Ontology-Integrated Natural Language Processing for Symptom Extraction and Disease Severity Staging from Clinical Notes: A Systematic Review
Abstract
African health systems are starting to digitise clinical records at a speed much faster than they can interpret the information. Even though most of the symptom evidence for disease staging is present in unstructured clinical narratives in the EMR, such evidence is often not accessible to automated decision support. Natural Language Processing (NLP) has advanced rapidly in clinical entity extraction, but the contribution of formal clinical ontologies to extraction and, crucially, to severity staging has not been systematically assessed. This review collates peer-reviewed evidence on ontology-integrated NLP for symptom extraction and severity classification. Six databases were searched using terms combining natural language processing, symptom extraction and severity staging (PubMed, Scopus, IEEE Xplore, ACM Digital Library, Web of Science, Google Scholar) for the period from January 2019 to June 2025. After applying pre-specified eligibility criteria, the records were screened and forty-one studies were included in the narrative synthesis, in accordance with PRISMA 2020 guidelines. Studies were characterised by clinical domain, corpus, ontology, integration strategy, architecture and evaluation design, and appraised using an adapted QUADAS-2 instrument. Three integration approaches were identified: dictionary and concept mapping, ontology-supervised distant labelling, and ontology-constrained neural architectures. Ontology integration consistently improved concept normalisation, cross-corpus transfer and annotation efficiency, with the largest gains in low-annotation settings. Only nine studies addressed severity staging; these were largely confined to cardiology and respiratory medicine and were seldom validated against clinical grading instruments. No included study used an African clinical corpus. The evidence base is therefore methodologically well developed for extraction, weak for staging and geographically narrow. For automated severity staging to support clinical decision-making at district-hospital level, African health systems need infrastructure as much as computation: de-identified African clinical corpora with local extensions to the terminology, ontology-guided staging validated against recognised grading instruments, and transparent reporting.
References
Adejumo, P., Thangaraj, P. M., Dhingra, L. S., Aminorroaya, A., Zhou, X., Brandt, C., & Khera, R. (2024). Natural language processing of clinical documentation to assess functional status in patients with heart failure. JAMA Network Open, 7(11), e2443925. https://doi.org/10.1001/jamanetworkopen.2024.43925
Ahmad, P. N., Shah, A. M., & Lee, K. (2023). A review on electronic health record text-mining for biomedical name entity recognition in healthcare domain. Healthcare, 11(9), 1268. https://doi.org/10.3390/healthcare11091268
Alahmar, A., AlMousa, M., & Benlamri, R. (2022). Automated clinical pathway standardization using SNOMED CT-based semantic relatedness. Digital Health, 8, 20552076221089796. https://doi.org/10.1177/20552076221089796
Arbabi, A., Adams, D. R., Fidler, S., & Brudno, M. (2019). Identifying clinical terms in medical text using ontology-guided machine learning. JMIR Medical Informatics, 7(2), e12596. https://doi.org/10.2196/12596
Ariyanto, A. D. P., Purwitasari, D., & Fatichah, C. (2024). A systematic review on semantic role labeling for information extraction in low-resource data. IEEE Access, 12, 57917-57946. https://doi.org/10.1109/ACCESS.2024.3392370
Bai, L., Mulvenna, M. D., Wang, Z., & Bond, R. (2021). Clinical entity extraction: Comparison between MetaMap, cTAKES, CLAMP and Amazon Comprehend Medical. In 2021 32nd Irish Signals and Systems Conference (pp. 1-6). IEEE. https://doi.org/10.1109/ISSC52156.2021.9467856
Berge, G. T., Granmo, O. C., Tveit, T. O., Ruthjersen, A. L., & Sharma, J. (2023). Combining unsupervised, supervised and rule-based learning: The case of detecting patient allergies in electronic health records. BMC Medical Informatics and Decision Making, 23(1), 188. https://doi.org/10.1186/s12911-023-02271-8
Bose, P., Srinivasan, S., Sleeman IV, W. C., Palta, J., Kapoor, R., & Ghosh, P. (2021). A survey on recent named entity recognition and relationship extraction techniques on clinical texts. Applied Sciences, 11(18), 8319. https://doi.org/10.3390/app11188319
Campillos-Llanos, L., Valverde-Mateos, A., & Capllonch-Carrion, A. (2025). Hybrid natural language processing tool for semantic annotation of medical texts in Spanish. BMC Bioinformatics, 26(1), 7. https://doi.org/10.1186/s12859-024-05949-6
Cazzaniga, G., Eccher, A., Munari, E., Marletta, S., Bonoldi, E., Della Mea, V., & L'Imperio, V. (2023). Natural language processing to extract SNOMED-CT codes from pathological reports. Pathologica, 115(6), 318. https://doi.org/10.32074/1591-951X-952
Chan, L., Beers, K., Yau, A. A., Chauhan, K., Duffy, A., Chaudhary, K., & Nadkarni, G. N. (2020). Natural language processing of electronic health records is superior to billing codes to identify symptom burden in haemodialysis patients. Kidney International, 97(2), 383-392. https://doi.org/10.1016/j.kint.2019.10.023
Chang, E., & Sung, S. (2024). Use of SNOMED CT in large language models: Scoping review. JMIR Medical Informatics, 12(1), e62924. https://doi.org/10.2196/62924
Dash, A., Darshana, S., Yadav, D. K., & Gupta, V. (2024). A clinical named entity recognition model using pretrained word embedding and deep neural networks. Decision Analytics Journal, 10, 100426. https://doi.org/10.1016/j.dajour.2024.100426
Dong, H., Suarez-Paniagua, V., Zhang, H., Wang, M., Casey, A., Davidson, E., & Wu, H. (2023). Ontology-driven and weakly supervised rare disease identification from clinical notes. BMC Medical Informatics and Decision Making, 23(1), 86. https://doi.org/10.1186/s12911-023-02181-9
Durango, M. C., Torres-Silva, E. A., & Orozco-Duque, A. (2023). Named entity recognition in electronic health records: A methodological review. Healthcare Informatics Research, 29(4), 286-300. https://doi.org/10.4258/hir.2023.29.4.286
Dyers, R. E., Mahomed, H., & Van Greunen, D. (2025). Design science research in quality improvement: Embedding rigour in digital health innovation. African Journal of Primary Health Care & Family Medicine, 17(2), 5194. https://doi.org/10.4102/phcfm.v17i2.5194
Gao, S., Alawad, M., Young, M. T., Gounley, J., Schaefferkoetter, N., Yoon, H. J., & Tourassi, G. (2021). Limitations of transformers on clinical text classification. IEEE Journal of Biomedical and Health Informatics, 25(9), 3596-3607. https://doi.org/10.1109/jbhi.2021.3062322
Kersloot, M. G., Lau, F., Abu-Hanna, A., Arts, D. L., & Cornet, R. (2019). Automated SNOMED CT concept and attribute relationship detection through a web-based implementation of cTAKES. Journal of Biomedical Semantics, 10(1), 14. https://doi.org/10.1186/s13326-019-0207-3
Mitha, S., Schwartz, J., Hobensack, M., Cato, K., Woo, K., Smaldone, A., & Topaz, M. (2023). Natural language processing of nursing notes: An integrative review. CIN: Computers, Informatics, Nursing, 41(6), 377-384. https://doi.org/10.1097/cin.0000000000000967
Ochoa, J. G. D., Layer, N., Mahr, J., Mustafa, F. E., Menzel, C. U., Muller, M., & Krohn, A. (2025). Optimized BERT-based NLP outperforms zero-shot methods for automated symptom detection in clinical practice. medRxiv, 2025-04. https://doi.org/10.1101/2025.04.21.25326037
Pethani, F., & Dunn, A. G. (2023). Natural language processing for clinical notes in dentistry: A systematic review. Journal of Biomedical Informatics, 138, 104282. https://doi.org/10.1016/j.jbi.2023.104282
Prakash, R., Dupre, M. E., Ostbye, T., & Xu, H. (2024). Extracting critical information from unstructured clinicians' notes data to identify dementia severity using a rule-based approach: Feasibility study. JMIR Aging, 7, e57926. https://doi.org/10.2196/57926
Quint, J. K., O'Leary, C., Venerus, A., Holmgren, U., Varghese, P., & Cabrera, C. (2021). Development and validation of a method to estimate COPD severity in multiple datasets: A retrospective study. Pulmonary Therapy, 7(1), 119-132. https://doi.org/10.1007/s41030-020-00139-0
Rohanian, O., Nouriborji, M., Jauncey, H., Kouchaki, S., Nooralahzadeh, F., & Clifton, L. (2024). Lightweight transformers for clinical natural language processing. Natural Language Engineering, 30(5), 887-914. https://doi.org/10.1017/S1351324923000542
Saigaonkar, S., & Narawade, V. (2024). Domain adaptation of transformer-based neural network model for clinical note classification in Indian healthcare. International Journal of Information Technology, 1-19. https://doi.org/10.1007/s41870-024-02053-z
Sarwar, T., Seifollahi, S., Chan, J., Zhang, X., Aksakalli, V., Hudson, I., & Cavedon, L. (2022). The secondary use of electronic health records for data mining: Data characteristics and challenges. ACM Computing Surveys, 55(2), 1-40. https://doi.org/10.1145/3490234
Scharp, D., Hobensack, M., Davoudi, A., & Topaz, M. (2024). Natural language processing applied to clinical documentation in post-acute care settings: A scoping review. Journal of the American Medical Directors Association, 25(1), 69-83. https://doi.org/10.1016/j.jamda.2023.09.006
Schulz, S., Case, J. T., Hendler, P., Karlsson, D., Lawley, M., Cornet, R., & Gao, Y. (2023). SNOMED CT and Basic Formal Ontology - convergence or contradiction between standards? The case of “clinical finding”. Applied Ontology, 18(3), 207-237. https://doi.org/10.3233/AO-230018
Slater, L. T., Bradlow, W., Ball, S., Hoehndorf, R., & Gkoutos, G. V. (2021). Improved characterisation of clinical text through ontology-based vocabulary expansion. Journal of Biomedical Semantics, 12(1), 7. https://doi.org/10.1186/s13326-021-00241-5
Solomon, M. D., Tabada, G., Allen, A., Sung, S. H., & Go, A. S. (2021). Large-scale identification of aortic stenosis and its severity using natural language processing on electronic health records. Cardiovascular Digital Health Journal, 2(3), 156-163. https://doi.org/10.1016/j.cvdhj.2021.03.003
Song, J., Hobensack, M., Bowles, K. H., McDonald, M. V., Cato, K., Rossetti, S. C., & Topaz, M. (2022). Clinical notes: An untapped opportunity for improving risk prediction for hospitalization and emergency department visit during home health care. Journal of Biomedical Informatics, 128, 104039. https://doi.org/10.1016/j.jbi.2022.104039
Steinkamp, J. M., Bala, W., Sharma, A., & Kantrowitz, J. J. (2020). Task definition, annotated dataset, and supervised natural language processing models for symptom extraction from unstructured clinical notes. Journal of Biomedical Informatics, 102, 103354. https://doi.org/10.1016/j.jbi.2019.103354
Thukral, A., Dhiman, S., Meher, R., & Bedi, P. (2023). Knowledge graph enrichment from clinical narratives using NLP, NER, and biomedical ontologies for healthcare applications. International Journal of Information Technology, 15(1), 53-65. https://doi.org/10.1007/s41870-022-01145-y
Topaz, M., Murga, L., Gaddis, K. M., McDonald, M. V., Bar-Bachar, O., Goldberg, Y., & Bowles, K. H. (2019). Mining fall-related information in clinical notes: Comparison of rule-based and novel word embedding-based machine learning approaches. Journal of Biomedical Informatics, 90, 103103. https://doi.org/10.1016/j.jbi.2019.103103
Vuokko, R., Vakkuri, A., & Palojoki, S. (2023). Systematized Nomenclature of Medicine - Clinical Terminology (SNOMED CT) clinical use cases in the context of electronic health record systems: Systematic literature review. JMIR Medical Informatics, 11, e43750. https://doi.org/10.2196/43750
Yang, X., Bian, J., Hogan, W. R., & Wu, Y. (2020a). Clinical concept extraction using transformers. Journal of the American Medical Informatics Association, 27(12), 1935-1942. https://doi.org/10.1093/jamia/ocaa189
Yang, X., PourNejatian, N., Shin, H. C., Smith, K. E., Parisien, C., Compas, C., & Wu, Y. (2022). GatorTron: A large language model for clinical natural language processing. medRxiv, 2022-02. https://doi.org/10.2196/19735
