Magnetic Resonance Imaging Classification Methods: A Review
Magnetic resonance imaging (MRI) is one of the most important medical diagnosis methods in the field of computer-aided detection of medical images. The MRI images help to find the presence of abnormal cells or tissues, referred as tumors. Prior to classif
- PDF / 732,628 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 6 Downloads / 237 Views
Abstract Magnetic resonance imaging (MRI) is one of the most important medical diagnosis methods in the field of computer-aided detection of medical images. The MRI images help to find the presence of abnormal cells or tissues, referred as tumors. Prior to classification, preprocessing of MRI images is performed. These preprocessing operations help to reduce undesired distortions and select only relevant features for further analysis, making the classification technique more accurate and efficient. In this paper, we present various image classification techniques used over MRI images and their performance. Keywords Magnetic resonance imaging (MRI) · Image processing · Supervised image classification · Unsupervised image classification · Artificial neural networks (ANN)
1 1. Introduction A. Medical Resonance Imaging MRI is a diagnosis tool for capturing images of internal body parts and tissues in detail [1]. These images are used for the diagnosis of various diseases. MRI image processing is the most challenging and innovative field. There are other techniques available for medical imaging, namely mammography, CT scan, and x-ray, but the MRI imaging outperforms these techniques in quality and detailing. It is best suited N. Pateria (B) · D. Kumar · S. Kumar Department of Computer Science & Engineering, NIT Jamshedpur, Jamshedpur, India e-mail: [email protected] D. Kumar e-mail: [email protected] S. Kumar e-mail: [email protected] © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 V. Nath and J. K. Mandal (eds.), Nanoelectronics, Circuits and Communication Systems, Lecture Notes in Electrical Engineering 692, https://doi.org/10.1007/978-981-15-7486-3_38
417
418
N. Pateria et al.
for soft tissue abnormality detection. MRI scan is useful in diagnosing brain tumors, brain injuries, diseases like Alzheimer’s, epilepsy, arthritis, bone infection, breast cancer, prostate, etc. [2]. Traditionally, human inspection was the only method used for the classification of MRI images. This method was impractical as MRI images contain noise due to improper operator assistance, and this noise leads to inaccuracies in classification of images. The MRI image data is large and complex. For appropriate diagnosis of MRI images, we have to preprocess these images and extract only relevant features by applying various feature reduction and feature selection techniques. In some emergency cases, diagnosis with wrong result or delay in accurate diagnosis may be fatal. MR images play an important role in human brain research [3]. MR imaging comes under clinical radiology and is one of the most effective medical imaging techniques. The classification of MR images can be done by using supervised techniques as well as unsupervised techniques. B. MRI Image Processing In MRI image processing, mainly the following steps are essential: • • • • •
Image acquisition Image preprocessing Image feature extraction Image feature reduction Image classification
Image acqui
Data Loading...
