Title |
Automated Quality Assessment of Structural Magnetic Resonance Brain Images Based on a Supervised Machine Learning Algorithm
|
---|---|
Published in |
Frontiers in Neuroinformatics, December 2016
|
DOI | 10.3389/fninf.2016.00052 |
Pubmed ID | |
Authors |
Ricardo A. Pizarro, Xi Cheng, Alan Barnett, Herve Lemaitre, Beth A. Verchinski, Aaron L. Goldman, Ena Xiao, Qian Luo, Karen F. Berman, Joseph H. Callicott, Daniel R. Weinberger, Venkata S. Mattay |
Abstract |
High-resolution three-dimensional magnetic resonance imaging (3D-MRI) is being increasingly used to delineate morphological changes underlying neuropsychiatric disorders. Unfortunately, artifacts frequently compromise the utility of 3D-MRI yielding irreproducible results, from both type I and type II errors. It is therefore critical to screen 3D-MRIs for artifacts before use. Currently, quality assessment involves slice-wise visual inspection of 3D-MRI volumes, a procedure that is both subjective and time consuming. Automating the quality rating of 3D-MRI could improve the efficiency and reproducibility of the procedure. The present study is one of the first efforts to apply a support vector machine (SVM) algorithm in the quality assessment of structural brain images, using global and region of interest (ROI) automated image quality features developed in-house. SVM is a supervised machine-learning algorithm that can predict the category of test datasets based on the knowledge acquired from a learning dataset. The performance (accuracy) of the automated SVM approach was assessed, by comparing the SVM-predicted quality labels to investigator-determined quality labels. The accuracy for classifying 1457 3D-MRI volumes from our database using the SVM approach is around 80%. These results are promising and illustrate the possibility of using SVM as an automated quality assessment tool for 3D-MRI. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
Australia | 1 | 20% |
United States | 1 | 20% |
Switzerland | 1 | 20% |
Unknown | 2 | 40% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Members of the public | 4 | 80% |
Scientists | 1 | 20% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
Netherlands | 1 | 1% |
United States | 1 | 1% |
Unknown | 89 | 98% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Ph. D. Student | 20 | 22% |
Researcher | 14 | 15% |
Student > Master | 12 | 13% |
Student > Bachelor | 5 | 5% |
Lecturer | 5 | 5% |
Other | 12 | 13% |
Unknown | 23 | 25% |
Readers by discipline | Count | As % |
---|---|---|
Engineering | 17 | 19% |
Computer Science | 11 | 12% |
Neuroscience | 10 | 11% |
Medicine and Dentistry | 8 | 9% |
Psychology | 4 | 4% |
Other | 12 | 13% |
Unknown | 29 | 32% |