Effect of Demographic Characteristics on Tension Type Headache and Migraine: A Machine Learning Based Analysis

Headache is an important neurological disorder that affects people's daily life. Headache types can be confused according to their symptoms. Thanks to the clinical decision support systems (CDSS) that can be developed, this confusion can be prevented and support can be provided to physicians. For this purpose, in this preliminary study, migraine and tension-type headache (TTH) types were classified using five different machine learning algorithms. In the classification phase, anamnesis data based on headache symptom information and demographic information were used. The classification process consists of different stages. In the first stage, the classification process was performed using only headache symptom data. Then, gender, age, and occupation information from demographic information were added to these data, and their effects were analyzed. As a result of the analysis, while the effect of gender and occupation positively affected the classification accuracy rate (2.31% increase), the effect of age negatively affected it. In addition, in this research, gender and occupational conditions that positively affect the results were analyzed within themselves, and a preliminary study of gender-specific or occupation-specific systems that could be developed in the future was carried out. This research is a preliminary study of a future CDSS. Thanks to the CDSS to be realized, headache types can be differentiated in regions where there is a medical shortage or a shortage of specialist physicians.