ISSN (Online): 2321-3418
server-injected
Medical Sciences and Pharmacy
Open Access

Machine Learning Prediction of BNP, hs-Troponin, and CRP from Routine Blood Tests in Heart Failure Patients: Evidence from Abakaliki, Ebonyi State, Nigeria

, , , ,
DOI: 10.18535/ijsrm/v13i09.mp03· Pages: 2324-2338· Vol. 13, No. 09, (2025)· Published: September 12, 2025
PDF
Views: 673 PDF downloads: 164

Abstract

Heart failure (HF) is a significant cause of morbidity and mortality worldwide, with a hefty burden in low-resource settings where diagnostic assays remain costly and limited in availability. This study examined whether routine haematology, coagulation, and biochemistry parameters can predict B-type Natriuretic Peptide, High-sensitivity Troponin and C-reactive Protein in HF patients using machine learning (ML).

Methods:

We prospectively enrolled 579 adults with HF at Alex Ekwueme Federal University Teaching Hospital, Abakaliki. We collected 10 mL of venous blood for laboratory investigations and obtained demographic and clinical data from their medical records. We processed the data in Python 3.12 and applied feature selection techniques including correlation thresholds, recursive feature elimination, its standard libraries. We trained and evaluated nine models, choosed the best model for each biomarker, and conducted sex-stratified analyses to compare performance between male and female participants.

Results:

Important predictors included Urea, creatinine, eGFR, D-dimer, fibrinogen, and neutrophil-to-lymphocyte ratio (NLR). Models that combined all the parameters outperformed single-domain models. CatBoost produced the best results for BNP (R² 0.30–0.33), ElasticNetCV for hs-Troponin (R² 0.09–0.12), and Ridge/ElasticNetCV for CRP (R² 0.53–0.54). SHAP analysis indicated that Urea and D-dimer strongly influenced BNP, while NLR, eGFR, and fibrinogen contributed most to the predictions of hs-Troponin and CRP. Sex-stratified models showed consistent behaviour across algorithms, with only minor differences in predictive strength.

Conclusion:

Routine laboratory data can estimate BNP, hs-Troponin, and CRP using ML in patients with HF from low-resource settings.

 

 

Keywords

Heart failureMachine learningBNPHigh-sensitivity troponinC-reactive proteinRoutine laboratory testsCardiac biomarker predictionLow-resource settings

References

  1. Parasar D, Ali A, Pillai NM, Shahi A, Alfurhood BS, Pant K. Detailed review on Integrated Healthcare Prediction System Using Artificial Intelligence and Machine Learning. In: 2023 3rd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE) [Internet]. 2023 [cited 2025 Aug 27]. p. 682–5. Available from: https://ieeexplore.ieee.org/document/10182709Google Scholar ↗
  2. James N, Gerrish L, Rokotyan N, Gladding PA. Machine Learning Applied to Routine Blood Tests and Clinical Metadata to Identify and Classify Heart failure [Internet]. medRxiv; 2021 [cited 2025 Aug 27]. p. 2021.07.26.21261115. Available from: https://www.medrxiv.org/content/10.1101/2021.07.26.21261115v1DOI ↗Google Scholar ↗
  3. Li J, Zhu P, Tang X, Jiang L, Li Y, Yan K, et al. Combined effect of D-dimer, hs-CRP, and Lp(a) on 5-year clinical outcomes after percutaneous coronary intervention: A large real-world study in China. iScience [Internet]. 2023 July 21 [cited 2025 Aug 27];26(7). Available from: https://www.cell.com/iscience/abstract/S2589-0042(23)01107-0Google Scholar ↗
  4. Duan S, Li Y, Yang P. Predictive value of blood urea nitrogen in heart failure: a systematic review and meta-analysis. Front Cardiovasc Med [Internet]. 2023 July 31 [cited 2025 Aug 27];10. Available from: https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2023.1189884/fullDOI ↗Google Scholar ↗
  5. Ogunmodede JA, Kolo PM, Bojuwoye MO, Dele-Ojo BF, Ogunmodede AJ, Omotoso AB. Characteristics of patients with acute heart failure in North Central Nigeria. Research Journal of Health Sciences. 2021 July 7;9(3):221–35.Google Scholar ↗
  6. Karaye KM, Dokainish H, ElSayed A, Mondo C, Damasceno A, Sliwa K, et al. Clinical Profiles and Outcomes of Heart Failure in Five African Countries: Results from INTER-CHF Study. Global Heart [Internet]. 2021 July 30 [cited 2025 Aug 27];16(1). Available from: https://globalheartjournal.com/articles/10.5334/gh.940DOI ↗Google Scholar ↗
  7. Feola M, Lombardo E, Testa M, Avogadri E, Piccolo S, Vado A. Prognostic factors of mid-term clinical outcome in congestive heart failure patients discharged after acute decompensation. Arch Med Sci. 2012 July 4;8(3):462–70.Google Scholar ↗
  8. Zhang Z, Cao L, Chen R, Zhao Y, Lv L, Xu Z, et al. Electronic healthcare records and external outcome data for hospitalized patients with heart failure. Sci Data. 2021 Feb 5;8(1):46.Google Scholar ↗
  9. McGranaghan P, Saxena A, Düngen HD, Rubens M, Appunni S, Salami J, et al. Performance of a cardiac lipid panel compared to four prognostic scores in chronic heart failure. Sci Rep. 2021 Apr 14;11(1):8164.Google Scholar ↗
  10. Hidayaturrohman QA, Hanada E. Impact of Data Pre-Processing Techniques on XGBoost Model Performance for Predicting All-Cause Readmission and Mortality Among Patients with Heart Failure. BioMedInformatics. 2024 Dec;4(4):2201–12.Google Scholar ↗
  11. Méndez-Bailón M, Iguarán-Bermúdez R, López-García L, Sánchez-Sauce B, Pérez-Mateos P, Barrado-Cuchillo J, et al. Prognostic Value of the PROFUND Index for 30-Day Mortality in Acute Heart Failure. Medicina (Kaunas). 2021 Oct 23;57(11):1150.Google Scholar ↗
  12. Chou RH, Huang PH, Hsu CY, Chang CC, Leu HB, Huang CC, et al. Circulating Fibroblast Growth Factor 21 is Associated with Diastolic Dysfunction in Heart Failure Patients with Preserved Ejection Fraction. Sci Rep. 2016 Sept 21;6(1):33953.Google Scholar ↗
  13. Kadowaki S, Watanabe T, Otaki Y, Narumi T, Honda Y, Takahashi H, et al. Combined assessment of myocardial damage and electrical disturbance in chronic heart failure. World Journal of Cardiology. 2017 May 26;9(5):457–65.Google Scholar ↗
  14. Spoto S, Argemi J, Di Costanzo R, Gavira Gomez JJ, Salterain Gonzales N, Basili S, et al. Mid-Regional Pro-Adrenomedullin and N-Terminal Pro-B-Type Natriuretic Peptide Measurement: A Multimarker Approach to Diagnosis and Prognosis in Acute Heart Failure. Journal of Personalized Medicine. 2023 July;13(7):1155.Google Scholar ↗
  15. Olasińska-Wiśniewska A, Urbanowicz T, Perek B, Misterski M, Grodecki K, Grygier M, et al. Predictive Value of Monocyte-To-Lymphocyte Ratio in Differentiating Heart Failure with Reduced Ejection Fraction in Patients with Severe Aortic Stenosis—A Retrospective Analysis. Journal of Clinical Medicine. 2024 Jan;13(20):6249.Google Scholar ↗
  16. Tuersun R, Abudouwayiti A, Li YX, Pan Y, Aimaier S, Wen ZY, et al. Serum CA125: a prognostic biomarker for mortality in chronic heart failure. BMC Cardiovascular Disorders. 2025 Mar 27;25(1):227.Google Scholar ↗
  17. Liyis BG de, Ciaves AF, Intizam MH, Jusuf PJ, Artha IMJR. Hematological Biomarkers of Troponin, Neutrophil-to-Lymphocyte Ratio, and Monocyte-toLymphocyte Ratio Serve as Effective Predictive Indicators of High-Risk Mortality in Acute. BioMedicine. 2023 Nov 27;13(4):32–43.Google Scholar ↗
  18. Ochi Y, Kubo T, Baba Y, Nakashima Y, Ueda M, Takahashi A, et al. Prediction of Medium-Term Mortality in Japanese Patients With Wild-Type Transthyretin Amyloidosis. Circulation Reports. 2020;2(6):314–21.Google Scholar ↗
  19. Xu Y, Song C, Wei W, Miao R. Development and validation of a cardiac surgery-associated acute kidney injury prediction model using the MIMIC-IV database. PLOS ONE. 2025 June 12;20(6):e0325151.Google Scholar ↗
  20. Nyawo TA, Dludla PV, Mazibuko-Mbeje SE, Mthembu SXH, Nyambuya TM, Nkambule BB, et al. A systematic review exploring the significance of measuring epicardial fat thickness in correlation to B-type natriuretic peptide levels as prognostic and diagnostic markers in patients with or at risk of heart failure. Heart Fail Rev. 2022 Mar 1;27(2):665–75.Google Scholar ↗
  21. Pérez-Herrero S, Lorenzo-Villalba N, Urbano E, Sánchez-Sauce B, Aguilar-Rodríguez F, Bernabeu-Wittel M, et al. Prognostic Significance of Lung and Cava Vein Ultrasound in Elderly Patients Admitted for Acute Heart Failure: PROFUND-IC Registry Analysis. Journal of Clinical Medicine. 2022 Jan;11(15):4591.Google Scholar ↗
  22. Rehman SU, James L. Januzzi J. Natriuretic Peptide Testing in Primary Care. Current Cardiology Reviews. 4(4):300–8.Google Scholar ↗
  23. Noveanu M, Breidthardt T, Potocki M, Reichlin T, Twerenbold R, Uthoff H, et al. Direct comparison of serial B-type natriuretic peptide and NT-proBNP levels for prediction of short- and long-term outcome in acute decompensated heart failure. Critical Care. 2011 Jan 5;15(1):R1.Google Scholar ↗
  24. Peeters JMPWU, Sanders-van Wijk S, Bektas S, Knackstedt C, Rickenbacher P, Nietlispach F, et al. Biomarkers in outpatient heart failure management; Are they correlated to and do they influence clinical judgment? Neth Heart J. 2014 Mar 1;22(3):115–21.Google Scholar ↗
  25. de Boer RA. Clinical evaluation of heart failure patients: what good are biomarkers bringing us? Neth Heart J. 2014 Mar 1;22(3):113–4.Google Scholar ↗
  26. Choi JS, Trinh TX, Ha J, Yang MS, Lee Y, Kim YE, et al. Implementation of Complementary Model using Optimal Combination of Hematological Parameters for Sepsis Screening in Patients with Fever. Sci Rep. 2020 Jan 14;10(1):273.Google Scholar ↗
  27. Truslow JG, Goto S, Homilius M, Mow C, Higgins JM, MacRae CA, et al. Cardiovascular Risk Assessment Using Artificial Intelligence-Enabled Event Adjudication and Hematologic Predictors. Circulation: Cardiovascular Quality and Outcomes. 2022 June;15(6):e008007.Google Scholar ↗
  28. Goriki Y, Tanaka A, Nishihira K, Kawaguchi A, Natsuaki M, Watanabe N, et al. A Novel Predictive Model for In-Hospital Mortality Based on a Combination of Multiple Blood Variables in Patients with ST-Segment-Elevation Myocardial Infarction. Journal of Clinical Medicine. 2020 Mar;9(3):852.Google Scholar ↗
  29. Yan Q, Chen S. Hemoconcentration is a valuable predictor of prognosis in patients with acute heart failure. Experimental and Therapeutic Medicine. 2020 Apr 1;19(4):2792–8.Google Scholar ↗
  30. Kim KH, Park JH, Ro YS, Hong KJ, Song KJ, Shin SD. Emergency department routine data and the diagnosis of acute ischemic heart disease in patients with atypical chest pain. PLOS ONE. 2020 Nov 5;15(11):e0241920.Google Scholar ↗
  31. Zhang Z, Shao B, Liu H, Huang B, Gao X, Qiu J, et al. Construction and Validation of a Predictive Model for Coronary Artery Disease Using Extreme Gradient Boosting. JIR. 2024 July 1;17:4163–74.Google Scholar ↗
  32. Vinci R, d’Aiello A, Costantino S, Capocchiano ND, Matamalas JT, Pedicino D. Editorial: Artificial intelligence applied to coronary artery diseases: from pathophysiology to precision medicine. Front Cardiovasc Med [Internet]. 2023 Aug 9 [cited 2025 Aug 27];10. Available from: https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2023.1247620/fullDOI ↗Google Scholar ↗
  33. Wang Z, Gu Y, Huang L, Liu S, Chen Q, Yang Y, et al. Construction of machine learning diagnostic models for cardiovascular pan-disease based on blood routine and biochemical detection data. Cardiovascular Diabetology. 2024 Sept 28;23(1):351.Google Scholar ↗
  34. Cadamuro J. Rise of the Machines: The Inevitable Evolution of Medicine and Medical Laboratories Intertwining with Artificial Intelligence—A Narrative Review. Diagnostics. 2021 Aug;11(8):1399.Google Scholar ↗
  35. Wickramaratne SD, Shaad Mahmud M. Bi-Directional Gated Recurrent Unit Based Ensemble Model for the Early Detection of Sepsis. In: 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) [Internet]. 2020 [cited 2025 Aug 27]. p. 70–3. Available from: https://ieeexplore.ieee.org/document/9175223Google Scholar ↗
  36. Olasińska-Wiśniewska A, Urbanowicz T, Perek B, Misterski M, Grodecki K, Grygier M, et al. Predictive Value of Monocyte-To-Lymphocyte Ratio in Differentiating Heart Failure with Reduced Ejection Fraction in Patients with Severe Aortic Stenosis—A Retrospective Analysis. Journal of Clinical Medicine. 2024 Jan;13(20):6249.Google Scholar ↗
  37. Pereira-Barretto AC, Oliveira Junior MT de, Strunz CC, Del Carlo CH, Scipioni AR, Ramires JAF. O nível sérico de NT-proBNP é um preditor prognóstico em pacientes com insuficiência cardíaca avançada. Arq Bras Cardiol. 2006;87:174–7.Google Scholar ↗
  38. Ikonomidis I, Michalakeas CA, Lekakis J, Paraskevaidis I, Th. Kremastinos D. Multimarker Approach in Cardiovascular Risk Prediction. Disease Markers. 2009;26(5–6):135423.Google Scholar ↗
  39. Oommen SG, Man RK, Talluri K, Nizam M, Kohir T, Aviles MA, et al. Heart Failure With Improved Ejection Fraction: Prevalence, Predictors, and Guideline-Directed Medical Therapy. Cureus. 16(6):e61790.Google Scholar ↗
  40. Akhtar M, Elliott PM. Risk Stratification for Sudden Cardiac Death in Non-Ischaemic Dilated Cardiomyopathy. Curr Cardiol Rep. 2019 Nov 25;21(12):155.Google Scholar ↗
  41. Sadahiro T, Kohsaka S, Okuda S, Inohara T, Shiraishi Y, Kohno T, et al. MRI and serum high-sensitivity C reactive protein predict long-term mortality in non-ischaemic cardiomyopathy. Open Heart [Internet]. 2015 Oct 19 [cited 2025 Aug 27];2(1). Available from: https://openheart.bmj.com/content/2/1/e000298Google Scholar ↗
  42. Méndez-Bailón M, Iguarán-Bermúdez R, López-García L, Sánchez-Sauce B, Pérez-Mateos P, Barrado-Cuchillo J, et al. Prognostic Value of the PROFUND Index for 30-Day Mortality in Acute Heart Failure. Medicina. 2021 Nov;57(11):1150.Google Scholar ↗
  43. Dong M, Wang L, Tse G, Dai T, Wang L, Xiao Z, et al. Effectiveness and safety of transcatheter aortic valve replacement in elderly people with severe aortic stenosis with different types of heart failure. BMC Cardiovascular Disorders. 2023 Jan 18;23(1):34.Google Scholar ↗
  44. Stock EO, Asztalos BF, Miller JM, He L, Creasy KT, Schwemberger R, et al. High-Density Lipoprotein Particles, Inflammation, and Coronary Heart Disease Risk. Nutrients. 2025 Jan;17(7):1182.Google Scholar ↗
  45. Soni N, Kaur P, Gurjar V, Bhargava A, Tiwari R, Chouksey A, et al. Unveiling the Interconnected Dynamics of Mitochondrial Dysfunction Associated With Age-Related Cardiovascular Risk: A Cross-Sectional Pilot Study. Cureus. 17(4):e82961.Google Scholar ↗
  46. Meyer L, Mulder D, Wallace J. A supervised machine learning approach with feature selection for sex-specific biomarker prediction. npj Syst Biol Appl. 2025 July 1;11(1):69.Google Scholar ↗
  47. Shou L, Huang WW, Barszczyk A, Wu SJ, Han H, Waese-Perlman A, et al. Blood Biomarkers Predict Cardiac Workload Using Machine Learning. BioMed Research International. 2021;2021(1):6172815.Google Scholar ↗
  48. Pan Y, Wei M, Jin M, Liang Y, Yi T, Tu J, et al. An interpretable machine learning model based on optimal feature selection for identifying CT abnormalities in patients with mild traumatic brain injury. eClinicalMedicine [Internet]. 2025 Apr 1 [cited 2025 Aug 27];82. Available from: https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(25)00124-5/fulltextGoogle Scholar ↗
Author details
Ikechukwu Ogbo
Department of Haematology and Blood Transfusion Science, Faculty of Health Science and Technology, Ebonyi State University (EBSU)
✉ Corresponding Author
👤 View Profile →
Nancy Chiatogu Ibeh
Department of Haematology and Blood Transfusion Science, Faculty of Medical Laboratory Science, Nnamdi Azikiwe University (NAU)
👤 View Profile →🔗 Is this you? Claim this publication
Grace Ifechukwu Amilo
Department of Haematology, Faculty of Basic Clinical Medicine, Nnamdi Azikiwe University (NAU)
👤 View Profile →🔗 Is this you? Claim this publication
Chizoba Okechukwu Okeke
Department of Haematology and Blood Transfusion Science, Faculty of Medical Laboratory Science, Nnamdi Azikiwe University (NAU)
👤 View Profile →🔗 Is this you? Claim this publication
Joelson Friday Onwe
Haematology and Blood Group Serology Unit, Department of Medical Laboratory Services, Alex Ekwueme Federal University Teaching Hospital, Abakaliki
👤 View Profile →🔗 Is this you? Claim this publication