Evaluating the Impact of Carboxyhemoglobin Level in the Human Body in Lagos State Using Machine Learning Algorithms

Omotayo Oluwatusin, Nsikan Ime Obot, Oloruntoba Ekun, Olabode Thomas Olakoyejo, Kehinde Orolu, Okwisilieze Uwadoka, Hussein Babalola, Adekunle Omolade Adelaja, Comfort Oluwaseyi Folorunso, Idowu O. Fadeyibi, Ayo A. Oyediran and Oluyemi Akinloye

Omotayo Oluwatusin, Nsikan Ime Obot, Oloruntoba Ekun, Olabode Thomas Olakoyejo, Kehinde Orolu, Okwisilieze Uwadoka, Hussein Babalola, Adekunle Omolade Adelaja, Comfort Oluwaseyi Folorunso, Idowu O. Fadeyibi, Ayo A. Oyediran and Oluyemi Akinloye

Keywords: Artificial Neural Network, Carboxyhaemoglobin, Machine Learning, Lagos State

Abstract

The detection of carboxyhaemoglobin (COHb), a remarkably stable yet harmful complex present in body cells, presents a significant challenge. Elevated COHb levels can cause symptoms like headaches, nausea, dizziness, and, in severe cases, coma or death. This study utilised thirteen predictive variables, including sex, body mass index, glucose, and blood pressure. The COHb levels in Lagos State, Nigeria, were classified using various machine learning algorithms and variables. Evaluation metrics such as accuracy, precision, and confusion matrices were employed for assessment. Highly varied but negatively correlated factors significantly influenced ML predictions of COHb. Glucose was identified as the most influential predictor, due to food oxidation, it combines with oxygen and dissociates carbon monoxide from the blood. While seven out of twelve models that did not overfit during the training phase were retained, the best-performing model was an artificial neural network (ANN) with seven hidden layers of six neurons each. Apart from being the only model that correctly classified the rare individual of the fourth group by avoiding misplacement into the first group of many persons in the confusion matrix, the ANN scheme achieved the highest scores of 70% and 64% in accuracy and precision, respectively, during generalisation, alongside other optimal performances.

References

Abbey, M., Amadi, S.C., Ocheche, U.S., Wekere, F.C.C., Altraide, B.O., Oloyede, O.A., Inimgba, N.M., and Akani, C.I. (2022). Maternal exposure to carbon monoxide in the first trimester (7-13+6 weeks) of pregnancy in the core Niger Delta. International Journal of Reproduction, Contraception, Obstetrics and Gynecology 11, 1839-1847.
Adedokun, O.A., and Owoade, O.K. (2019). Source apportionment of vehicular exhaust emissions in Lagos, Nigeria. Sustainable Cities and Society, 47, 101484.
Bemtgen, X., Rilinger, J., Jäckel, M., Zotzmann, V., Supady, A., Benk, C., Bode, C., Wengenmayer, T., Lother, A., and Staudacher, D.L. (2021). Admission blood glucose level and outcome in patients requiring venoarterial extracorporeal membrane oxygenation. Clinical Research in Cardiology 110, 1484–1492.
Bickler, M.P., and Rhodes, L.J. (2018). Accuracy of detection of carboxyhemoglobin and methemoglobin in human and bovine blood with an inexpensive, pocket-size infrared scanner. PloS one 13, e0193891.
Crecelius, A.R., Kirby, B.S., and Dinenno, F.A. (2015). Intravascular ATP and regulation of blood flow and oxygen delivery in humans. Exercise and Sport Sciences Review 43, 5–13.
Guarnaccia, C., Cerón Bretón, J.G., Quartieri, J., Tepedino, C. and Cerón Bretón, R.M. (2014). An application of time series analysis for forecasting and control of carbon monoxide concentrations. International Journal of Mathematical Models and Methods in Applied Sciences 8, 505–515.
Hampson, N.B., and Hauff, N.M. (2008). Carboxyhemoglobin levels in carbon monoxide poisoning: do they correlate with the clinical picture? The American Journal of Emergency Medicine 26, 665–669.
Hanis, T.M., Islam, M.A., and Musa, K.I. (2022). Diagnostic accuracy of machine learning models on mammography in breast cancer classification: a meta-analysis. Diagnostics 12, 1643.
Huzar, T.F., George, T., and Cross, J.M. (2013). Carbon monoxide and cyanide toxicity: etiology, pathophysiology, and treatment in inhalation injury. Expert Review of Respiratory Medicine 7, 159–170.
Isa, R.O., Odigure, J.O., Okafor, J.O., Abdulkareem, A.S., Abdulfatai, J., and Afolabi, A.S. (2013). Mathematical modeling of road traffic carbon monoxide pollutant: a case study of Minna Metropolis, Nigeria. International Review of Chemical Engineering 5, 57– 65.
Jeon, S.B., Sohn, C.H., Seo, D.W., Oh, B.J., Lim, K.S., Kang, D.W., and Kim, W.Y. (2018). Acute brain lesions on magnetic resonance imaging and delayed neurological sequelae in carbon monoxide poisoning. JAMA Neurology 75, 436–443.
Jones, D.P., and Kennedy, F.G. (1982). Intracellular oxygen supply during hypoxia. American Journal of Physiology 243, C247.
Lee, I., Park, N., Lee, H., Hwang, C., Kim, J.H. and Park, S., (2021). Systematic review on human skin-compatible wearable photoplethysmography sensors. Applied Sciences 11, 2313.
Mattiuzzi, C., and Lippi, G. (2020). Worldwide epidemiology of carbon monoxide poisoning. Human and Experimental Toxicology 39, 387–392.
Maynard, R.L., and Robert, W. (1999). Carbon monoxide. In Air pollution and Health (pp. 749-796). Academic Press.
McNair, H.M., Miller, J.M., and Snow, N.H. (2019). Basic gas chromatography. John Wiley & Sons.
Murphy, K.P. (2012). Machine learning: a probabilistic perspective. The MIT Press.
Obot, N.I. (2024). Regression models and hybrid intelligent systems for estimating clear-sky downward longwave radiation in Equatorial Africa. Journal of Earth and Space Physics 49, 143–159.
Obot, N.I., Olubgon, B., Humphrey, I., and Akeem, R.A. (2023). Equatorial all-sky downward longwave radiation modelling. Communication in Physical Sciences 9, 111–124.
Oghenejoboh, K. M., and Adiotomre, K. O. (2012). Finite volume modelling of carbon monoxide from vehicular emissions at Oshodi market in Lagos Metropolis of Nigeria. Journal of Computations and Modelling 2, 55–77.
Oluwatusin, O., Orolu, K., Ekun, O., Akinloye, O., and Oyediran, O.O. (2019). Mathematical model to evaluate the effect of carbon monoxide exposure as a function of gender, age, and height. Journal of Applied Sciences and Environmental Management 23, 799–803.
Oluwole, O., Arinola, G., Ologun, G., and Fatusi, A. (2016). The impact of air pollution on respiratory and cardiovascular health in Lagos, Nigeria. Journal of Environmental and Public Health 8086069.
Piantadosi, C.A. (2004). Carbon monoxide poisoning. Undersea and Hyperbaric Medicine 31, 167–177.
Prockop, L.D., and Chichkova, R.I. (2007). Carbon monoxide intoxication: an updated review. Journal of the Neurological Sciences 262, 122–130.
Rodkey, F.L., O’Neal, J.D., Collison, H.A., and Uddin, D.E. (1974). Relative affinity of hemoglobin S and hemoglobin A for carbon monoxide and oxygen. Clinical Chemistry 20, 83–84.
Ryter, S.W., Alam, J., and Choi, A.M., (2006). Heme oxygenase 1/carbon monoxide: from basic science to therapeutic applications. Physiological Reviews 86, 583–650.
Samuel, J.M., Kahl, J.H., Zaney, M.E., Hime, G.W., and Boland, D.M. (2021). Comparison of spectrophotometric methods for the determination of carboxyhemoglobin in post-mortem blood. Journal of Analytical Toxicology 45, 885–891.
Shi, B., Zhou, T., Lv, S., Wang, M., Chen, S., Heidari, A.A., Huang, X., Chen, H., Wang, L. and Wu, P. (2022). An evolutionary machine learning for pulmonary hypertension animal model from arterial blood gas analysis. Computers in Biology and Medicine 146, 105529.
Sobamowo, M. G. (2016). Prediction of the effects of combustion–generated pollutant on human health: mathematical models and numerical solutions. Iranian (Iranica) Journal of Energy and Environment 7, 233–242.
Wu, A.H., (2019). “On vivo” and wearable clinical laboratory testing devices for emergency and critical care laboratory testing. Journal of Applied Laboratory Medicine 4(2), 254–263.