A Lightweight Deep Learning Approach for Autonomous Detection of Risky Infant Sleep Postures
Resumo
Sudden Infant Death Syndrome (SIDS) and sleep-related fatalities remain a leading cause of post-neonatal mortality, often linked to hazardous sleep positions such as prone positioning and airway obstruction. While continuous monitoring is essential for prevention, existing wearable sensors pose risks of skin irritation and displacement, while traditional video monitors depend on fallible human vigilance. This study addresses these limitations by developing a high-performance, non-invasive vision-based system for the autonomous detection of risky infant postures. Utilizing the YOLOv8n architecture, we implemented a lightweight framework optimized for real-time classification of “Safe” and “Danger” positions. Our findings demonstrate high detection reliability, with the model achieving a recall of 95.11% and a precision of 96.07% for hazardous postures, ensuring a critical reduction in false negatives while maintaining caregiver trust. To mitigate the “black-box” nature of deep learning, we applied Gradient-weighted Class Activation Mapping (Grad-CAM). Qualitative analysis confirmed that the model’s predictive logic is grounded in clinically relevant anatomical features, such as the infant’s head and torso, rather than environmental noise. These results establish a robust computational foundation for future privacy-preserving, localized monitoring solutions that can provide immediate, life-saving alerts in home environments.
Palavras-chave:
Infant Posture Detection, Computer Vision, YOLOv8, Risk Detection, SIDS, Safe Sleep
Referências
Abdelfattah, M., Zhou, L., Sum-Ping, O., Hekmat, A., Galati, J., Gupta, N., Adaimi, G., Marwaha, S., Parekh, A., Mignot, E., et al. (2025). Automated detection of isolated rem sleep behavior disorder using computer vision. Annals of Neurology, 97(5):860–872.
Alam, H., Burhan, M., Gillani, A., Haq, I. U., Arshed, M. A., Shafi, M., and Ahmad, S. (2023). Iot based smart baby monitoring system with emotion recognition using machine learning. Wireless Communications and Mobile Computing, 2023(1):1175450.
Bharati, V. (2021). An efficient edge deep learning computer vision system to prevent sudden infant death syndrome. In 2021 IEEE International Conference on Smart Computing (SMARTCOMP), pages 286–291. IEEE.
Cay, G., Solanki, D., Al Rumon, M. A., Ravichandran, V., Hoffman, L., Laptook, A., Padbury, J., Salisbury, A. L., and Mankodiya, K. (2022). Neowear: An iot-connected e-textile wearable for neonatal medical monitoring. Pervasive and Mobile Computing, 86:101679.
Chandnani, K., Tripathy, S., Parbhakar, A. K., Takiar, K., Singhal, U., Sasikumar, P., and Maheswari, S. (2025). A novel smart baby cradle system utilizing iot sensors and machine learning for optimized parental care. Scientific Reports, 15(1):19080.
Ge, Z., Liu, S., Wang, F., Li, Z., and Sun, J. (2021). Yolox: Exceeding yolo series in 2021. arXiv preprint arXiv:2107.08430.
Huang, X. and Huang, M. (2024). Classification and recognition of infant sleeping positions based on dual model feature fusion. In Proc. IEEE Int. Conf. on Artificial Intelligence, Automation and High Performance Computing (AIAHPC), pages 57–63.
Izulla, P., Muriuki, A., Kiragu, M., Yahner, M., Fonner, V., Nitu, S. N. A., Osir, B., Bello, F., and de Graft-Johnson, J. (2023). Proximate and distant determinants of maternal and neonatal mortality in the postnatal period: A scoping review of data from low-and middle-income countries. PLoS One, 18(11):293–479.
Li, C., Ren, G., and Wang, Z. (2025). Sleep posture recognition method based on sparse body pressure features. Applied Sciences, 15(9).
Maugeri, A., Barchitta, M., Schillaci, G., and Agodi, A. (2025). Spatial patterns and temporal trends in stillbirth, neonatal, and infant mortality: an exploration of country-level data from 2000 to 2021. Journal of Global Health, 15:04034.
Moon, R. Y., Darnall, R. A., Feldman-Winter, L., Goodstein, M. H., and Hauck, F. R. (2016). SIDS and other sleep-related infant deaths: Updated 2016 recommendations for a safe infant sleeping environment. Pediatrics, 138(5):e20162938.
Ogbo, F. A., Ezeh, O. K., Awosemo, A. O., Ifegwu, I. K., Tan, L., Jessa, E., Charwe, D., and Agho, K. E. (2019). Determinants of trends in neonatal, post-neonatal, infant, child and under-five mortalities in tanzania from 2004 to 2016. BMC public health, 19(1):1243.
Pereira, A. G. (2025). Uma abordagem leve para detecção de embarcações utilizando redes neurais sem pesos e regressão. Dissertação de mestrado, Universidade Federal do Rio de Janeiro (UFRJ), COPPE, Programa de Engenharia de Sistemas e Computação, Rio de Janeiro, RJ, Brasil. Acesso em: 7 maio 2026.
Tan, J. H. and Goh, C. P. (2024). Enhancing child safety: Computer vision-based accident detection for infants and toddlers. In 2024 3rd International Conference on Digital Transformation and Applications (ICDXA), pages 1–5. IEEE.
Udoko, A. N., Dyess, N. F., Hwang, S. S., and Fisher, C. R. (2026). Sudden unexpected infant death and safe sleep practices. NeoReviews, 27(1):e1–e10.
Wu, L., Guo, S., Han, L., and Jia, J. (2025). Human sleep motion recognition based on multi-sensor measurement data fusion, measurement. Measurement, 253, Parte C(1):117746.
Yamada, N. K., Szyld, E., Strand, M. L., Finan, E., Illuzzi, J. L., Kamath-Rayne, B. D., Kapadia, V. S., Niermeyer, S., Schmölzer, G. M., Williams, A., et al. (2024). 2023 american heart association and american academy of pediatrics focused update on neonatal resuscitation: an update to the american heart association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation, 149(1):157–166.
Yun, I., Jeung, J., Kim, M., Kim, Y.-S., and Chung, Y. (2020). Ultra-low power wearable infant sleep position sensor. Sensors, 20(1).
Alam, H., Burhan, M., Gillani, A., Haq, I. U., Arshed, M. A., Shafi, M., and Ahmad, S. (2023). Iot based smart baby monitoring system with emotion recognition using machine learning. Wireless Communications and Mobile Computing, 2023(1):1175450.
Bharati, V. (2021). An efficient edge deep learning computer vision system to prevent sudden infant death syndrome. In 2021 IEEE International Conference on Smart Computing (SMARTCOMP), pages 286–291. IEEE.
Cay, G., Solanki, D., Al Rumon, M. A., Ravichandran, V., Hoffman, L., Laptook, A., Padbury, J., Salisbury, A. L., and Mankodiya, K. (2022). Neowear: An iot-connected e-textile wearable for neonatal medical monitoring. Pervasive and Mobile Computing, 86:101679.
Chandnani, K., Tripathy, S., Parbhakar, A. K., Takiar, K., Singhal, U., Sasikumar, P., and Maheswari, S. (2025). A novel smart baby cradle system utilizing iot sensors and machine learning for optimized parental care. Scientific Reports, 15(1):19080.
Ge, Z., Liu, S., Wang, F., Li, Z., and Sun, J. (2021). Yolox: Exceeding yolo series in 2021. arXiv preprint arXiv:2107.08430.
Huang, X. and Huang, M. (2024). Classification and recognition of infant sleeping positions based on dual model feature fusion. In Proc. IEEE Int. Conf. on Artificial Intelligence, Automation and High Performance Computing (AIAHPC), pages 57–63.
Izulla, P., Muriuki, A., Kiragu, M., Yahner, M., Fonner, V., Nitu, S. N. A., Osir, B., Bello, F., and de Graft-Johnson, J. (2023). Proximate and distant determinants of maternal and neonatal mortality in the postnatal period: A scoping review of data from low-and middle-income countries. PLoS One, 18(11):293–479.
Li, C., Ren, G., and Wang, Z. (2025). Sleep posture recognition method based on sparse body pressure features. Applied Sciences, 15(9).
Maugeri, A., Barchitta, M., Schillaci, G., and Agodi, A. (2025). Spatial patterns and temporal trends in stillbirth, neonatal, and infant mortality: an exploration of country-level data from 2000 to 2021. Journal of Global Health, 15:04034.
Moon, R. Y., Darnall, R. A., Feldman-Winter, L., Goodstein, M. H., and Hauck, F. R. (2016). SIDS and other sleep-related infant deaths: Updated 2016 recommendations for a safe infant sleeping environment. Pediatrics, 138(5):e20162938.
Ogbo, F. A., Ezeh, O. K., Awosemo, A. O., Ifegwu, I. K., Tan, L., Jessa, E., Charwe, D., and Agho, K. E. (2019). Determinants of trends in neonatal, post-neonatal, infant, child and under-five mortalities in tanzania from 2004 to 2016. BMC public health, 19(1):1243.
Pereira, A. G. (2025). Uma abordagem leve para detecção de embarcações utilizando redes neurais sem pesos e regressão. Dissertação de mestrado, Universidade Federal do Rio de Janeiro (UFRJ), COPPE, Programa de Engenharia de Sistemas e Computação, Rio de Janeiro, RJ, Brasil. Acesso em: 7 maio 2026.
Tan, J. H. and Goh, C. P. (2024). Enhancing child safety: Computer vision-based accident detection for infants and toddlers. In 2024 3rd International Conference on Digital Transformation and Applications (ICDXA), pages 1–5. IEEE.
Udoko, A. N., Dyess, N. F., Hwang, S. S., and Fisher, C. R. (2026). Sudden unexpected infant death and safe sleep practices. NeoReviews, 27(1):e1–e10.
Wu, L., Guo, S., Han, L., and Jia, J. (2025). Human sleep motion recognition based on multi-sensor measurement data fusion, measurement. Measurement, 253, Parte C(1):117746.
Yamada, N. K., Szyld, E., Strand, M. L., Finan, E., Illuzzi, J. L., Kamath-Rayne, B. D., Kapadia, V. S., Niermeyer, S., Schmölzer, G. M., Williams, A., et al. (2024). 2023 american heart association and american academy of pediatrics focused update on neonatal resuscitation: an update to the american heart association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation, 149(1):157–166.
Yun, I., Jeung, J., Kim, M., Kim, Y.-S., and Chung, Y. (2020). Ultra-low power wearable infant sleep position sensor. Sensors, 20(1).
Publicado
19/07/2026
Como Citar
PIROTELLO, Gabriel L.; AKABANE, Ademar T.; GONZÁLEZ, Diana C..
A Lightweight Deep Learning Approach for Autonomous Detection of Risky Infant Sleep Postures. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO UBÍQUA E PERVASIVA (SBCUP), 18. , 2026, Gramado/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1-11.
ISSN 2595-6183.
DOI: https://doi.org/10.5753/sbcup.2026.22577.
