| pubmed-article:15613224 | pubmed:abstractText | Artificial neural networks present a technique for modeling relationships between variables in complex systems. The negative effects of headache are determined by many "biopsychosocial" elements that represent a complex system. Artificial neural networks may therefore be useful for examining psychological factors in headache. To test this hypothesis, we trained an artificial neural network to predict life-style interference attributed to headache from psychological measures of anger, depression, and coping appraisal and strategies. The artificial neural network demonstrated a better fit of the data than that obtained by multiple regression, and predicted interference levels to within 10% error for 80% of novel cases. Artificial neural networks may be a useful technique for examining psychological correlates of headache. | lld:pubmed |
