Chronic Pain Diagnosis Using Machine Learning, Questionnaires, and QST: A Sensitivity Experiment

• DOI: 10.3390/diagnostics10110958
• Publication Date: 2020-11-17T12:23:28Z
• Authors: Alex Novaes Santana; Charles Novaes de...; Pedro Montoya
• Keywords: Ensemble Learning

In the last decade, machine learning has been widely used in different fields, especially because of its capacity to work with complex data. With support techniques, studies have using data-driven approaches better understand some syndromes like mild cognitive impairment, Alzheimer's disease, schizophrenia, and chronic pain. Chronic pain is a disease that can recurrently be misdiagnosed due comorbidities other which it shares symptoms. Within context, several suggesting algorithms classify or predict conditions. Those were fed diversity data types, from self-report based on questionnaires most advanced brain imaging techniques. this study, we assessed sensitivity datasets classifying syndromes. Together assessment, highlighted important methodological steps should taken into account when an experiment conducted. The best results obtained by ensemble-based dataset containing greatest information, resulting area under receiver operating curve (AUC) values around 0.85. addition, performance strongly related hyper-parameters. Thus, good strategy for hyper-parameter optimization extract algorithm. These findings notion powerful tool