Prediction of Infectious Diseases in LMICs Using Electronic Health Record Data

NCT07432061 | Completed

• 1 other identifier: MUTM 2024-084-01
• Study Type: observational
• Target: 1,000
• Locations: 1 country
• Sites: 1

Brief Summary

Dengue is a rapidly emerging infectious disease in South and Southeast Asia. Definitive diagnosis requires laboratory testing (PCR or antigen testing) which are often unavailable in settings with highest incidence. Correctly identifying patients who have dengue, and the small number of patients with dengue who will progress to severe disease is important to ensure prompt institution of appropriate treatments. Existing models use a combination of clinical and laboratory features. A model developed and tested on data from 397 patients admitted to the Hospital for Tropical Diseases in Bangkok in 2013 - 2014 used Bayesian modelling of variables (liver and full blood count) and clinical symptoms (including fever, petechiae, bleeding) to distinguish dengue from other febrile illness. The resultant model performed had an AUC of 0.75 which improved to 0.8 when NS1 was included. The Sequential Organ Failure (SOFA) scores, or modified versions use vital sign and blood test (liver, renal and haematology) data and are good indicators of those likely to die. However, they function less well in moderately severe diseases (e.g. predicting need for ICU admission). These approaches are promising, but are limited by limited generalizability, use of multiple blood tests and clinical symptoms. A low-cost easy tool able to rapidly diagnose dengue and predict disease severity would be of great value in the region. With modern machine learning methods, this is now feasible and previously identified barriers such as the requirement for large amounts of training data can now be overcome. For example, models can be created from large datasets, but then optimized for smaller different datasets (data either from other locations/conditions, or with less input data). We've previously shown that data-driven machine learning algorithms could generalize across multiple United Kingdom (UK) National Health Service (NHS) Trusts (for predicting COVID-19). Whilst initially trained on data from over 77,000 patients, we created a model requiring only vital sign data and bedside blood count able to predict COVID-19 diagnosis in patients presenting at UK hospitals. We have demonstrated ability to adapt this model for a lower middle-income country (LMIC) setting using data from two Vietnamese hospitals. The adapted models achieved AUROCs around 0.75 and AUPRCs around 0.89 (similar to UK sites where much larger amounts of data were available). Performing "transfer learning," whereby a small subset of UK data was used to support model development in Vietnam, improved performances between 5-10%. We also found that using statistical methods for addressing missing values can further improve predictive performance by 2-5%. This machine learning model can also function as a 'baseline model' and be adapted for a new task i.e. dengue.

Conditions

Dengue Fever; Electronic Health Records; Prediction

Outcome Measures

Primary Outcomes (2)

• Dfferentiate dengue from unspecified causes of acute febrile illness

  To create AI models able to differentiate dengue from unspecified causes of acute febrile illness in terms of clinical diagnosis and characteristics

  At baseline (time of initial clinical presentation)

• Prediction of severe dengue

  To predict the development of severe dengue using routinely available clinical data

  At baseline (time of initial clinical presentation)

Study Arms (1)

Records of patients diagnosed with dengue and non-dengue infections

Medical record between 1January 2016 to 30 September 2024

Other: No intervention

Interventions

No intervention OTHER

No intervention

Records of patients diagnosed with dengue and non-dengue infections

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

All anonymized medical records of inpatients and out-patients adults aged ≥18 years visiting the Hospital for Tropical Diseases in Bangkok between January 1, 2016, and September 30, 2024, will be included in this study.

You may qualify if:

• Dengue-related ICD codes: A90-94, A910, A911, A919, A970-972, A979
• Non-dengue ICD codes: R78.81, A79.9, A27, B34.9, A49.9

You may not qualify if:

• Medical records with significant missing values, as determined by the Principal Investigators (PIs) and co-investigators.
• Records of patients diagnosed with mixed infections (causative agents ≥ 2)

Sponsors & Collaborators

• Mahidol University: lead
• National Hospital for Tropical Diseases, Hanoi, Vietnam: collaborator
• Hospital for Tropical Diseases, Ho Chi Minh City, Vietnam: collaborator

Study Sites (1)

Hospital for Tropical Diseases, Faculty of Tropical Medicine

Bangkok, 10400, Thailand

Location

Related Publications (7)

• Yang J, Dung NT, Thach PN, Phong NT, Phu VD, Phu KD, Yen LM, Thy DBX, Soltan AAS, Thwaites L, Clifton DA. Generalizability assessment of AI models across hospitals in a low-middle and high income country. Nat Commun. 2024 Sep 27;15(1):8270. doi: 10.1038/s41467-024-52618-6.

  PMID: 39333515 RESULT

• Yang J, Clifton L, Dung NT, Phong NT, Yen LM, Thy DBX, Soltan AAS, Thwaites L, Clifton DA. Mitigating machine learning bias between high income and low-middle income countries for enhanced model fairness and generalizability. Sci Rep. 2024 Jun 10;14(1):13318. doi: 10.1038/s41598-024-64210-5.

  PMID: 38858466 RESULT

• Soltan AAS, Yang J, Pattanshetty R, Novak A, Yang Y, Rohanian O, Beer S, Soltan MA, Thickett DR, Fairhead R, Zhu T, Eyre DW, Clifton DA; CURIAL Translational Collaborative. Real-world evaluation of rapid and laboratory-free COVID-19 triage for emergency care: external validation and pilot deployment of artificial intelligence driven screening. Lancet Digit Health. 2022 Apr;4(4):e266-e278. doi: 10.1016/S2589-7500(21)00272-7. Epub 2022 Mar 9.

  PMID: 35279399 RESULT

• Yang J, Soltan AAS, Clifton DA. Machine learning generalizability across healthcare settings: insights from multi-site COVID-19 screening. NPJ Digit Med. 2022 Jun 7;5(1):69. doi: 10.1038/s41746-022-00614-9.

  PMID: 35672368 RESULT

• McBride A, Vuong NL, Van Hao N, Huy NQ, Chanh HQ, Chau NTX, Nguyet NM, Ming DK, Ngoc NT, Nhat PTH, Phong NT, Tai LTH, Tho PV, Trung DT, Tam DTH, Trieu HT, Geskus RB, Llewelyn MJ, Thwaites CL, Yacoub S. A modified Sequential Organ Failure Assessment score for dengue: development, evaluation and proposal for use in clinical trials. BMC Infect Dis. 2022 Sep 3;22(1):722. doi: 10.1186/s12879-022-07705-8.

  PMID: 36057771 RESULT

• Luvira V, Silachamroon U, Piyaphanee W, Lawpoolsri S, Chierakul W, Leaungwutiwong P, Thawornkuno C, Wattanagoon Y. Etiologies of Acute Undifferentiated Febrile Illness in Bangkok, Thailand. Am J Trop Med Hyg. 2019 Mar;100(3):622-629. doi: 10.4269/ajtmh.18-0407.

  PMID: 30628565 RESULT

• Sa-Ngamuang C, Haddawy P, Luvira V, Piyaphanee W, Iamsirithaworn S, Lawpoolsri S. Accuracy of dengue clinical diagnosis with and without NS1 antigen rapid test: Comparison between human and Bayesian network model decision. PLoS Negl Trop Dis. 2018 Jun 18;12(6):e0006573. doi: 10.1371/journal.pntd.0006573. eCollection 2018 Jun.

  PMID: 29912875 RESULT

MeSH Terms

Conditions

Dengue

Condition Hierarchy (Ancestors)

Mosquito-Borne Diseases; Vector Borne Diseases; Infections; Arbovirus Infections; Virus Diseases; Flavivirus Infections; Flaviviridae Infections; RNA Virus Infections; Hemorrhagic Fevers, Viral

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: November 15, 2025
• First Posted: February 25, 2026
• Study Start: November 14, 2024
• Primary Completion: September 15, 2025
• Study Completion: September 15, 2025
• Last Updated: February 25, 2026

Record last verified: 2026-02

Locations

TH (1)