Human core temperature prediction method integrated with iTransformer and transfer learning for high-temperature environments

doi:10.16265/j.cnki.issn1003-3033.2026.07.1118

Abstract

Abstract:

To enhance the ability to prevent and control heat stress risks among operators in high-temperature environments, a method for human core temperature prediction integrated with iTransformer with transfer learning was proposed. Large-scale simulated datasets were generated using the joint system thermoregulation model (JOS-3), which served as the foundation for pre-training the iTransformer deep learning model. Subsequently, experimental data from climate chambers were employed to fine-tune the pre-trained model with transfer learning strategies, ultimately realizing the prediction of human core temperature. The results show that the proposed method achieves a mean absolute error (MAE) of 0.103 ℃ and a root mean square error (RMSE) of 0.321 ℃ on the test set. It significantly outperforms Transformer, Reformer, Informer and Long Short-Term Memory (LSTM) methods with the identical transfer learning strategies, as well as deep learning methods trained exclusively on climate chamber experimental data, and the numerical method based on JOS-3.

Key words: iTransformer, transfer learning, high-temperature environments, human core temperature, deep learning

CLC Number:

X968高低温防护

Hu Xiaofeng, Tang Ruijun, Cong Ao, Sheng Chunlei. Human core temperature prediction method integrated with iTransformer and transfer learning for high-temperature environments[J]. China Safety Science Journal, 2026, 36(7): 251-259.

Figures/Tables 8

Fig.1

Table 1

Table 2

Fig.2

Table 3

Table 4

Fig.3

Table 5

References 25

[1]	Lynn J, Peeva N. Communications in the IPCC's Sixth Assessment Report cycle[J]. Climatic Change, 2021, 169(1/2): 1-11. doi: 10.1007/s10584-021-03220-y
[2]	田水承, 周润康, 杨杰. 高温环境下降温服对消防员热反应的影响[J]. 中国安全科学学报, 2020, 30(4):166-171. doi: 10.16265/j.cnki.issn1003-3033.2020.04.026
	Tian Shuicheng, Zhou Runkang, Yang Jie. Effects of cooling garments on firefighters' thermal responses under high temperature[J]. China Safety Science Journal, 2020, 30(4): 166-171. doi: 10.16265/j.cnki.issn1003-3033.2020.04.026
[3]	Givoni B, Goldman R F. Predicting rectal temperature response to work, environment, and clothing[J]. Journal of Applied Physiology, 1972, 32(6): 812-822. pmid: 5034345
[4]	Gagge A P. An effective temperature scale based on a simple model of human physiological regulatory response[J]. American Society of Heating, Refrigerating and Air-Conditioning Engineers Transactions, 1971, 77: 247-262.
[5]	Wissler E H. A mathematical model of the human thermal system[J]. The Bulletin of Mathematical Biophysics, 1964, 26(2): 147-166. doi: 10.1007/BF02476835
[6]	Takada S, Kobayashi H, Matsushita T. Thermal model of human body fitted with individual characteristics of body temperature regulation[J]. Building and Environment, 2009, 44(3): 463-470. doi: 10.1016/j.buildenv.2008.04.007
[7]	Du Chenqiu, Li Baizhan, Li Yongqiang, et al. Modification of the Predicted Heat Strain (PHS) model in predicting human thermal responses for Chinese workers in hot environments[J]. Building and Environment, 2019, 165: DOI:10.1016/j.buildenv.2019.106349.
[8]	Takahashi Y, Nomoto A, Yoda S, et al. Thermoregulation model JOS-3 with new open-source code[J]. Energy and Buildings, 2021, 231: DOI:10.1016/j.enbuild.2020.110575.
[9]	Niedermann R, Wyss E, Annaheim S, et al. Prediction of human core body temperature using non-invasive measurement methods[J]. International Journal of Biometeorology, 2014, 58(1): 7-15. doi: 10.1007/s00484-013-0687-2 pmid: 23760405
[10]	Buller M J, Tharion W J, Cheuvront S N, et al. Estimation of human core temperature from sequential heart rate observations[J]. Physiological Measurement, 2013, 34(7): 781-798. doi: 10.1088/0967-3334/34/7/781
[11]	De Korte J Q, Veenstra B J, Van Rijswick M, et al. A heart rate based algorithm to estimate core temperature responses in elite athletes exercising in the heat[J]. Frontiers in Sports and Active Living, 2022, 4: DOI:10.3389/FSPOR.2022.882254.
[12]	WellesA P, Xu Xiaojiang, Santee W R, et al. Estimation of core body temperature from skin temperature, heat flux, and heart rate using a Kalman filter[J]. Computers in Biology and Medicine, 2018, 99: 1-6.
[13]	Egbert J, Krenz J, Sampson P D, et al. Accuracy of an estimated core temperature algorithm for agricultural workers[J]. Archives of Environmental & Occupational Health, 2022, 77(10): 809-818.
[14]	胡啸峰, 黄玲. 基于机器学习的户外作业警务人员热应激预测模型[J]. 中国安全科学学报, 2024, 34(11): 220-228. doi: 10.16265/j.cnki.issn1003-3033.2024.11.0171
	Hu Xiaofeng, Huang Ling. Heat stress prediction model for outdoor policeman based on machine learning[J]. China Safety Science Journal, 2024, 34(11): 220-228. doi: 10.16265/j.cnki.issn1003-3033.2024.11.0171
[15]	Han Xinge, Wu Jiansong, Hu Zhuqiang, et al. Forecasting human core and skin temperatures: a long-term series approach[J]. Big Data and Cognitive Computing, 2024, 8(12): DOI:10.3390/BDCC8120197.
[16]	Laxminarayan S, Buller M J, Tharion W J, et al. Human core temperature prediction for heat-injury prevention[J]. IEEE Journal of Biomedical and Health Informatics, 2014, 19(3): 883-891.
[17]	Gribok A V, Buller M J, Hoyt R W, et al. A real-time algorithm for predicting core temperature in humans[J]. IEEE Transactions on Information Technology in Biomedicine, 2010, 14(4): 1039-1045. doi: 10.1109/TITB.2010.2043956 pmid: 20371418
[18]	Gribok A V, Buller M J, Reifman J. Individualized short-term core temperature prediction in humans using biomathematical models[J]. IEEE Transactions on Biomedical Engineering, 2008, 55(5): 1477-1487. doi: 10.1109/TBME.2007.913990 pmid: 18440893
[19]	Lu Jie, Behbood V, Hao Peng, et al. Transfer learning using computational intelligence: a survey[J]. Knowledge-Based Systems, 2015, 80: 14-23. doi: 10.1016/j.knosys.2015.01.010
[20]	Talo M, Baloglu U B, Yildirim O, et al. Application of deep transfer learning for automated brain abnormality classification using MR images[J]. Cognitive Systems Research, 2019, 54: 176-188. doi: 10.1016/j.cogsys.2018.12.007
[21]	Pan Sinno Jialin, Yang Qiang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 22(10): 1345-1359. doi: 10.1109/TKDE.2009.191
[22]	Raffel C, Shazeer N, Roberts A, et al. Exploring the limits of transfer learning with a unified text-to-text transformer[J]. Journal of Machine Learning Research, 2019:DOI:10.48550/arXiv.1910.10683.
[23]	ISO 8996-2021 Ergonomics of the thermal environment: determination of metabolic rate[S].
[24]	ISO 9886-2004 Ergonomics: evaluation of thermal strain by physiological measurements[S].
[25]	Liu Yong, Hu Tengge, Zhang Haoran, et al. iTransformer: inverted transformers are effective for time series forecasting[C]. The Twelfth International Conference on Learning Representations, 2024: DOI:10.48550/arXiv.2310.06625.

性别	身高/cm	体质量/kg	体脂率/%	年龄/岁
男	180.3±3.8	76.7±11.6	20.7±4.3	21.7±0.5
女	164.8±3.4	60.8±3.7	30.7±2.2	21.5±0.5

测量仪器	测量项目	测量范围	测量精度	最高采集频率
E-Celsius^®核心胶囊、e-viewer^®性能监视器	核心温度/℃	25~45℃	±0.1	1次/s
Ds1922L皮肤温度贴片、记录仪	皮肤温度/℃	-45~85	±0.5	1次/s
Polar心率测量带	心率/(次/min)	30~240	±1	1次/s
Wx-Sq5微型气象站	环境温度/℃	-50~80	±0.3	1次/min
	相对湿度/%	0~100	±3	1次/min
	风速/(m/s)	0~30	±0.3	1次/min

算例	MAE/℃	RMSE/℃	R²
JOS-3	0.511	0.620	-1.37
iTransformer (S₀)	0.131	0.399	0.457
iTransformer (S₁)	0.397	0.630	0.110
iTransformer (S₂)	0.103	0.321	0.649
Reformer (S₀)	0.284	0.462	0.271
Reformer (S₁)	0.291	0.549	0.366
Reformer (S₂)	0.233	0.355	0.570
Transformer (S₀)	0.271	0.465	0.264
Transformer (S₁)	0.642	0.858	0.071
Transformer (S₂)	0.264	0.445	0.324
Informer (S₀)	0.435	0.592	-0.193
Informer (S₁)	0.366	0.549	0.064
Informer (S₂)	0.270	0.397	0.463
LSTM (S₀)	0.168	0.483	0.652
LSTM (S₁)	0.225	0.542	0.710
LSTM (S₂)	0.135	0.391	0.664