Sage Journals: Discover world-class research

Abstract

The understanding of association football head collision mechanics is of interest in the context of increased neurodegenerative disease risk in ex-professional players. This study sought to measure and characterise micromechanical pressure wave propagation from football head collisions; a distinct energy transfer mechanism from, and occurring before the start of, macromechanical gross head kinematics. A surrogate head setup, instrumented with a hydrophone pressure sensor, was exposed to footballs travelling at match realistic velocities. Pressure waves emanating from the region of first contact were measured within the cranial cavity. Their mean peak-to-peak magnitudes were up to 31.0 kPa. Across a range of 20 footballs of different materials, constructions and assembly methods, all of which satisfied the Laws of the Game, a 9.1-fold difference in peak-to-peak pressure and 54.7-fold difference in energy transfer, was observed. A 5-fold increase in peak-to-peak pressure was observed for a 77% increase in ball inbound velocity. Hydrophilic leather ball types demonstrated up to 4.2-fold greater peak-to-peak pressures when tested in wet, compared to dry conditions. These data highlight that pressure wave energy transfer may be influenced by the materials and construction of the ball, with scope to affect this without requiring alteration to the game or its Laws. The influence of these pressure waves on acute and long-term brain health requires further exploration.

Keywords

soccer ball-type pressure wave head impacts collision mechanics

Introduction

Association football is the most popular sport globally, played by over 260 million players¹ and followed by an estimated 3.5 billion fans.² Whilst playing football offers significant cognitive and physiological health benefits, recent epidemiological studies have highlighted an elevated risk of neurodegenerative disease development and mortality in later life among former professional footballers.^3–8 Repetitive ball-to-head collisions are widely assumed to contribute to this elevated risk.

The mechanics of head impacts have been extensively studied, enabling the characterisation of collision phenomena which associate with acute symptomatic brain injuries such as skull fracture and concussion. However, the mechanisms that link football heading to long-term neurodegenerative outcomes remain poorly understood, largely due to the absence of immediate or observable symptoms. Studies involving human participants,^9–19 laboratory impact recreations,^11,20–28 and computational models^29–33 have been conducted to quantify or predict head kinematics and impact forces associated with football heading. These investigations have largely focussed on macromechanical responses, treating the head and ball as discrete bodies and analysing their relative motions, deformations and strain distributions under impact. While informative, such approaches do not generally capture the micromechanical dynamics, including local stress oscillations, wave propagation and higher-order tissue responses that occur on a spatial scale relative to brain microstructures such as those within the neurovascular unit. These phenomena may be critical for understanding subtle or subclinical brain injury mechanisms.

Collisions between solids induce rapid localised mechanical disturbances at the point of contact.^34,35 These disturbances generate pressure waves that propagate through the surrounding media, transferring energy and exerting mechanical stresses on structures within the impacted field.^34,36–38 Pressure waves are an established cause of acute and long-term degenerative brain injury in blast-exposed military personnel, in severe and repetitive low-level contexts, respectively.^39–48 Through computational modelling, Perkins et al.³² suggested that pressure waves could arise from football head collisions, but these have not been further explored or studied experimentally.³² Detailed analysis of force and contact area through a football head collision revealed peak contact pressure within the first 10% of contact time.⁴⁹ Moreover, impact neurotrauma studies indicate that potentially damaging strain magnitudes in the brain develop prior to the onset of gross head kinematics.^32,50,51 Football head collisions have been associated with acute non-concussive changes to the brain, including altered cerebrovascular reactivity and cortical control processes.^52,53 This study sought to experimentally test the hypothesis that football head collisions generate measurable intracranial pressure waves and to characterise their properties.

Pressure waves are commonly studied using laboratory re-enactments with human surrogate skulls filled with homogenous gel, which mimics the acoustic transmission characteristics of cerebrospinal fluid and brain tissue. Because the timescales over which pressure waves propagate within the dimensions of the head are orders of magnitude shorter than those associated with gross head kinematics, any acoustic signal is independent of, and fully captured prior to, any head displacement. Consequently, biomechanical fidelity is primarily required for the geometry and material properties of the surrogate skull, rather than for head or neck motion. Surrogates designed for this purpose are not required to mimic the inertial response of the head and are therefore typically rigidly mounted and may be sectioned (e.g. half-skull setups) for ease of sensor placement. Hydrophones or similar piezoelectric pressure sensors are positioned within the propagation field to measure the energy transmitted from the impact stimuli.^{32,50,51,54–58}

Considerable work has focussed on developing biofidelic skull surrogates for sport and related applications. Studies by Stone et al.,⁵⁹ England and Falland-Chueng et al.^60,61 have validated the mechanical properties of additively manufactured thermoplastic polymers, most commonly acrylonitrile butadiene styrene (ABS), as representative of cranial bone.⁶² These validations encompass flexural and tensile moduli, flexural and tensile strengths and density, making it an appropriate mechanical analogue. The longitudinal speed of sound in cranial bone (2200–2600 m/s^63,64) is comparable to that of additively manufactured ABS (2121–2770 m/s⁶⁵). As the brain and cerebrospinal fluid are predominantly water-based (acoustic impedance ≈ 1.5 × $10^{6}$ kg/m² s), water soluble gels with similar acoustic propagation characteristics are widely used as practical brain simulants.

The magnitude and characteristics of pressure waves generated during impact are determined by the mass, velocity and material properties of the colliding bodies.^37,66–69 In football, the Laws of the Game issued by the International Football Association Board (IFAB) specify the ball’s allowable mass at the start of play but do not define its materials or mechanical properties.⁷⁰ Historically, footballs were constructed from uncoated, hand-stitched leather panels, whereas modern designs utilise synthetic stitchless laminate constructions.^71–75 Ball velocity at the moment of head impact varies widely depending on the scenario, ranging from 5 m/s for throw-ins and close interplay up to ~30 m/s for shots and clearances.^23,74,76 Ball mass can change during play as a result of water exposure with different ball types exhibiting distinct mechanisms and magnitudes of water uptake.⁷¹

This study aimed to experimentally examine pressure wave propagation from football head collisions. It did not seek to determine the effect of pressure waves on the brain, nor draw any conclusions regarding brain health. Given the range of ball designs used through the history of the game it was necessary to include a variety of ball types within the experimental programme. The specific research questions were as follows.

RQ1: Do football head collisions generate measurable pressure waves and what are their fundamental characteristics?

RQ2: How are these characteristics modulated by ball type, inbound velocity or water uptake?

Methods

This study projected footballs, at game realistic velocities, against an instrumented human head surrogate to measure and characterise resultant pressure wave propagation within the cranial cavity.

Experimental apparatus and method

A bespoke pneumatic actuator (Red Streak Series STIWS) projected the ball 2.10 m in a horizontal plane to impact the surrogate skull. Inbound velocity was controlled by actuator pressure and measured prior to the collision using a pair of light gates (separation distance = 0.10 m). A single Photron Fastcam Sa1.1 (Photron, San Diego, USA) monochrome high-speed video camera (resolution = 512 × 352 pixels; frame rate=30,000 fps; shutter speed = 1/30,000 s), fitted with a Sigma MM4646-6 zoom lens (Sigma Corporation, Kanagawa, Japan), was positioned perpendicular to the ball direction of travel, at the front edge of the skull, to verify first contact between the skull and ball occurred within 3 mm of the intended position on the frontal bone of the forehead. The scene was illuminated using a single PIXAPRO LED panel light (PIXAPRO, Brierley Hill, UK) and a single Arri Arrisun 12+ light (Arri, Munich, Germany). A full schematic of the experimental setup is shown in Figure 1(a). Both hydrophone and video recordings were initiated by the light gate signal, allowing temporal alignment of hydrophone and video data for detection of the point of first contact (depicted in Figure 1(c)).

Figure 1.

(a) Plan view schematic diagram of experimental setup and apparatus for pressure wave impact testing, (b) illustration of skull surrogate showing cross section of half-skull setup with propagation medium within skull cavity and positioning of hydrophone relative to impact location, (c) video footage showing point of first contact between ball and head model and (d) representations of pressure-time and PPSI-time traces with key extraction metrics annotated.

Representative human skull geometry (25-year-old Japanese male; height = 173 cm; weight 65 kg), was obtained from an MRI scan from Tokyo Institute of Technology, in accordance with relevant guidelines and regulations, with appropriate ethical approval and informed consent.⁷⁷ Following the method of Stone and England,^62,78 this geometry was used to produce a skull surrogate. A sectioned half-skull configuration was developed, representing the forehead region, with a 50 mm deep cranial cavity. The skull was additively manufactured from ABS filament, using a Bambu Lab X1-Carbon (Bambu Lab, Shenzhen, China) 3D printer (filament material = 1.75 mm white ABS; nozzle diameter = 0.4 mm; layer height = 0.08 mm; infill density = 100%). This cavity was filled with degassed Aquasonic 100 ultrasound gel⁷⁹ (density = 1080 kg/m³; speed of sound = 1490 m/s; acoustic impedance=1.6 $\times 10^{6}$ kg/m² s), almost identically matching the acoustic properties of human cerebrospinal fluid, as identified through adult human studies (density = 995 kg/m³; propagation velocity = 1518 m/s; acoustic impedance = 1.5 $\times 10^{6}$ kg/m² s for adult males),⁸⁰ and bulk brain tissue (density = 1040 kg/m³; propagation velocity = 1540 m/s; acoustic impedance = 1.6 $\times 10^{6}$ kg/m² s).^81,82

The skull was rigidly mounted to a square steel plate (200 mm × 200 mm × 15 mm), secured to an aluminium box frame (440 mm × 330 mm × 260 mm), bonded to a 2 tonne concrete block. Two equidistant steel rods extended from the rear of the plate (length = 140 mm; diameter = 20 mm), onto which a flat measurement plate was attached (150 mm × 30 mm × 10 mm). This plate housed a depth micrometre (RS Components, Corby, UK; resolution = ±0.001 mm; barrel length = 25 mm).

Pressure signals were measured using a 4 mm diameter needle hydrophone (Precision Acoustics model NH4000, Dorchester, UK) connected to a pre-amplifier and DC coupler via a 10 m cable. The hydrophone was submerged in the acoustic gel within the skull cavity and positioned 5 mm from the interior skull surface; a distance greater than the subarachnoid space approximating a frontal brain location,⁸³ to capture the direct pressure wave path before reflections. The barrel of the hydrophone was fitted into an aluminium tube, which aligned its position to the impact position and velocity vector of the inbound ball. The hydrophone was immersed within the medium for at least 10 min prior to data acquisition to allow thermal and pressure acclimatisation. Signals were acquired using a Tektronix MDO4000C Series Mixed Domain Oscilloscope (Tektronix, Beaverton, USA) at a sampling rate of 10 MHz.

Unfiltered voltage-time signals from the hydrophone, calibrated to international ISO standards,⁸⁴ were converted to pressure-time traces using a linear calibration factor (12,590 mV/MPa), and analysed using custom MATLAB algorithms (version R2021b, MathWorks, Natick, USA). Metric-based signal characterisation enabled comparisons between balls, velocities and conditions. Both pressure-time and energy-time metrics were reported. Signal magnitude was characterised by the peak-to-peak pressure, determined by the difference between the peak overpressure and peak underpressure. The rise time was defined as the time from the initial disturbance of the noise floor, characterised and verified for stability using moving mean and moving standard deviation, to peak overpressure. In the energy-time domain, pulse pressure squared integral (PPSI) curves were derived from pressure-time data, calculated across the recorded signal from 1 ms before, to 1.5 ms after, the noise floor disturbance.⁸⁵ The 90% PPSI level ( ${PPSI}_{90}$ ) was defined as the difference between the fifth and ninety-fifth percentile of cumulative energy within the PPSI curves.^85,86 This metric quantifies the magnitude of energy transferred from the collision. The time interval between these percentiles ( ${PPSI}_{90}$ Time) indicates how long it took for 90% of the signal energy to be transferred.

Experimental design

Ball types

Twenty footballs were tested, all conforming to IFAB Law II.⁷⁰ They spanned seven ball types, representative of those used throughout the history of the game, encompassing a sample space of materials, constructions and assembly methods, representative of different “eras of play.” Each ball type was assigned an identification letter (A-G) and each ball model within this assigned a number (for example A1, A2). Both elite and recreational samples were included. The study was conducted independently of any ball manufacturer. Leather footballs degrade over time, such that original historical samples would not accurately reflect the properties of their respective era. Therefore, modern handmade replicas were used, carefully selected to reflect authentic shell construction, including hand-stitching, laces (where appropriate) and full-grain vegetable-tanned leather of suitable thickness, without anachronistic shell materials.

All 20 balls were pre-conditioned at room temperature for 24 h prior to testing (24°C, 30% humidity) and subsequently tested under standard conditions: dry, at room temperature, with an inflation pressure of 1.00 bar (the only inflation pressure always present through history when specified in IFAB Law II⁷⁰) and an inbound velocity of ~18 m/s. This velocity is typical of a corner kick,⁸⁷ a common preceding event to a header.⁸⁸ Ten sequential trials were conducted for each ball. Details of each ball are provided in the Supplemental Materials (Table 1).

Table 1.

Details of seven balls used to represent each ball type.

Ball type	Synthetic thermally bonded	Synthetic fuse welded	Synthetic machine stitched	Synthetic hand stitched	Synthetic moulded	Laceless leather hand stitched	Laced leather hand stitched
Representative colour and ID	A	B	C	D	E	F	G
Ball ID	A1	B1	C1	D1	E1	F1	G1
Stitched or stitchless	Stitchless	Stitchless	Stitched	Stitched	Stitchless	Stitched	Stitched
Dry mass (g)	436.5	436.9	427.3	424.3	437.0	439.2	449.6
Wet mass (g) (% increase from dry)	448.7 (2.6%)	444.7 (2.2%)	461.3 (6.9%)	474.6 (10.4%)	423.5 (1.4%)	506.0 (34.3%)	719.1 (59.9%)
Diameter (mm)	218.83	218.21	218.15	219.03	220.14	223.56	230.94
Representative era of play	2006–Present	2020–Present	2016–Present	2000–2012	1980s	1955–1975	1920–1955
Level of play	Elite	Elite	Elite	Elite	Recreational	Elite	Elite

Note. Colours presented are those used to represent each ball type in Figures later.

Inbound velocity

To efficiently examine the influence of velocity and mass on the pressure signal, one example ball from each of the seven ball types was selected (ball number one for each category, see Table 1). This selection was not intended to be indicative of ball type, but rather to provide a diverse range of samples for comparison. Each ball was projected at three inbound velocities: ~13 m/s (aerial pass), ~18 m/s (corner kick) and ~23 m/s (shot).⁸⁷ Five consecutive trials were conducted for each combination of ball and velocity.

Wet conditions

The same seven balls were also tested under wet conditions. Immediately prior to testing, each ball was inflated to 1.0 bar and fully submerged for 90 min in a water bath (water temperature = 16°C), consistent with previous water uptake testing protocols.⁷¹ Each ball was then allowed to drain for 10 min before testing, to remove excess surface water not representative of ball mass. Ball mass was measured prior to each impact using an EB Series Compact electronic scale (Ohaus, Parsippany, NJ, USA), following the procedure of FIFA Quality Programme for Footballs Test 05.⁸⁹ The skull surrogate was dried between collisions. Ten consecutive trials were conducted for each ball under both dry and wet conditions.

Statistical analysis

Results from repeated trials for each ball and condition were aggregated and reported as mean values with 95% confidence intervals. Differences between group means of individual balls under comparable conditions were assessed using pairwise Welch t-tests at α = 0.05, with Bonferroni correction to account for multiple comparisons.

To assess the influence of inbound velocity and ball type on pressure wave outcome metrics, a two-factor repeated measures analysis of variance (ANOVA) was conducted with velocity as a within-subjects factor and ball a between-subjects factor. The model assessed main effects of velocity, ball and their interaction; the latter indicating whether relationships between velocity and pressure wave characteristics varied based on the ball tested. Statistical significance of the relationships was reported through p-values assessed at the α = 0.05 significance level, with Greenhouse-Geisser correction. Effect sizes were determined using partial eta squared (η²). For each ball, post hoc one-way repeated measures ANOVAs were performed to investigate these differences.

When comparing dry and wet conditions, repeated measures analysis of variance (ANOVA) was used to assess the main effect of the ball, its condition and their interaction on each pressure wave outcome metric. Statistical significance of the relationships was reported through p-values assessed at the α = 0.05 significance level, with Greenhouse-Geisser correction, and effect sizes were determined using partial eta squared (η²). Post hoc paired t-tests (α = 0.05) were used to assess statistical significance within balls between dry and wet conditions.

Results

Pressure wave characteristics

Each of the 430 collisions conducted within this study resulted in a pressure measurement at a 5 mm depth inside the surrogate cranial cavity. Figure 2 shows a representative pressure-time signal for each of the seven balls in Table 1, and examples from individual balls are available in the Supplemental Materials. The first point of contact was timestamped as $t = 0$ , as visually depicted in Figure 1(c). Prior to this point, a continuous equilibrium pressure was observed, which was deemed the noise floor. The first disturbance from this noise floor – the direct path arrival – occurred a finite time after the point of first contact. This observed delay was never shorter than 6 μs, the theoretical propagation time from the outer surface of the skull to the hydrophone, and never greater than 100 μs. These findings confirm that the measured pressure wave was generated by, and propagated from, the region of initial contact between the ball and head. The exact timing of the delay varied between balls and may be related to surface geometry. Detailed investigation of these micromechanical effects was beyond the scope of this study.

Figure 2.

Representative pressure-time traces for an example ball from each ball type. The start of contact between the ball and head is timestamped as t = 0. Each subplot shows 500 μs of signal following this. Vertical axes are common between each subplot to visualise differences in pressure magnitudes between balls.

Within the first few hundred microseconds, the signals showed a rapid increase to a compressional overpressure peak, followed by a tensional underpressure. Ninety percent of energy within each signal was contained within the first 500 μs, before the pressure returned to equilibrium. The signals then decayed, returning to equilibrium within 500 μs (0.5 ms) of the time of first contact. Across repeated trials with the same ball at the same velocity, the signals exhibited comparable acoustic signatures and repeatable magnitudes, with peak-to-peak values yielding a coefficient of variation between 2.2% and 29.1%.

Differences between balls

Pressure signals were measured from collisions with 20 different balls, projected 10 times each, at 18.20 m/s ± 0.27 (1 SD) m/s, in dry conditions. The mean peak to peak pressure from each ball was found to vary from a minimum of 3.4 kPa (G1) to a maximum of 31.0 kPa (C1, Figure 3(a)). The moulded ball D1 generated the highest mean peak-to-peak pressure, 9.1-fold greater than the uncoated laced leather ball (G1, 3.4 kPa) and 1.6-fold greater than the laceless coated leather ball (F1, 5.6 kPa). Substantial variability was observed among synthetic balls, and even between balls of the same type, with a mean peak-to-peak pressure range of 12.6 kPa for synthetic thermally bonded balls and 20.0 kPa for synthetic hand stitched balls. The elite synthetic machine stitched ball (C1) exhibited the lowest peak-to-peak pressure of any synthetic ball (5.7 kPa), significantly lower than all other synthetic balls.

Figure 3.

Results for all balls tested in standard conditions (dry, room temperature, 1.0 bar inflation pressure, 18 m/s velocity) for: (a) peak-to-peak magnitude, (b) peak-to-peak time, (c) ${PPSI}_{90}$ and (d) ${PPSI}_{90} Time$ . Error bars represent the 95% confidence interval across 10 impacts.

The mean rise time varied between 2.7 and 84.9 μs representing a 31.4-fold difference between balls (Figure 3(b)). Leather balls exhibited significantly longer rise times than all synthetic balls, except for the machine stitched ball C1. The uncoated laced leather ball G1 exhibited the longest rise time – more than twice that of its laceless coated counterpart F1 (p = 0.004). When both pressure-time characteristics were considered, the machine stitched ball C1 behaved comparably to the leather balls and was a clear outlier to the synthetic balls. The shortest mean rise time was observed in the fuse welded synthetic ball B1. Notably, among the synthetic samples, many exhibited similar rise times, but significantly different peak-to-peak magnitudes, indicating little or no dependency between pressure wave magnitudes and temporal characteristics.

When energy-time characteristics were considered, the ${PPSI}_{90}$ exhibited 54.7-fold difference between balls (Figure 3(c)). The magnitude of these differences reflected the quadratic relationship between energy and the measured pressure. Trends between balls were similar to those observed for pressure, with leather balls producing the lowest energy transfer, significantly lower than all synthetic balls, with the exception of the machine-stitched synthetic ball. However, the difference between the two leather balls was not significant (p = 0.852). The three highest energy transfers were recorded for the synthetic moulded ball D1 (3601 Pa² s), hand stitched synthetic ball E4 (1792 Pa² s) and thermally bonded synthetic ball A4 (1081 Pa² s), each statistically greater than all remaining balls.

The duration over which the 90% of the energy contained within the signal was delivered ranged from 137 to 494 μs (Figure 3(d)). The uncoated laced leather ball exhibited the longest mean ${PPSI}_{90} Time$ , 77.7% longer than its coated laceless counterpart (p < 0.001); however, characteristic differences between leather and synthetic ball types were not distinguishable using this metric. Variability was greatest within the thermally bonded ball type where, for example, two balls of similar construction (A2 and A4), manufactured 10 years apart, exhibited a 3.4-fold difference (p = 0.064). The coefficient of variance range here was 12.1%–146.9%, a far larger range than for ${PPSI}_{90}$ (14.8%–57.6%).

Effect of velocity

A sample of seven balls was tested at three different inbound velocities, with five impacts per ball, per velocity. Measured velocities were 13.90 ± 0.18 (1 SD) m/s, typical of an aerial pass, 18.29 ± 0.25 (1 SD) m/s, typical of a corner kick and 23.46 ± 0.15 (1 SD) m/s, representative of a shot. Across the range of velocities, the general characteristics of the pressure-time signals depicted in Figure 2 were retained, including the delay from the start of contact and typical signal signatures. However, increasing the inbound velocity notably affected the characteristics of the pressure signals.

Peak-to-peak pressure increased up to 5-fold for a 10 m/s (22.37 mph; 36.00 km/h) increase in velocity, approximately corresponding to the difference between heading a typical cross-field pass and blocking a shot (Figure 4(a)). Repeated measures ANOVA showed that peak-to-peak pressure was influenced by both velocity (η² = 0.85; p < 0.001) and ball (η² = 0.98; p < 0.001). A significant interaction effect (η² = 0.67; p < 0.001) between these also revealed the effect of velocity was different between balls tested.

Figure 4.

Changes in: (a) peak-to-peak pressure, (b) peak-to-peak duration, (c) ${PPSI}_{90}$ and (d) ${PPSI}_{90} Time$ with inbound velocity. Each marker presents a mean value for a given ball at a specific velocity. Error bars represent the 95% confidence interval across 10 impacts.

Increasing velocity significantly increased peak-to-peak magnitudes for each ball (p < 0.001 in each case). Different balls responded differently to velocity such that their order changed at different velocities. The difference between the maximum and minimum mean peak-to-peak pressures across all balls at 18 m/s was 27.6 kPa, representing a 9.1-fold difference. At 13 m/s this difference decreased to 16.3 kPa, corresponding to an 11.2-fold difference while at 23 m/s it was 23.3 kPa, representing a 5.5-fold difference.

Both ball (η² = 0.87, p < 0.001), velocity (η² = 0.78; p < 0.001) and their interaction (η² = 0.68; p < 0.001) statistically significantly influenced rise times. The range of mean rise times between balls decreased with inbound velocity (Figure 4(b)). At 13 m/s, the difference in rise time between balls exceeded 90 μs whereas at 23 m/s the balls clustered, with a range of only 7.8 μs. Each ball exhibited significantly shorter rise times with increased inbound velocity, indicating that as the ball shell was deformed more rapidly, the influence of materials and construction on temporal characteristics of the resulting pressure signal reduced. For example, leather balls, which differed from synthetic balls at lower velocities, behaved comparably to them at higher velocities.

Additionally, ball (η² = 0.98, p < 0.001), velocity (η² = 0.85; p < 0.001) and their interaction (η² = 0.92; p < 0.001) statistically significantly influenced ${PPSI}_{90}$ . With increasing velocity, the magnitude of energy transfer statistically significantly increased for each ball tested (p ≤ 0.002 in each case). The synthetic thermally bonded and moulded balls exhibited an approximately linear increase with velocity, whereas the synthetic fuse-welded and machine-stitched balls showed a non-linear increase. For the moulded ball, the plateau observed at higher velocities for peak-to-peak pressure was not present for PPSI₉₀. The difference between maximum and minimum PPSI₉₀ values across balls was 536 Pa² s at 13 m/s, 2823 Pa² s at 18 m/s and 4409 Pa² s at 23 m/s (Figure 4(c)).

For ${PPSI}_{90} Time$ , weaker main effects were observed for ball (η² = 0.67; p < 0.001), velocity (η² = 0.16; p = 0.009) and their interaction (η² = 0.54; p< 0.001). Both leather balls delivered their energy over significantly longer durations than all synthetic balls except for the thermally bonded ball. The range in ${PPSI}_{90} Time$ was 429 μs at 13 m/s, decreased to 223 μs at 18 m/s, and increased to 878 μs at 23 m/s, with both leather balls exhibiting longer energy transfer times at the highest velocity (Figure 4(d)).

Differences between dry and wet conditions

Seven balls were tested in both dry and wet conditions. Repeated measures ANOVA revealed that the ball (η² = 0.94; p < 0.001) showed a larger main effect on peak-to-peak pressure than condition (η² = 0.39; p < 0.001). However, a strong significant effect of their interaction (η² = 0.86; p < 0.001) highlighted that the influence of condition was not consistent between balls. Post hoc pairwise comparisons revealed significant differences between dry and wet conditions for each ball except B1 (p = 0.083), however the direction of difference differed between balls. The most pronounced differences between conditions were observed in the leather balls (For F1, T = −23.0, p < 0.001; for G1, T = −5.9, p < 0.001). Representative pressure-time traces for leather balls under dry and wet conditions are shown in Figure 5. Mean peak-to-peak amplitudes increased by 277% and 317% for the uncoated and coated balls, respectively (Figure 6(a)). The mean peak-to-peak pressure was 23.1 kPa for the wet coated laceless ball, exceeding the pressure measured in 17 of the 20 balls tested under standard conditions, and 12.9 kPa for the uncoated laced ball, exceeding 9 of 20 balls under standard conditions.

Figure 5.

Representative pressure-time traces for leather footballs in (a and c) dry and (b and d) wet conditions. Axes scales are consistent between each trace to visualise differences. Dry traces for these balls are visualised in greater detail in Figure 1.

Figure 6.

Percentage differences from dry to wet conditions, for each ball, for: (a) peak-to-peak-pressure, (b) rise time, (c) ${PPSI}_{90}$ and (d) ${PPSI}_{90} Time$ . Statistically significant differences between dry and wet conditions for a given ball, as identified through post hoc t-tests, are indicated by asterisk icons. Error bars represent 95% confidence intervals.

Ball (η² = 0.90, p < 0.001), condition (η² = 0.22; p< 0.001) and their interaction (η² = 0.58; p < 0.001) also statistically significantly influenced rise times. Leather balls demonstrated significantly shorter rise times when wet (Figure 6(b)), with mean values of 15.2 and 56.4 μs for the coated laceless (T = 2.6; p = 0.031) and uncoated laced balls (T = 9.4; p < 0.001), respectively. These values were comparable to those of synthetic balls under both dry and wet conditions. Therefore, while pressure-time characteristics of leather balls differed from synthetic balls when dry, their behaviour when wet was more similar to that of synthetic balls.

Whilst there was no main effect of wet condition on ${PPSI}_{90}$ (η² = 0.03; p = 0.16), both the ball (η² = 0.88; p < 0.001) and interaction variable (η² = 0.76; p < 0.001) showed significant effects, where the magnitude of energy transferred during the collision differed significantly between dry and wet for each ball (Figure 6(c)). For four synthetic balls, the energy decreased by 38%–70% (p < 0.039). In contrast, the leather balls exhibited the largest changes, with energy increasing by 836%–1519% (p < 0.002). Leather balls, which showed the lowest energy magnitudes in dry conditions (mean values of 68.3 and 65.8 Pa² s for the coated laceless and uncoated laced balls, respectively) increased to values that exceeded most synthetic balls when wet (1106.4 and 616.1 Pa² s, respectively), with the exception of the synthetic moulded ball.

The ball (η² = 0.74; p < 0.001) and interaction variable (η² = 0.63; p < 0.001) showed significant main effects on the ${PPSI}_{90} Time$ , with no main effect for condition (η² < 0.01; p = 0.680). For six of the seven balls, the mean ${PPSI}_{90} Time$ decreased in wet conditions compared with dry (statistically significant for A1, F1 and D1). Synthetic hand stitched ball E1 was the exception, increasing by 115% when wet (T = −5.0; p< 0.001). While the duration decreased significantly for the coated laceless leather ball (T = 4.9; p < 0.001), this effect was not observed for the uncoated laced leather ball (T = 0.9; p = 0.373) (Figure 6(d)).

Discussion

This study experimentally documented pressure wave propagation from football head collisions. Following impact, pressure waves arrived at the hydrophone inside the skull cavity within 100 μs of the first contact with the head, emanating from the region of impact. Acoustic signatures, magnitudes and temporal characteristics were repeatable, with energy propagated within 0.5 ms – before gross head kinematics typically begin. These findings are consistent with computational simulations and animal neurotrauma models,^32,50 indicating that energy from a football collision is transferred earlier and over a shorter temporal and more localised spatial domain than previously considered.

Although this study did not seek to investigate the effects of pressure waves on brain tissue directly, several contextual observations can be made. Pressure signals were comparable to those documented in some military gunshots and blast scenarios.^90,91 Peak-to-peak pressures of up to 32.9 kPa were measured, whereas neurological modifications have been observed in military personnel at blast overpressures as low as 10 kPa.⁹² The US Department of Defence blast overpressure exposure limit for training, measured outside the head, is 27 kPa, and is currently under review due to associations between low-level blast exposure and neurodegeneration.⁹³ Peak overpressures of typical military and civilian firearms are approximately 2 kPa, again measured outside the head, and can exhibit comparable acoustic signatures and temporal evolution to signals measured from football head collisions.^91,94,95 The location of the measurement within the model approximates the frontal brain region, which appears most susceptible to neurodegeneration in former professional footballers, as evidenced through post-mortem analysis,⁹⁶ dementia screening tests,⁷ and neuroradiology examinations.⁹⁷ Animal impact neurotrauma studies have measured blood-brain barrier disruption, and simulations have predicted the greatest brain strain in this region, prior to the onset of gross head kinematics.^50,51 Blood-brain barrier disfunction is a “hallmark pathology” of Chronic Traumatic Encephalopathy, which has been identified in several former footballers.^96,98–100 While these observations do not imply causation between pressure waves and neurodegeneration in footballers, they suggest that pressure waves cannot be excluded as a potential contributing mechanism. Features of the measured signals may have clinical implications. For example, greater pressure and energy magnitudes will likely increase mechanical stress on brain tissue subjected to pressure waves. Shorter rise times would indicate higher loading rates delivered to viscoelastic tissue, which may result in greater damage susceptibility.

Large differences were observed between footballs conforming to IFAB Law II.⁷⁰ Previous studies comparing peak force or peak head accelerations resulting from impacts of different footballs at similar velocities reported differences no greater than 20%.^24,87,101 In contrast, this study observed a 9-fold difference in peak-to-peak pressure, and 54-fold difference in energy transfer between balls. These findings highlight the influence of ball shell materials and construction on the magnitude and duration of pressure waves and suggest a complex micromechanical interaction during impact. Greater apparent trial-to-trial variability with some balls suggests even local orientation differences may influence energy transfer. Variability in the delay between the start of contact and direct path arrival of the pressure wave, as well as differences in signal complexity between balls, support the need for further detailed micromechanical investigation. Further analysis of local ball material properties would be required to explain observed differences in pressure waves between balls.

Inbound velocity significantly influenced pressure magnitudes. Relationships with velocity varied between balls. These differences likely reflect underlying dynamic high-strain rate dependent compressive micromechanics within the ball shell. These may also explain temporal characteristics such as the lower range of rise times at high velocities, where stiffness responses may be rate dependent. The variable range of energy and pressure magnitudes between velocities may be consequential of local ball geometry and layer thicknesses, but these underlying mechanics require further investigation. The apparent non-linear relationships of some balls with velocity support the need for future testing to be conducted over a range of game-realistic velocities.

Leather balls produced the lowest pressure magnitudes in dry conditions, but these magnitudes increased substantially when wet. Unlike synthetic balls, leather balls absorb water directly into their outer panels and via ingress between seams, increasing local mass at the point of contact.⁷¹ Water absorption may also increase local dynamic stiffness due to the incompressibility of water within the fibrous matrix of leather, as well as surface properties, however it is not possible to decouple these confounding factors. Further investigation is required to explain why some synthetic balls showed small reductions in pressure magnitudes in wet conditions, and may support the need for future testing, including new or existing standards tests, to be conducted in varied conditions.

This study has several limitations. A simplified head and brain model was used which did not incorporate detailed microstructure. In-vivo validation is lacking and will be challenging to achieve. The use of a single skull geometry limits generalisability of absolute pressure values to individuals with different cranial anatomies, such as different genders and ages. Further, the half-skull configuration provides a first-order approximation of the intracranial pressure field and does not facilitate complex 3D wave reflections. Further studies mapping these patterns may reveal detailed spatial distributions, constructive interference and focussing which may present pressure hotspots within brain geometry. Although pressure waves, particularly those from low-level blast, have been studied in military contexts, their effects on human brain tissue, specifically from football collisions, remain unproven. In addition, this study assessed a subset of balls broadly representative of different ball types however, to fully capture variability within and between ball types, wider ball testing is required. Some balls, for some metrics, demonstrated notable trial-to-trial variability. The signals measured in this study were complex, and despite tight experimental control, variance may be induced by subtle impact location, environmental variability or microscopic ball geometry.

Further studies exposing human tissue or bioengineered structures to these phenomena are required to determine their relevance to acute and long-term brain health. Integration of computational modelling to assess pressure wave propagation and energy distribution within the brain would be valuable. Wider scale ball testing and detailed micromechanical characterisation of footballs are needed to fully explore the future engineering, health or governance implications of these findings. Exploration of local, high strain rate, material stiffness and damping properties within the construction of the ball shell, and how they influence characteristics such as pressure magnitudes and rise times, may inform avenues for ball designs that would mitigate energy transfer characteristics.

The findings of this study suggest that pressure wave energy from football head collisions may be influenced by the materials and construction of the ball, and there is scope to affect this energy transfer without requiring alteration to the game or its Laws.

Conclusion

This study experimentally measured repeatable pressure wave propagation from football head collisions. The magnitude and duration of these waves were influenced by ball type, inbound velocity and water uptake. These findings provide evidence for a previously unidentified component of the mechanical insult experienced during a football head collision, and that it is distinct from macroscopic gross head kinematics. These findings provide a new mechanistic basis for future safety considerations, with potential to influence football design and equipment standards, moving beyond the current focus on head kinematics.

Supplemental Material

sj-docx-1-pip-10.1177_17543371261438388 – Supplemental material for Pressure wave propagation from association football head collisions

Supplemental material, sj-docx-1-pip-10.1177_17543371261438388 for Pressure wave propagation from association football head collisions by Ieuan Phillips, Séan Mitchell, Paul Lepper and Andy Harland in Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology

Footnotes

Acknowledgements

The authors wish to thank The Football Association for their donation that supported this research.

ORCID iD

Ieuan Phillips

Author contributions

All authors collaboratively developed the fundamental ideas and methodology of the study. IP was responsible for data collection, analysis and manuscript preparation, under the close supervision of AH, SM and PL, who also edited the final version.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was conducted following a philanthropic donation from The Football Association.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Data available from authors upon reasonable request and subject to appropriate sharing agreements. Contact A.R.Harland@lboro.ac.uk with enquiries.

Supplemental material

Supplemental material for this article is available online.

References

FIFA. FIFA big count 2006. FIFA Commun Div Inf Serv 2007; 31: 1–12.

Miller

Fan engagement: football clubs consider themselves content providers. They need to start acting like it. Financial TImes, https://www.ftstrategies.com/en-gb/insights/fan-engagement/#:~:text=Young%20or%20old%2C%20players%20or,to%20dispute%20football’s%20global%20appeal (2024, accessed 24 September 2024).

Mackay

Russell

Stewart

, et al. Neurodegenerative disease mortality among former professional soccer players. N Engl J Med 2019; 381: 1801–1808.

Russell

MacKay

Stewart

, et al. Association of field position and career length with risk of neurodegenerative disease in male former professional soccer players. JAMA Neurol 2021; 78: 1057–1063.

Ueda

Pasternak

Lim

C-E

, et al. Neurodegenerative disease among male elite football (soccer) players in Sweden: a cohort study. Lancet Public Health. Epub ahead of print April 2023. https://doi.org/10.1016/s2468-2667(23)00027-0

Orhant

Carling

Chapellier

, et al. A retrospective analysis of all-cause and cause-specific mortality rates in French male professional footballers. Scand J Med Sci Sports 2022; 32: 1389–1399.

Macnab

Espahbodi

Hogervorst

, et al. Cognitive impairment and self-reported dementia in UK retired professional soccer players: a cross sectional comparative study. Sports Med Open 2023; 9: 43.

Espahbodi

Hogervorst

Macnab

T-MP

, et al. Heading frequency and risk of cognitive impairment in retired male professional soccer players. JAMA Netw Open 2023; 6: e2323822.

McCuen

Svaldi

Breedlove

, et al. Collegiate women’s soccer players suffer greater cumulative head impacts than their high school counterparts. J Biomech 2015; 48: 3720–3723.

10.

Caccese

Lamond

Buckley

, et al. Reducing purposeful headers from goal kicks and punts may reduce cumulative exposure to head acceleration. Res Sports Med 2016; 24: 407–415.

11.

Shewchenko

Withnall

Keown

, et al. Heading in football. Part 2: Biomechanics of ball heading and head response. Br J Sports Med 2005; 39: 26–32.

12.

Lynall

Clark

Grand

, et al. Head impact biomechanics in women’s college soccer. Med Sci Sports Exerc 2016; 48: 1772–1778.

13.

Dezman

ZDW

Ledet

Kerr

HA.

Neck strength imbalance correlates with increased head acceleration in soccer heading. Sports Health 2013; 5: 320–326.

14.

Sokol-Randell

Stelzer-Hiller

Allan

, et al. Heads up! A biomechanical pilot investigation of soccer heading using instrumented mouthguards (iMGs). Appl Sci 2023; 13: 2639. https://doi.org/10.3390/app13042639

15.

Barnes-Wood

McCloskey

Connelly

, et al. Investigation of head kinematics and brain strain response during soccer heading using a custom-fit instrumented mouthguard. Ann Biomed Eng 2024; 52(4): 934–945. https://doi.org/10.1007/s10439-023-03430-8

16.

Miyazaki

Sakuraba

Kimachi

, et al. Comparison of brain strain in ball heading and concussion-level sports head impact cases. In: ISEA 2024 - The Engineering of Sport 15. Loughborough University, 2024, pp. 183–184.

17.

Hanlon

Bir

Validation of a wireless head acceleration measurement system for use in soccer play. J Appl Biomech 2010; 26: 424–431.

18.

Lamond

Caccese

Buckley

, et al. Linear acceleration in direct head contact across impact type, player position, and playing scenario in collegiate women’s soccer players. J Athl Train 2018; 53: 115–121.

19.

Shewchenko

Withnall

Keown

, et al. Heading in football. Part 1: development of biomechanical methods to investigate head response. Br J Sports Med 2005; 39: 10–25.

20.

Cecchi

Monroe

Moscoso

, et al. Effects of soccer ball inflation pressure and velocity on peak linear and rotational accelerations of ball-to-head impacts. Sports Eng 2020; 23: 12–17.

21.

Rowson

Duma

SM.

A review of head injury metrics used in automotive safety and sports protective equipment. J Biomech Eng 2022; 144: 110801. https://doi.org/10.1115/1.4054379

22.

Naunheim

Ryden

Standeven

, et al. Does soccer headgear attenuate the impact when heading a soccer ball? Acad Emerg Med 2003; 10: 85–90.

23.

Ferdousi

Bilodeau

Robidoux

Comparison of frequency and impact magnitude of heading in 1966 and 2018 international professional men’s soccer matches. University of Ottawa, 2021.

24.

England

Liverseidge

Miyazaki

, et al. Does reduced ball inflation pressure in association football decrease head impact kinematics? Ann Biomed Eng 2025; 53(11): 3109–3125. https://doi.org/10.1007/s10439-025-03804-0.

25.

Mills

Billingham

Choppin

, et al. Repeatability of a piezoelectric force platform to measure impact metrics for a single model of football. Sports Eng 2022; 25: 24. https://doi.org/10.1007/s12283-022-00389-y.

26.

Auger

Markel

Pecoski

, et al. Factors affecting peak impact force during soccer headers and implications for the mitigation of head injuries. PLoS ONE 2020; 15: e0240162. https://doi.org/10.1371/journal.pone.0240162

27.

Shewchenko

Withnall

Keown

, et al. Heading in football. Part 3: effect of ball properties on head response. Br J Sports Med 2005; 39: 33–39.

28.

Withnall

Shewchenko

Wonnacott

, et al. Effectiveness of headgear in football. Br J Sports Med 2005; 39: 40–48.

29.

Christenson

Casa

DJ.

Analysis on the effect of ball pressure on head acceleration to ensure safety in soccer. Proc West Mark Ed Assoc Conf 2020; 49: 3.

30.

Queen

Weinhold

Kirkendall

, et al. Theoretical study of the effect of ball properties on impact force in soccer heading. Med Sci Sports Exerc 2003; 35: 2069–2076.

31.

Sharma

Smith

The effect of pressure and kick speed in soccer on head injury and goalkeeper movement. In: Reilly

Lees

Davids

, et al. (eds) The engineering of sport. Liverpool: Routledge, 2022, p.15.

32.

Perkins

Bakhtiarydavijani

Ivanoff

, et al. Assessment of brain injury biomechanics in soccer heading using finite element analysis. Brain Multiphys 2022; 3: 100052. https://doi.org/10.1016/j.brain.2022.100052.

33.

Huber

Patton

Maheshwari

, et al. Finite element brain deformation in adolescent soccer heading. Comput Methods Biomech Biomed Engin 2024; 27: 1239–1249.

34.

Bayman

Model of the behavior of solid objects during collision. Am J Phys 1976; 7: 671–676.

35.

Hertz

On the contact of elastic solids. Miscellaneous Papers, Capter V 1881, pp. 146–162.

36.

Maury

Wright

Nelson

Point sources. University of Southampton Institute of Sound and Vibration Research, https://blog.soton.ac.uk/soundwaves/wave-basics/point-sources-inverse-square-law/ (2024, accessed 3 December 2024).

37.

McLaskey

Glaser

SD.

Hertzian impact: experimental study of the force pulse and resulting stress waves. J Acoust Soc Am 2010; 128: 1087.

38.

Chandra

Ganpule

Kleinschmit

, et al. Evolution of blast wave profiles in simulated air blasts: experiment and computational modeling. Shock Waves 2012; 22: 403–415.

39.

Jakkula

Ganzenmüller

Beisel

, et al. Investigating slow shock in low-impedance materials using a direct impact Hopkinson bar setup. EPJ Web Conf 2021; 250: 06009.

40.

Säljö

Mayorga

Bolouri

, et al. Mechanisms and pathophysiology of the low-level blast brain injury in animal models. Neuroimage 2011; 54: S83–S88. https://doi.org/10.1016/j.neuroimage.2010.05.050

41.

Stone

Avants

Tustison

, et al. Functional and structural neuroimaging correlates of repetitive low-level blast exposure in career breachers. J Neurotrauma 2020; 37: 2468–2481.

42.

Elder

Gama Sosa

De Gasperi

, et al. The Neurovascular Unit as a Locus of Injury in Low-Level Blast-Induced Neurotrauma. Int J Mol Sci 2024; 25: 1150. https://doi.org/10.3390/ijms25021150

43.

Heyburn

Dahal

Abutarboush

, et al. Differential effects on TDP-43, piezo-2, tight-junction proteins in various brain regions following repetitive low-intensity blast overpressure. Front Neurol 2023; 14: 1237647. https://doi.org/10.3389/fneur.2023.1237647

44.

Gama Sosa

De Gasperi

Pryor

, et al. Low-level blast exposure induces chronic vascular remodeling, perivascular astrocytic degeneration and vascular-associated neuroinflammation. Acta Neuropathol Commun 2021; 9: 167. https://doi.org/10.1186/s40478-021-01269-5

45.

Ravula

Rodriguez

Younger

, et al. Animal model of repeated low-level blast traumatic brain injury displays acute and chronic neurobehavioral and neuropathological changes. Exp Neurol 2022; 349: 113938. https://doi.org/10.1016/j.expneurol.2021.113938

46.

Gama Sosa

De Gasperi

Perez Garcia

, et al. Low-level blast exposure disrupts gliovascular and neurovascular connections and induces a chronic vascular pathology in rat brain. Acta Neuropathol Commun 2019; 7: 6.

47.

Kilgore

Hubbard

WB.

Effects of low-level blast on neurovascular health and cerebral blood flow: current findings and future opportunities in neuroimaging. Int J Mol Sci 2024; 25: 642. https://doi.org/10.3390/ijms25010642

48.

Heyburn

Abutarboush

Goodrich

, et al. Repeated low-level blast overpressure leads to endovascular disruption and alterations in TDP-43 and Piezo2 in a rat model of blast TBI. Front Neurol 2019; 10: 766. https://doi.org/10.3389/fneur.2019.00766

49.

Phillips

Harland

Mitchell

, et al. Non-linear temporal contact pressure for quantification of impact severity from Assocication Football Collisions. In: ISEA 2024 - The Engineering of Sport 15. Loughborough University, https://repository.lboro.ac.uk/articles/conference_contribution/Non-linear_temporal_contact_pressure_for_quantification_of_impact_severity_from_association_football_collisions/27045508?file=49243951 (8–11 July 2024, accessed 18 October 2024).

50.

Tagge

Fisher

Minaeva

, et al. Concussion, microvascular injury, and early tauopathy in young athletes after impact head injury and an impact concussion mouse model. Brain 2018; 141: 422–458.

51.

Watts

Long

, et al. Spatiotemporal changes in blood-brain barrier permeability, cerebral blood flow, T2 and diffusion following mild traumatic brain injury. Brain Res 2016; 1646: 53–61.

52.

Svaldi

Joshi

McCuen

, et al. Accumulation of high magnitude acceleration events predicts cerebrovascular reactivity changes in female high school soccer athletes. Brain Imaging Behav 2020; 14: 164–174.

53.

Svaldi

McCuen

Joshi

, et al. Cerebrovascular reactivity changes in asymptomatic female athletes attributable to high school soccer participation. Brain Imaging Behav 2017; 11: 98–112.

54.

Lam

Dong

Shokrollahi

, et al. Finite element analysis of soccer ball-related ocular and retinal trauma and comparison with abusive head trauma. Ophthalmol Sci 2022; 2: 100129. https://doi.org/10.1016/j.xops.2022.100129

55.

Gâteau

Marsac

Pernot

, et al. Transcranial ultrasonic therapy based on time reversal of acoustically induced cavitation bubble signature. IEEE Trans Biomed Eng 2010; 57: 134–144.

56.

Clement

White

King

, et al. A magnetic resonance imaging-compatible, large-scale array for trans-skull ultrasound surgery and therapy. J Ultrasound Med 2005; 24: 1117–1125.

57.

Estrada

Rebling

Razansky

Prediction and near-field observation of skull-guided acoustic waves. Phys Med Biol 2017; 62: 4728–4740.

58.

Dc Eames

Hananel

Snell

, et al. Trans-cranial focused ultrasound without hair shaving: feasibility study in an ex vivo cadaver model, http://www.jtultrasound.com/content/1/1/24 (2013, accessed 19 September 2025).

59.

Stone

Mitchell

Miyazaki

, et al. A destructible headform for the assessment of sports impacts. Proc IMechE, Part P: J Sports Engineering and Technology 2023; 237: 7–18.

60.

Falland-Cheung

Waddell

Chun Li

, et al. Investigation of the elastic modulus, tensile and flexural strength of five skull simulant materials for impact testing of a forensic skin/skull/brain model. J Mech Behav Biomed Mater 2017; 68: 303–307.

61.

Falland-Cheung

Scholze

Lozano

, et al. Mechanical properties of the human scalp in tension. J Mech Behav Biomed Mater 2018; 84: 188–197.

62.

England

RM.

Advanced biofidelic headforms for high fidelity re-enactment of real-world head impacts in cricket. Loughborugh: Loughborugh University, 2023.

63.

Wydra

Malyarenko

Shapoori

, et al. Development of a practical ultrasonic approach for simultaneous measurement of the thickness and the sound speed in human skull bones: a laboratory phantom study. Phys Med Biol 2013; 58: 1083–1102. https://iopscience.iop.org/article/10.1088/0031-9155/58/4/1083/pdf (2013, accessed 21 October 2024).

64.

Pichardo

Moreno-Hernández

Andrew Drainville

, et al. A viscoelastic model for the prediction of transcranial ultrasound propagation: application for the estimation of shear acoustic properties in the human skull. Phys Med Biol 2017; 62: 6938–6962.

65.

Antoniou

Evripidou

Giannakou

, et al. Acoustical properties of 3D printed thermoplastics. J Acoust Soc Am 2021; 149: 2854–2864.

66.

Zhu

Tutorial on hertz contact stress. The University of Arizona, https://wp.optics.arizona.edu/optomech/wp-content/uploads/sites/53/2016/10/OPTI-521-Tutorial-on-Hertz-contact-stress-Xiaoyin-Zhu.pdf (2012, accessed 10 June 2025).

67.

McLaskey

Lockner

Kilgore

, et al. A robust calibration technique for acoustic emission systems based on momentum transfer from a ball drop. Bull Seismol Soc Am 2015; 105: 257–271.

68.

Mizutani

Nishiwaki

Takemoto

, et al. Vibration and acoustic emission monitoring of ball impact on a composite plate. AIP Publishing, New York, 2008, pp. 9–14.

69.

Petersen

TB.

Acoustic emission from impacts of rigid bodies, https://www.ndt.net/?id=10858 (2009, accessed 19 September 2025).

70.

IFAB. Laws of the Game 20/21, https://www.theifab.com/ (2020, accessed 12 March 2022).

71.

Phillips

Harland

Mitchell

, et al. Investigating static water uptake characteristics and mechanisms for leather and synthetic association footballs. Sports Eng 2025; 28: 40.

72.

Hanson

Harland

The 16th century football: 450 year old sports technology assessed by modern standards. In: 9th conference of the international sports engineering association (ISEA). Lowell, MA, 9–13 July 2012.

73.

Hanson

Material and construction influences on football impact behaviour. Thesis, Loughborough University, 2014.

74.

Neilson

The dynamic testing of soccer balls. PhD, Sports Technology Institute, Loughborough University, https://repository.lboro.ac.uk/articles/thesis/The_dynamic_testing_of_soccer_balls/9531353 (2003, accessed 26 July 2023).

75.

Price

Jones

Harland

AR.

Advanced finite-element modelling of a 32-panel soccer ball. Proc IMechE Part C: J Mechanical Engineering Science 2007; 221: 1309–1319.

76.

Cook

Goff

JE.

Parameter space for successful soccer kicks. Eur J Phys 2006; 27: 865–874.

77.

Pramudita

Miyazaki

Kouchi

, et al. Brain responses to rotational and translational impact, and influence of individual differences in head shape. Trans Jpn Soc Mech Eng A 2010; 761: 44–51.

78.

Stone

BW.

An assessment of the mechanics of protected and unprotected head impacts in cricket using representative impact conditions and a sport specific anthropomorphic test device. Loughborough: Loughborough University, 2019.

79.

Parker Laboratories. Aquasonic 100 Blue Technical Datasheet, https://bio-medical.com/media/support/aquasonic-100-blue-technical-datasheet.pdf (2025, accessed 8 May 2025).

80.

Palaniappan

Velusamy

Ultrasonic study of human cerebrospinal fluid. Indian J Pure Appl Phys 2004; 42: 591–594.

81.

Abdi

Sánchez-Molina

García-Vilana

, et al. Revealing the role of material properties in impact-related injuries: investigating the influence of brain and skull density variations on head injury severity. Heliyon 2024; 10: e29427. https://doi.org/10.1016/j.heliyon.2024.e29427

82.

Labuda

Newman

Hoffmeister

, et al. Two-dimensional mapping of the ultrasonic attenuation and speed of sound in brain. Ultrasonics 2022; 124: 106742. https://doi.org/10.1016/j.ultras.2022.106742

83.

Libicher

Troger

US measurement of the subarachnoid space in infants: normal values. Radiology 1992; 184: 749–751.

84.

ISO. ISO/IEC 17025: general requirements for the competence of testing and calibration laboratories, https://www.iso.org (2017, accessed 5 January 2026).

85.

Preston

RC.

Output measurements for medical ultrasound. Cham: Springer-Verlag, 1991.

86.

Robinson

Wang

Cheong

, et al. Acoustic characterisation of unexploded ordnance disposal in the North Sea using high order detonations. Mar Pollut Bull 2022; 184: 114178. https://doi.org/10.1016/j.marpolbul.2022.114178

87.

Ferdousi

Comparison of frequency and impact magnitude of heading in 1966 and 2018 international professional men’s soccer matches. Ottawa, University of Ottawa, 2021.

88.

The

, Premier League. Professional football heading in training guidance. The FA, Premier League, 2021.

89.

FIFA. FIFA quality programme for footballs (outdoor, futsal and beach soccer footballs). Testing Manual. FIFA Quality Programme, https://inside.fifa.com/innovation/standards/footballs (2018, accessed 10 June 2025).

90.

Wolf

Bebarta

Bonnett

, et al. Blast injuries. Lancet 2009; 374: 405–415.

91.

Maher

RC.

Acoustical characterization of gunshots. In: 2007 IEEE workshop on signal processing applications for public security and forensics, Washington, DC, USA, 11–13 April 2007. New York, NY: IEEE.

92.

Säljö

Svensson

Mayorga

, et al. Low levels of blast raises intracranial pressure and impairs cognitive function in rats. J Neurotrauma 2009; 27(2): 383–389.

93.

Hicks

Requirements for Managing Brain Health Risks from Blast Overpressure. Washington, https://denix.osd.mil/auth/soh/programs/bop/ (2024, accessed 8 August 2024).

94.

Flamme

Wong

Liebe

, et al. Estimates of auditory risk from outdoor impulse noise II: civilian firearms. Noise Health 2009; 11: 231–242.

95.

Paddan

Howell

MJ.

A comparison of noise produced by blank and live rounds fired from military rifles. Appl Acoust 2022; 189: 108599. https://doi.org/10.1016/j.apacoust.2021.108599

96.

Eaton

Coroner cites football as reason for brain injury. BMJ 2002; 375: 1133.

97.

Song

Fleysher

, et al. Orbitofrontal gray-white interface injury and the association of soccer heading with verbal learning. JAMA Netw Open 2025; 8: e2532461.

98.

Doherty

O’Keefe

Wallace

, et al. Blood-brain barrier dysfunction as a hallmark pathology in chronic traumatic encephalopathy. J Neuropathol Exp Neurol 2016; 75: 656–662.

99.

Manning

First former MLS player diagnosed with CTE. Concussion Legacy Foundation, https://concussionfoundation.org/news/press-release/scott-vermillion-mls-cte (2022, accessed 24 August 2023).

100.

BBC Sport. Nobby stiles: ‘no doubt’ heading caused ex-footballer’s. Football, 2020, pp. 1–7.

101.

Mills

KL.

Understanding the mechanisms that govern the impact behaviour of a football. Sheffield Hallam University, https://shura.shu.ac.uk/35661/ (2024, acces-sed 9 June 2025).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.34 MB

0.00 MB