noninvasively from the scalp and classifies whether the operator is mentally fatigued in real time. In benchmark tests, the model achieved a classification accuracy of 98.81 percent and, critically, maintained strong performance even when the EEG recordings were deliberately corrupted with noise at a signal-to-noise ratio of minus six decibels, a level of interference far harsher than most laboratory experiments ever attempt.

The problem the researchers set out to solve is fundamentally one of signal integrity. EEG is famously vulnerable to contamination. On a construction site, the electrical systems of large machinery, radio transmissions, vibration transmitted through the operator’s body, and the operator’s own muscle movements and eye blinks all inject artifacts into the faint electrical potentials measured at the scalp. Classical decompositions of EEG signals, such as empirical mode decomposition, have long been used to tease apart the mixed frequency content of brain recordings, but they are notoriously sensitive to noise. When noise corrupts the decomposition process itself, the resulting components lose their physical meaning, and downstream classifiers inherit the error. Previous work, including studies on driver fatigue detection, has struggled with exactly this weakness, which is why many published fatigue-detection systems perform well in quiet laboratories but falter in the field.

The team’s answer to this challenge is multivariate variational mode decomposition, or MVMD. Unlike methods that process each EEG channel independently, MVMD jointly decomposes all channels at once, searching for a set of intrinsic mode functions, or IMFs, that are frequency-aligned across every recording electrode. In practical terms, this means that the alpha-band oscillation detected over the frontal cortex and the alpha-band oscillation detected over the central electrodes are extracted as matching components of the same decomposition, preserving the spatial relationships between brain regions that conventional approaches tend to scramble. Because variational mode decomposition formulates the problem as an optimization with explicit bandwidth constraints, it is also far more robust against noise corruption than empirical mode decomposition, a property that earlier studies in motor-imagery brain-computer interfaces had already hinted at. Applying it to fatigue detection in a genuinely noisy setting is one of the paper’s key contributions.

Once the EEG signals have been decomposed, the question becomes which of the resulting components actually carry the signature of fatigue. Not all frequency bands are equally informative, and the answer can shift from person to person and from moment to moment. To handle this, the researchers designed an IMF attention mechanism that adaptively evaluates the contribution of each extracted frequency component, automatically weighting the bands that best discriminate fatigue from alertness. Neuroscientifically, this makes good sense: mental fatigue is typically accompanied by changes in relative spectral power, such as increases in theta and alpha activity over frontal regions and alterations in beta activity associated with sustained attention. By letting the network learn which IMFs matter, the system avoids the brittleness of hand-picked frequency features.

Alongside frequency attention, the architecture incorporates a channel attention module, inspired in part by squeeze-and-excitation networks developed in computer vision, that emphasizes critical brain regions. The authors report that the model places its strongest weight on frontal and central areas of the scalp, precisely where neurophysiological research locates much of the neural activity associated with cognitive fatigue, vigilance decrements, and the executive control of sustained attention. This spatial selectivity strengthens the model’s representation of the brain’s fatigue signature while suppressing electrodes that contribute mostly noise. The combination means the network effectively learns both where to look on the scalp and which rhythms to listen for, two questions that traditional pipelines answer with fixed, hand-engineered rules.

The final stage of processing is perhaps the most futuristic element: a spiking neural network, or SNN. Unlike conventional artificial neural networks that transmit continuous-valued activations, spiking networks communicate through discrete spikes, mirroring the way biological neurons actually fire. This choice serves two purposes. First, spiking neurons are inherently event-driven, meaning they respond to changes in their input rather than to every sample, which makes them energy efficient and well suited to portable, battery-powered monitoring hardware that an operator could realistically wear on a job site. Second, spiking dynamics preserve the temporal structure of the EEG signal, retaining information about how brain activity evolves over milliseconds, information that frame-based networks tend to discard. The researchers point to earlier work on low-power driving fatigue monitoring with spiking networks as a foundation for this design philosophy.

The experimental results are the reason this study is attracting attention. On mental fatigue classification tasks, MVMD-FCASNN reached 98.81 percent accuracy, a figure the authors describe as significantly outperforming traditional methods. More striking is the model’s resilience: at a noise level of minus six decibels, where the interference is actually stronger than the underlying EEG signal, the system continued to classify fatigue reliably, while conventional comparison methods degraded sharply. For a technology intended to run continuously in environments where electrical noise from cranes, welding equipment, and generators is unavoidable, this anti-interference capability may matter more than headline accuracy. The study also reports informed consent from all participants and ethics approval consistent with the Helsinki declaration, and the authors declare no conflicts of interest.

The broader context explains why this line of research is urgent. Tower crane accidents remain a persistent source of fatalities and injuries in the construction industry, and studies in Hong Kong, Indonesia, and elsewhere have tied a substantial share of unsafe hoisting behavior to operator fatigue and excessive mental workload. Earlier detection systems have relied on eye-tracking glasses, facial landmark analysis, smart cushions, heart rate variability, and other physiological proxies, each with tradeoffs in comfort, invasiveness, and robustness. EEG offers a direct window into cognitive state, but until now its practical deployment in high-noise industrial settings has been hampered by exactly the interference problem this new architecture is built to withstand. Related work by overlapping research groups, including flexible dry electrodes designed for real driving environments, suggests a growing ecosystem of wearable brain-monitoring hardware maturing alongside the algorithms.

The researchers, funded by the National Natural Science Foundation of China and supported by Northeast Electric Power University and the Jilin City Science and Technology Bureau, argue that their approach offers a reliable and efficient solution for real-time mental fatigue monitoring in high-noise construction scenarios. If such systems can be integrated into crane cabins as a routine safeguard, an operator whose brainwaves betray dangerous drowsiness could be alerted, or their work schedule could be flagged for review, before a lapse in attention translates into a dropped load. The work also points beyond construction: the same pipeline of joint multichannel decomposition, dual attention mechanisms, and spiking computation could plausibly be adapted to truck drivers, pilots, air traffic controllers, and any profession where sustained vigilance is a matter of life and death. For now, the result stands as a striking demonstration that brain-inspired computing can withstand the messiness of the real world, not just the laboratory, and that the human brain’s own warning signs of exhaustion can be read clearly enough to act on them in time.

Subject of Research: EEG-based detection of mental fatigue in tower crane operators using multivariate variational mode decomposition and a frequency and channel attention spiking neural network in real construction environments

Article Title: Study on mental fatigue detection of crane operators based on frequency and channel attention spiking neural network with multivariate variational mode decomposition in real construction environments

Article References: Yao, W., Gu, T., & Wang, F. (2026). Study on mental fatigue detection of crane operators based on frequency and channel attention spiking neural network with multivariate variational mode decomposition in real construction environments. Medical & Biological Engineering & Computing. https://doi.org/10.1007/s11517-026-03669-y

Image Credits: AI Generated

DOI: 10.1007/s11517-026-03669-y

Keywords: mental fatigue, crane operators, EEG, spiking neural network, multivariate variational mode decomposition, attention mechanism, construction safety, brain-computer interface, signal processing, neural decoding, occupational health, machine learning