Automatic classification of computed tomography brain images using ANN, k-NN and SVM

Automated Detection and Screening of Traumatic Brain Injury (TBI) Using Computed Tomography Images: A Comprehensive Review and Future Perspectives

“…Pre-processing is employed to remove the irrelevant information in the brain CT images, such as from the skull, head holder, soft tissue edema, and background, which can significantly introduce noise and degrade the performance of the CAD system. Techniques, such as intensity based thresholding, morphological operation, and connected component analysis, are used individually or combined to obtain the enhanced CT images [23,44,[49][50][51]. Commonly used noise reduction techniques include median filtering and gradient magnitude filtering [23,44,50,51].…”

“…Techniques, such as intensity based thresholding, morphological operation, and connected component analysis, are used individually or combined to obtain the enhanced CT images [23,44,[49][50][51]. Commonly used noise reduction techniques include median filtering and gradient magnitude filtering [23,44,50,51]. Contrast limited adaptive histogram equalisation (CLAHE) is also utilised as a technique for enhancing image quality [51].…”

“…A detailed description of various stages involved in CAD systems is furnished in subsequent sections. Figure 5 shows the schema of a typical feature learning-based approach for TBI [ 18 , 23 , 41 , 42 , 43 , 44 ].…”

“…Texture feature extraction is widely used as a feature extraction technique, and the inclusion of shape features [ 67 ] and statistical features have proven useful for improved diagnosis of TBI [ 44 , 51 , 57 , 68 , 69 ]. Raghavendra et al [ 70 ] extracted a set of nonlinear features based on entropy to detect the presence of intracranial hematoma in CT images and obtained an accuracy of 97.37%.…”

“…Tong et al [ 51 ] presented a midline formation technique to diagnose hematoma, in which LBP texture features and histogram features of both hemispheres were extracted and compared, and a recall rate of 84.86 was achieved. Rajini and Bhavani [ 44 ] proposed a model based on DWT featuring for the diagnosis of hematoma in TBI patients. Li et al [ 56 ] concluded that the distance transform with respect to five different landmarks, along with the Bayesian classifier, can distinguish normal versus subarachnoid hematoma (SAH).…”