Introduction: Headache is common at emergency services and neuroimaging can help in timely diagnosis of life-threatening pathologies, although overuse can generate increased healthcare costs and irradiation, which is why red flag signs are used to try to guide the neuroimaging indication. We decided to evaluate predictive variables of abnormal neuroimaging to derive and validate a scale that would have appropriate diagnostic performance and is reproducible.Materials and methods: Observational Analytical, retrospective, single-center study.Results: 626 patients with non-traumatic headache treated in the emergency room, 15·5% with abnormal neuroimaging. The variables with the highest OR were: age > 40 years (4·92), nuchal rigidity (15·80), motor deficit (5·4), visual deficit (3·2) and gait disturbance (2·27). 3 abnormal neuroimaging prediction models have been derived, which had AUC/ROC Model 1: 0·71, Model 2: 0·75, Model 3: 0·72. All models have Hosmer-Lemeshow > 0·05. Scale is performed with model 1, which is validated internally and a cut-off point of 0·179, the AUC/ROC of 0·757 is obtained with a diagnostic accuracy of 0·79 (0·73-0·85).Conclusion: Clinically and statistically significant variables associated with abnormal neuroimaging have been identified in patients who came to the emergency room for headache and a scale has been derived and internally validated that has high accuracy, diagnostic performance and predictive capacity, which allows discriminating patients who require neuroimaging of those who do not. This scale has a better overall diagnostic performance vs red flag signs.Funding Information: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. The authors received no financial support for the research, authorship, and/or publication of this article.Declaration of Interests: The Authors declares that there is no conflict of interest. Ethics Approval Statement: The institutional biomedical research ethics committee approved this study applying the declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects. This study adhered to the standards of the STROBE guidelines.