JaundiceBOT: An Explainable Offline Artificial Intelligent Chatbot for Early Risk Stratification of Neonatal Jaundice in Rural Areas
DOI:
https://doi.org/10.54536/ajsts.v5i2.8559Keywords:
Artificial Intelligence, Chatbot, Jaundice, Neonatal, Natural Language ProcessingAbstract
A custom chatbot; JaundiceBOT was developed to provide accessible, speedy and intelligent medical diagnosis of Jaundice for neonates. Jaundice is a condition of hyperbilirubinemia affecting 60% of term and 80% of preterm infants. The research aimed at the development of an educational and informational tool as a clinical support for neonatal caregivers. The system was developed to engage users in a conversation to understand symptoms, then make a diagnosis. It employed a specialized Hybrid Natural Language Processing (NLP) system implemented through client-side JavaScript (ES6+), the core functionality depends on a robust, rule-based artificial intelligence technique and a structured knowledge base created from processing 214 real hospital cases. The architecture is free of any form of application programming interface (API-free), which makes it function 100% offline and minimizes operational cost. The user begins to interact using detail questionnaire to obtain essential health information, such as the age and the specific symptoms. This data enables the chatbot to perform risk stratification and offer personalized health service and evidence-based emergency protocols. The frontend of the application was developed by employing HTML, CSS, JavaScript, and a component-based design to ensure a responsive and user-friendly interface. Evaluation shows an accuracy of 94.7% for the NLP system, with a 2.3 sec. start-up time and 0.8 sec response time. The study has presented a portable infant health assistance which could be accessed in rural communities, thereby bringing neonatal health care close to everyone.
Downloads
References
Abiha, U., Banerjee, D. S., & Mandal, S. (2023). Demystifying non-invasive approaches for screening jaundice in low resource settings: A review. Frontiers in Pediatrics, 11, 1292678.
Althnian, A., Almanea, N., & Aloboud, N. (2021). Neonatal jaundice diagnosis using a smartphone camera based on eye, skin, and fused features with transfer learning. Sensors, 21(21), 7038.
Chao W., Liang H., Weiwei H., Weiling L., Xinfu J., Zebin H. & Qingping D (2021): Research and design of neonatal jaundice detector based on colour sensor, DMIP, Kyoto Japan ’21(4), 12-15.
Choo, S., Yoo, S., Endo, K., Truong, B., & Son, M. H. (2025). Advancing clinical chatbot validation using AI-powered evaluation with a new 3-bot evaluation system: Instrument validation study. JMIR Nursing, 8(1), e63058.
Eliazer M., Amaran S., Sreekumar K., Vikram A., Gyanendra P. J. & Cho W. (2025): Integrating vision transformer-based deep learning model with kernel extreme learning machine for non-invasive diagnosis of neonatal jaundice using biomedical images. Scientific Reports (2025) 15:25493 https://doi.org/10.1038/s41598-025-08342-2
Hazarika, C. J., Borah, A., Gogoi, P., Ramchiary, S. S., Daurai, B., Gogoi, M., & Saikia, M. J. (2024). Development of non-invasive biosensors for neonatal jaundice detection: A review. Biosensors, 14(5), 254.
Hernandez-Peña, P., Poullier, J. P., Van Mosseveld, C. J. M., Van de Maele, N., Cherilova, V.,
Indikadahena, C., & Zoidze, V. (2013). Health worker remuneration in WHO Member States. Bulletin of the World Health Organization, 91(11), 808–815.
Moosa, A. S., Ngeow, A. J. H., Yang, Y., Poon, Z., Ng, D. X., Yi Ling, E. K., & Tan, N. C. (2023). A novel smartphone app for self-monitoring of neonatal jaundice among postpartum mothers: Qualitative research study. JMIR mHealth and uHealth, 11(12), e53291.
Montenegro, J. L. Z., da Costa, C. A., & Da Rosa Righi, R. (2019). Survey of conversational agents in health. Expert Systems with Applications, 129, 56–67.
Ngeow, A. J. H., Tan, M. G., Dong, X., Hong, H. X., Lim, Y., Soh, Z. Y. & Tan, N. C. (2023). Validation of a smartphone-based screening tool (Biliscan) for neonatal jaundice in a multi-ethnic neonatal population. Journal of Paediatrics and Child Health, 59(2), 288–297.
Ngeow, A. J. H., Moosa, A. S., Tan, M. G., Zou, Z., Goh, M. M. R., Lim, G. H., & Tan, N. C.(2024). Development and validation of a smartphone app for neonatal jaundice screening. JAMA Network Open, 7(12), e2450260.
Odeyemi, C. S. ., Ubochi, B. C. ., Olowu, A. J. ., & Apena, W. O. . (2026). Artificial Intelligence-Based Computer-Aided Diagnosis of Neonatal Sepsis: CurrentAdvances, Challenges and Clinical Opportunities in Nigeria. American Journal of Data Science and Artificial Intelligence, 2(2), 46-53. https://doi.org/10.54536/ajdsai.v2i2.7798.
Odeyemi C. S, Olaniyan O. M., Sobowale A. A and Samuel I B. (2024): Artificial Intelligence (AI) Based Techniques for Reducing Neonatal Mortality in Nigeria: A Descriptive Review. UNIOSUN Journal of Engineering and Environmental Science (UNIOSUNJEET) 6(2) 86-94.
Odeyemi C S, Olaniyan O M, Omodunbi B., Samuel I B, Soladoye A & Olawanle David (2026): Hybrid Artificial Intelligence System for Reducing Neonatal Mortality in Nigeria, International Journal of Medical Informatics 206(2026) 106162. doi.org/10.1016/j.ijmedinf.2025.106162.
Okwundu, C. I., Olowoyeye, A., Uthman, O. A., Smith, J., Wiysonge, C. S., Bhutani, V. K., Fiander, M., & Gautham, K. S. (2023). Transcutaneous bilirubinometry versus total serum bilirubin measurement for newborns. Cochrane Database of Systematic Reviews, Issue 5, Art. No.: CD012660.
Olorunmo F. R., Afolalu S. A., Jesumileye M. R., Akpor O. A. & Ogunkorode A. O. (2025): AI-Driven mobile health solutions for educating mothers on neonatal jaundice, Journal of Science and Technology Research, 7(2), 3842-3848, doi.org/10.37933/nipes/7.4.2025.S1476.
Salami, F. O., Muzammel, M., Mourchid, Y., & Othmani, A. (2025). Artificial Intelligence non-invasive methods for neonatal jaundice detection: A review. Artificial Intelligence In Medicine, 162, 103088.
Samiee-Zafarghandy, S., Dmochowski, K., & Macdonald, L. K. (2014). The effect of skin color on the performance of the BiliChek transcutaneous bilirubinometer. Journal of Perinatology, 34(8), 613–617.
Yan, Q., Gong, Y., Luo, Q., Yin, X., Yang, L., Wang, H., Feng, J., Xing, K., Huang, Y., Huang, C., & Fan, L. (2022). Effects of a smartphone-based out-of-hospital screening app for neonatal hyperbilirubinemia on neonatal readmission rates and maternal anxiety: Randomized controlled trial. Journal of Medical Internet Research, 24(11), e37843.
Zhang, M., Tang, J., He, Y., Li, W., Chen, Z., Xiong, T., Qu, Y., Li, Y., & Mu, D. (2021). Systematic review of global clinical practice guidelines for neonatal hyperbilirubinemia. BMJ Open, 11(1), e040182.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Odeyemi C.S., Olaniyan O. M.
This work is licensed under a Creative Commons Attribution 4.0 International License.
