Evaluation of an NLP-Based Chatbot for Informational Support in Bronchopulmonary Dysplasia (BPD) in neonates
Abstract
Bronchopulmonary Dysplasia (BPD) is a chronic lung condition that affects neonates, especially those born prematurely. Despite its relevance, the lack of accessible and up-to-date information hampers the proper management of this condition for both families and healthcare professionals. This study proposes the development of a chatbot based on Natural Language Processing (NLP) technique, named Soprinho IA, to provide informational and empathetic support to both mothers and healthcare professionals, focusing on BPD information. The methodology involves data collection from various sources, text preprocessing, and training of three different NLP architectures, such as Recurrent Neural Networks (RNNs), transformers, and other state-of-the-art approaches. The Sentence-Transformer model, based on BERT, achieved the best results, with high accuracy and semantic relevance. The chatbot is integrated into a responsive web interface, offering multimodal interaction features, including audio support. The evaluation of the models demonstrated the chatbot’s effectiveness in providing contextualized and easily understandable responses, with potential to improve communication and the management of neonates with BPD. The study also highlights limitations, such as the need to improve speech-to-text conversion, and suggests future directions for the integration of AI models to support clinical decision-making.
References
Amil, S., Da, S.-M.-A.-R., Plaisimond, J., Roch, G., Sasseville, M., Bergeron, F., and Gagnon, M.-P. (2025). Interactive conversational agents for perinatal health: A mixed methods systematic review. Healthcare, 13(4).
Balest, A. L. (2023a). Displasia broncopulmonar (dbp) — versão para profissionais de saúde. Last Accessed: July 2025.
Balest, A. L. (2023b). Displasia broncopulmonar (dbp) — versão para profissionais o paciente. Last Accessed: July 2025.
Bancalari, E., Claure, N., and Sosenko, I. R. (2003). Bronchopulmonary dysplasia: changes in pathogenesis, epidemiology and definition. In Seminars in neonatology, volume 8, pages 63–71. Elsevier.
Banerjee, S. and Lavie, A. (2005). Meteor: An automatic metric for mt evaluation with improved correlation with human judgments. In Proceedings of the acl workshop on intrinsic and extrinsic evaluation measures for machine translation and/or summarization, pages 65–72.
Leal, S. E. and Aluísio, S. M. (2024). Complexidade textual e suas tarefas relacionadas. Processamento de linguagem natural: conceitos, técnicas e aplicações em português.
Leitner, K., Cutri-French, C., Mandel, A., Christ, L., Koelper, N., McCabe, M., Seltzer, E., Scalise, L., Colbert, J. A., Dokras, A., Rosin, R., and Levine, L. (2025). A conversational agent using natural language processing for postpartum care for new mothers: Development and engagement analysis. JMIR AI, 4:e58454.
Lipton, Z. C., Berkowitz, J., and Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019.
Nenkova, A. (2005). Automatic text summarization of newswire: Lessons learned from the document understanding conference. In AAAI, volume 5, pages 1436–1441.
Nguyen, Q. C., Aparicio, E. M., Jasczynski, M., Channell Doig, A., Yue, X., Mane, H., Srikanth, N., Gutierrez, F. X. M., Delcid, N., He, X., and Boyd-Graber, J. (2024). Rosie, a health education question-and-answer chatbot for new mothers: Randomized pilot study. JMIR Form Res, 8:e51361.
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., and Liu, P. J. (2020). Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of machine learning research, 21(140):1–67.
Reimers, N. and Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084.
Rivera Rivera, J. N., AuBuchon, K. E., Smith, M., Starling, C., Ganacias, K. G., Danielson, A., Patchen, L., Rethy, J. A., Blumenthal, H. J., Thomas, A. D., and Arem, H. (2024). Development and refinement of a chatbot for birthing individuals and newborn caregivers: Mixed methods study. JMIR Pediatr Parent, 7:e56807.
Siffel, C., Kistler, K. D., Lewis, J. F. M., and and, S. P. S. (2021). Global incidence of bronchopulmonary dysplasia among extremely preterm infants: a systematic literature review. The Journal of Maternal-Fetal & Neonatal Medicine, 34(11):1721–1731. PMID: 31397199.
Stanford Children’s Health. Bronchopulmonary dysplasia program. [link]. Last accessed: June 2025.
Stoll, B. J., Hansen, N. I., Bell, E. F., Shankaran, S., Laptook, A. R., Walsh, M. C., Hale, E. C., Newman, N. S., Schibler, K., Carlo, W. A., et al. (2010). Neonatal outcomes of extremely preterm infants from the nichd neonatal research network. Pediatrics, 126(3):443–456.
Thébaud, B., Goss, K. N., Laughon, M., Whitsett, J. A., Abman, S. H., Steinhorn, R. H., Aschner, J. L., Davis, P. G., McGrath-Morrow, S. A., Soll, R. F., et al. (2019). Bronchopulmonary dysplasia. Nature reviews Disease primers, 5(1):78.
(WHO), W. H. O. (2023). Born too soon: decade of action on preterm birth. Available at: [link].
Zhou, L., Bao, J., Setiawan, I. M. A., Saptono, A., Parmanto, B., et al. (2019). The mhealth app usability questionnaire (mauq): development and validation study. JMIR mHealth and uHealth, 7(4):e11500.
