MRI Type Identification

MRI sequence and orientation identification app

Brain Magnetic Resonance Imaging (MRI) scans are available in a wide variety of sequences, orientations, and magnet strengths. A necessary preprocessing step for any automated diagnosis is to identify the MRI sequence, orientation, and magnet strength of the acquired image. Automatic identification of the MRI sequence can be useful in labeling massive online datasets used by data scientists in the design and development of Computer Aided Diagnosis (CAD) tools.

This UI utilizes Deep Learning (DL) approach for brain MRI sequence and orientation identification using scans of different data types as input. A 12-class classification system is used for commonly used MRI scans, including T1, T2-weighted, Proton Density (PD), Fluid Attenuated Inversion Recovery (FLAIR) sequences in axial, coronal and sagittal orientations.

The tool has been tested on fresh data from online and hospital sources with a satisfactory performance accuracy of 99.84% & 86.49%, respectively.

The UI is extremely user friendly with a push button to input MRI. The MRI file types accepted by the tool include JPG, PNG, BMP, and DICOM. Digital Imaging and Communications in Medicine (DICOM) files are generally preferred by radiologists/ neurologists and are accessed using specialized software like RADIANT DICOM Viewer. After receiving the MRI file from user, the app displays the MRI image along with information including sequence, orientation, deep brain/ near skull image and time taken to process the image.