Improving Automatic Migraine Classification Performance with Naive Bayes

Year 2024, Volume: 28 Issue: 4, 816 - 823, 31.08.2024

Arzum Karataş

https://doi.org/10.16984/saufenbilder.1332882

Abstract

The expertise of the physician and the patient's ongoing observation are the two primary contributing factors in diagnosing migraines. However, individuals who experience migraines in the early stages frequently visit emergency rooms or different outpatient clinics, such as internal medicine, ophthalmology, and family medicine. Additionally, the type of migraine is frequently misdiagnosed due to the severity of the symptoms being misjudged or because the five-to-ten-minute examination period is insufficient for achieving an accurate diagnosis. Incorrect treatment of this type can have adverse effects on the patient's health. The majority of research in this field has concentrated on the study of brainwaves, leading to the development of complex tests that are only available to a small proportion of the population. However, one study has made progress in automatic migraine classification. The study, which demonstrates 97% classification performance above that of previous studies and produces findings in a timely manner, provides a decision support mechanism that will assist clinicians in the proper classification of migraine type. Given that over 20% of Turkey's population suffers from migraines, our study concentrated on the same issue to enhance classification performance in terms of accuracy and training time. The Naive Bayes model was employed in the study to categorize the various types of migraines, and the performance of the model was evaluated using data from actual migraine sufferers. The classification model utilized exhibited superior classification performance compared to previous studies, with 99% accuracy and precision. Additionally, the model's training time in the same dataset was the shortest when compared to other benchmarked classifier models. The application of the Naive Bayes classifier to the classification of migraines represents a highly effective technique that can facilitate rapid, accurate clinical diagnoses, thereby enabling physicians to provide their patients with precise diagnoses.

Keywords

Migraine type diagnosis, Naïve Bayes, Classification

References

• M. Ashina, Z. Katsarava, T. P. Do, D. C. Buse, P. Pozo-Rosich, A Özge, A. V Krymchantowski, E. R. Lebedeva, K. Ravishankar, S. Yu, S. Sacco, S. Ashina, S. Younis, T. J Steiner, R. B Lipton, “Migraine: epidemiology and systems of care,” The Lancet, vol. 397, no. 10283, pp. 1485-1495, 2021.
• J. Gupta, S. S. Gaurkar, “Migraine: An Underestimated Neurological Condition Affecting Billions,” Cureus, vol. 14, no.8, Art no. e28347, 2022.
• J. Berg, L. J. Stovner, L. J., “Cost of migraine and other headaches in Europe,” European Journal of Neurology, vol. 12, no. June 2005, pp. 59-62, 2005.
• P. J. Goadsby, “Recent advances in understanding migraine mechanisms, molecules and therapeutics,” Trends in Molecular Medicine, vol. 13, no. 1, pp. 39-44, 2007.
• P. Parisi, A. Verrotti, M. C. Paolino, A. Ferretti, F. D. Sabatino, F.D., “Obesity and Migraine in Children,” in Omega-3 Fatty Acids in Brain and Neurological Health, R.R. Watson and F. D. Meester, Eds. Cambridge, UK: Cambridge Academic Press, 2014, pp. 277-286.
• Headache Classification Committee of the International Headache Society, “The International Classification of Headache Disorders, 3rd edition,” Cephalalgia, vol. 38, no.1, pp.1-211, 2018
• T. Shimizu, F. Sakai, H. Miyake, T. Sone, M. Sato, S. Tanabe, Y. Azuma, D. W. Dodick,“ Disability, quality of life, productivity impairment and employer costs of migraine in the workplace,” The Journal of Headache and Pain, vol. 22, no. 29., Art no. 29, 2021.
• C. Wöber, W. Brannath, K. Schmidt, M. Kapitan, E. Rude, P. Wessely, Ç. Wöber-Bingöl, the PAMINA Study Group, “Prospective Analysis of Factors Related to Migraine Attacks: The PAMINA Study,” Cephalalgia, vol. 17, no. 4, pp. 304-314, 2007.
• P. T. Fukui, T. R. T. Goncalves, C. G. Strabelli, N. M. F. Lucchino, F. C. Matos, C. Fernanda, J. P. M. D. Santos, E. Zukerman, V. Zukerman-Guendler, J. P. Mercante, M. R. Masruha, “Trigger factors in migraine patients,” Arquivos de Neuro-Psiquiatria, vol. 66, no. September 2008, pp. 494-499, 2008 .
• J. M. Pavlovic, D. C. Buse, C.M. Sollars, S. Haut, R.B. Lipton, “Trigger Factors and Premonitory Features of Migraine Attacks: Summary of Studies,” Headache The Journal of Head and Face Pain, vol. 54, no. 10, pp. 1670-1679, 2014.
• Turkish Neurological Society. (2019, Jun. 23). Türk Nöroloji Derneği Basin Bülteni 22 Temmuz “Dünya Beyin Günü - Migren”[Online]. Available: https://www.noroloji.org.tr/haber/586/turk-noroloji-dernegi-basin-bulteni-22-temmuz-dunya-beyin-gunu-migren.
• R. Agosti, “Migraine burden of disease: From the patient's experience to a socio‐economic view,” The Journal of Headache and Pain, vol. 58, pp. 17-32, 2018.
• M. Bonafede, S. Sapra, N. Shah, S. Tepper, K. Cappell, P. Desai, “Direct and indirect healthcare resource utilization and costs among migraine patients in the United States,” The Journal of Headache and Pain, vol. 58, no.5, pp. 700-714, 2018.
• T. Takeshima, Q. Wan, Y. Zhang, M. Komori, S. Stretton, N. Rajan, T. Treuer, K. Ueda, “Prevalence, burden, and clinical management of migraine in China, Japan, and South Korea: a comprehensive review of the literature,” The Journal of Headache and Pain, vol. 20, Art no. 111, 2019.
• L. P. Wong, H. Alias, N. Bhoo-Pathy, I. Chung, Y. C. Chong, S. Kalra, Z. U. B. S. Shah, “Impact of migraine on workplace productivity and monetary loss: a study of employees in banking sector in Malaysia,” The Journal of Headache and Pain, vol. 21, no. 68, 2020.
• Headache Classification Subcommittee of the International Headache Society, “The International Classification of Headache Disorders: 2nd edition,” Cephalalgia, vol. 24 (Suppl 1), pp. 9-160, 2004.
• S. Tarantino, A. Capuano, R. Torriero, M. Citti, C. Vollono, S. Gentile, F. Vigevano, M. Valeriani, “Migraine Equivalents as Part of Migraine Syndrome in Childhood,” Pediatric Neurology, vol. 51, no. 5, pp. 645-649, 2014.
• S. B. Akben, D. Tuncel, A. Alkan, “Classification of multi-channel EEG signals for migraine detection,” Biomedical Research., vol. 27, no.3, pp. 743-748, 2016.
• P. A. Sanchez-Sanchez, J. R. García-González, J. M. R. Ascar, “Automatic migraine classification using artificial neural networks”, F1000 Research, vol. 16, no. 9, Art no. 618, 2020.
• P. Domingos, M. Pazzani, “On the Optimality of the Simple Bayesian Classifier under Zero-One Loss,” Machine Learning, vol. 29, pp. 103-130, 1997.
• P. A. Sánchez-Sánchez, J. R. García-González, J. M. R. Ascar. (2020). Migraine Classification Model [Online]. Available: https://codeocean.com/capsule/1269964/tree/v1.
• F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, no.85, pp. 2825-2830, 2011.
• Zenodo.(2020). Pandas [computer software library] [Online]. Available: https://pandas.pydata.org/.