Fast and Accurate Feature Extraction-Based Segmentation Framework for Spinal Cord Injury Severity Classification

Abstract

Detection of spinal cord injury (SCI) is one of the major problems in MRI images to detect the affected portion of spinal cord regions using feature sets. Automatic detection of spinal cord atrophy is complex due to change in structure, size, and white matter. Delineating gray matter and white matter are the essential factors that influence the detection of spinal cord atrophy and its severity. Automatic segmentation and classification are accurate methods for detecting the severity of the SCI. Hierarchical segmentation, partitioning segmentation, graph, and watershed segmentation methods are used to find the SCI segments in static fixed positions. Also, these segmentation models result in a high false positive rate due to over segmentation features and noise in the segmented regions. Furthermore, these classification methods fail to segment and detect the severity level in the affected region due to over segmentation. In order to overcome these issues, a novel segment-based classification model is required to find the severity of the injury and to predict the disease patterns on the over segmented regions and features. In the present model, a hybrid image threshold technique is used to segment the spinal cord regions for non-linear SVM classification approach. Among the traditional feature segmentation-based classification models, the proposed thresholdbased non-linear SVM has better accuracy for SCI detection. The results proved that the present model is more efficient than the earlier approaches in terms of true positive rate (TP = 0.9783) and accuracy (0.9683).

Spinal Cord MRI Segmentation Techniques and Algorithms: A Survey

“…Ahammad et al [40] designed a novel segmentationbased classification model. In this model the author used the hybrid image threshold-based segmentation technique for feature extraction.…”

A test architecture design for SoCs using ATAM method

“…All limitations as demonstrated in trouble 1, anyway as opposed to the amount of compass chains entire furthermore the span l,h in support of every yield series k, the whole integer of breadth flip disappointments is known. Choose a T-A-M building moreover a wrapper plan for each one element with the true objective to facilitate the general SOC-level test time (in clock cycles) is constrained as well as W max isn't outperformed [18][19].…”

Estimation of Quality and Intelligibility of a Speech Signal with varying forms of Additive Noise

“…Certain names for colours, such as purple, green, were then given to noise with other spectral profiles, often referring to the colour of light with different spectra The White noise is known as any type of audio signal given a constant spectral density of energy at different frequencies of equal frequency [8][9][10]. Inside of the frequency spectrum, pink noise is known as a signal, for which, if it's power levels of each frequency interval is increased, the signal frequency of it will show a decline and if the signal frequency is increased, the power levels will show a decline [8][9][10]. When many sound particles move at a different pace and also in random directions, we call that type of motion which is encountered by them as Brownian motion.…”

“…When many sound particles move at a different pace and also in random directions, we call that type of motion which is encountered by them as Brownian motion. Brown (or) red noise follows in steps of Brownian motion, so, its alternatively called as random walk noise [8][9][10]. Although the same signal is represented by these sounds, they have their own variations.…”

“…The noise disparity is illustrated are White Noise is widely distributed throughout the sound spectrum, including sounds of low frequency, medium, high frequency. Example: frequency ranges between 39 Hz-60 Hz contains same amount of sound power as the sound power range between 390 Hz and 410 Hz, as both intervals are 20 Hz wide [9][10][11][12][13][14].…”