Recently, radiomics combined with machine learning method has been widely used in clinical practice. Compared with traditional imaging studies that explore the underlying mechanisms, the machine learning method focuses on classification and prediction to propose personalized diagnosis and treatment strategies. However, these studies were based on thin-section research-quality brain MR imaging with section thickness of \< 2 mm. Clinical, the usage of thick-section clinical setting instead of thin-section research setting is especially important to shorten the acquisition time to reduce the patient's suffering. Here investigators want to build multiparametric diagnostic model of migraineurs without aura using radiomics features extracted from thick-section clinical-quality brain MR images.
Study Type
OBSERVATIONAL
Enrollment
400
using radiomics features from multiparametric thick-section clinical-quality brain MRI to distinguish migraineurs from health controls.
Xijing Hospital
Xi'an, Shaanxi, China
accuracy
a measure of statistical bias which measures the proportion of health controls and migrainures that are correctly identified as such.
Time frame: 2018.7-2019.12
sensitivity
true positive rate of detection in migraineurs which measures the proportion of actual migraineurs that are correctly identified as such.
Time frame: 2018.7-2019.12
specificity
true negative rate measures the proportion of actual health controls that are correctly identified as such.
Time frame: 2018.7-2019.12
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.