Reliability of Bilateral Thermal Asymmetry During Shortened Acclimatization

Executive summary. Daily thermal screening has an operational constraint: established preparation guidance asks athletes to wait at least 15 minutes before imaging. This can make repeatable pre-training screening difficult in time-limited club settings. This study did not test whether shortened preparation replaces laboratory thermography. It tested a narrower operational question: whether the group-level bilateral asymmetry signal (ΔT) shows systematic change before the 16-minute reference acquisition.

Key interpretation. At 8 and 10 minutes, group mean ΔT showed no significant difference from 16 minutes and mean bias was near zero. Individual agreement remained moderate, so the finding supports standardized operational screening, not replacement of laboratory acclimation guidance.

Study at a glance

1. The operational problem

Infrared thermography is used in sport to describe superficial thermal distribution and bilateral asymmetry. It is non-invasive and repeatable, but its practical value depends on whether clubs can fit standardized preparation into daily schedules.

ATS internal imaging guidance specifies 15–20 minutes of room acclimation; the TISEM sports-thermography consensus recommends a minimum 15-minute acclimation period. Those recommendations concern stabilization of skin temperature under controlled measurement conditions.

Clinical interpretation often relies on the difference between corresponding left and right regions, ΔT = |T_left − T_right|. Absolute skin temperature and bilateral asymmetry are related but distinct endpoints. This paper asks whether ΔT can provide a stable operational signal earlier than absolute temperature fully equilibrates.

2. Objective and methods

The objective was to quantify the stability of bilateral thermal asymmetry from baseline to 16 minutes, and to characterize 8- and 10-minute acquisitions against the 16-minute reference for operational screening.

Thirty-one physically active male athletes completed nine consecutive thermal acquisitions at two-minute intervals. The protocol included 32 bilateral anatomical regions across anterior and posterior views. Athletes were prepared in a controlled room (24 ± 1 °C; relative humidity 40–60%), wearing athletic shorts and no footwear. Imaging used a handheld FLIR ONE Edge system.

The primary endpoint was ΔT. Group change over time was tested with repeated-measures ANOVA, Greenhouse–Geisser correction, and a Friedman sensitivity analysis. The 8- and 10-minute acquisitions were compared with 16 minutes using ICC(2,1), Bland–Altman bias and 95% limits of agreement (LoA), SEM, and MDC95. Missing source measurements were retained as missing; paired analyses use available 8'/16' or 10'/16' observations.

3. Results

Group-level asymmetry did not shift over time. Across 917 complete athlete-region trajectories, the repeated-measures model did not detect a time effect (F(8, 7328) = 1.06, p = .392; Greenhouse–Geisser p = .390). The non-parametric sensitivity analysis was consistent (Friedman χ² = 10.04, p = .262). These results describe stability of the group mean; they do not imply identical readings for every individual athlete or region.

Bias was near zero at both operational timepoints and neither comparison showed a significant mean difference from 16 minutes. The LoA and moderate ICC values, however, demonstrate individual-level spread: the reason this paper frames the protocol as screening support rather than full methodological substitution.

Reliability differs across regions. The strongest 8'/16' reproducibility was observed in Groin (ICC .72), Lower Abdomen (.71), Rear Upper Arm (.70), and Ankle (back) (.69). No region reaches the ≥ .90 laboratory-validation target. This regional variation reinforces the need for standardized capture, longitudinal interpretation, and cautious individual decision-making.

4. What the findings mean

Two observations can be true at once. First, there is no evidence that mean bilateral asymmetry shifts systematically across the measured acclimation window. Second, individual agreement with the 16-minute reference is only moderate. ANOVA answers the first question; ICC and Bland–Altman analysis answer the second.

This evidence supports an 8-minute, standardized operational screening workflow when clubs cannot practically wait for a full laboratory preparation period. It does not demonstrate equivalence to a 15–20 minute protocol for absolute-temperature assessment, diagnostic use, or clinical decision-making.

5. Practical implementation for clubs

The operational benefit is not a claim that clubs should abandon established preparation standards. It is a practical option for a defined use case: frequent, standardized screening in which teams review bilateral patterns over time and escalate atypical findings rather than make diagnoses from a single thermal image. Against the 16-minute study reference, an 8-minute screening acquisition means 50% less waiting before capture.

6. Strengths and limitations

Strengths

• Repeated within-session measurements at nine timepoints
• Thirty-one athletes and up to 32 bilateral regions
• Complementary group-level, agreement, and regional analyses
• A question anchored in real club workflow constraints

Limitations

• Male athletes only and a moderate sample size
• One controlled session; no inter-day reliability
• A single environmental context and device workflow
• ICC values indicate moderate, not high, individual-level reproducibility
• Findings should not be generalized to diagnosis, pathology detection, or unrestricted field conditions

7. Conclusion

Bilateral thermal asymmetry showed no significant group-level temporal shift from baseline to 16 minutes. At 8 and 10 minutes, comparison with the 16-minute acquisition produced near-zero mean bias and no significant paired difference.

The appropriate operational interpretation is deliberately narrow: under standardized conditions, an 8-minute acclimatization period can provide a useful bilateral-asymmetry signal for time-constrained club screening. It is not evidence that shortened preparation replaces established laboratory thermography guidance, nor that one scan should drive a clinical conclusion.

References

1. Moreira DG, Costello JT, Brito CJ, et al. Thermographic imaging in sports and exercise medicine: A Delphi study and consensus statement on the measurement of human skin temperature. Journal of Thermal Biology. 2017;69:155–162. doi:10.1016/j.jtherbio.2017.07.006
2. Athlete Thermography Screening. Imaging Guidelines: Thermographic screening — preparation and capture protocol. Internal protocol, 2026.
3. ATS validation dataset: 31 athletes, 0–16 minute acclimatization protocol; internal analysis workbook, 2026.

For screening only. Not a medical diagnosis.