Non-contact infrared thermography for core body temperature prediction faces challenges from measurement uncertainties caused by inter-device variability and environmental factors. This study develops a machine learning framework to establish the relationship between superficial thermal patterns and core temperature using empirical data from FLIR and ICI infrared imaging systems. The framework integrates multi-region facial thermographic data, device-specific metadata, and ambient environmental parameters. A feature set was developed incorporating statistical descriptors of regional temperature distributions, device-specific correction factors, environmental parameters, and spatiotemporally derived attributes. Three machine learning algorithms—Support Vector Regression, Extreme Gradient Boosting, and Random Forests—were compared for cross-device core temperature estimation. The optimized Support Vector Regression model achieved the highest predictive accuracy, with results most consistent with clinical reference measurements in both cross-device generalization and environmental robustness tests. The model demonstrated consistent performance across different device types and environmental conditions, and effectively characterized the interactive effects of device heterogeneity and environmental complexity. The integration of data-driven modeling with biothermal principles provides a framework for advancing accuracy in multi-device infrared thermometry.

Zhang Yucheng, Liu Hao, Luo Shunan. Research on a Multi-Device Infrared Body Temperature Prediction System Based on Support Vector Machine and Error Optimization. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 10: 14-22. https://doi.org/10.25236/AJCIS.2025.081003.

[1] Alessandra Pontiero, Cecilia Bulgarelli, Elena Ciuffoli, Francesca Buzzurra, Agnese Villani, Roberta Troìa, Massimo Giunti. Triage body temperature predicts outcome in cats at emergency department admission: a retrospective study of 1440 cases (January 2018 to December 2021).[J].Journal of feline medicine and surgery,2025,27(1):1098612X241291295.

[2] Rakhimov A.Ya., Manturova N.E.,Verbo E.V.. Prediction and prevention of inadvertent perioperative hypothermia in plastic surgery[J].Plastic Surgery and Aesthetic Medicine.,2025,(2):76-82.

[3] Kun Zhu, Zi Xuan Zhang, Miao Zhang. Application value of machine learning models in predicting intraoperative hypothermia in laparoscopic surgery for polytrauma patients.[J].World journal of clinical cases,2024,12(24):5513-5522.

[4] Yonghong Wang, Na Yang, Jiying Zhang, Wenyong Han, Xia Li. Construction and validation of a postoperative hypothermia prediction model in elderly patients undergoing colorectal surgery[J]. Heliyon, 2024,10(12):e32391.

[5] Emmanuel Daniel, Saada A Seif, Walter C Millanzi. Predictors of neonatal hypothermia within six hours of birth and exploring preventive practices among post-natal mothers in Kilimanjaro region: Explanatory sequential mixed method protocol.[J].PloS one,2024,19(11):e0313432.

[6] Lou Jiaqi, Fan Youfen, Cui Shengyong, Huang Neng, Jin Guoying, Chen Cui, Zhang Chun, Li Jiliang. Development and validation of a nomogram to predict hypothermia in adult burn patients during escharectomy under general anesthesia.[J].Burns : journal of the International Society for Burn Injuries, 2023,50(1):93-105.

[7] Li Leilei, Lu Yubing, Yang Li Li, Xu Wei, Yu Jing Kai. Construction and validation of postoperative hypothermia prediction model for patients undergoing joint replacement surgery[J].Journal of clinical nursing, 2023,32(13-14):3831-3839.

[8] Kubota Jun, Higurashi Norimichi, Hirano Daishi, Okabe Shiro, Yamauchi Kento, Kimura Rena, Numata Haruka, Suzuki Takayuki, Kakegawa Daisuke, Ito Akira, Hamano Shin ichiro. Body temperature predicts recurrent febrile seizures in the same febrile illness[J].Brain and Development,2021,43(7):768-774.

[9] Luo Lan, She Xichen, Cao Jiexuan, Zhang Yunlong, Li Yijiang, Song Peter X K. Detection and Prediction of Ovulation From Body Temperature Measured by an In-Ear Wearable Thermometer.[J]. IEEE transactions on bio-medical engineering,2020,67(2):512-522.

[10] Forristal Chantal, Van Aarsen Kristine, Columbus Melanie, Wei James, Vogt Kelly, Mal Sameer. Predictors of Hypothermia upon Trauma Center Arrival in Severe Trauma Patients Transported to Hospital via EMS.[J].Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors,2020,24(1):15-22.

[11] The Institute of Statistical Mathematics, Tokyo, Japan., The Institute of Statistical Mathematics, Tokyo, Japan., National Institute for Environmental Studies, Tsukuba, Japan., QOL Corporation, Nagano, Japan., The Institute of Statistical Mathematics, Tokyo, Japan.. A self-excited threshold autoregressive state-space model for menstrual cycles: Forecasting menstruation and identifying within-cycle stages based on basal body temperature.[J].Statistics in medicine,2019,38(12):2157-2170.

[12] Nomura Osamu, Ihara Takateru, Sakakibara Hiroshi, Hirokoshi Yuho, Inoue Nobuaki. Predicting serious bacterial infection in febrile young infants utilizing body temperature.[J].Pediatrics international : official journal of the Japan Pediatric Society,2019,61(5):449-452.

[13] Parlin Adam F., Schaeffer Paul J. Plastron‐mounted loggers predict terrestrial turtle body temperature better than carapace‐mounted loggers[J]. Wildlife Society Bulletin,2019,43(1):152-158.