China
Africa
Explore chapters and articles related to this topic
Significant Advancements in Cancer Diagnosis Using Machine Learning
Published in Meenu Gupta, Rachna Jain, Arun Solanki, Fadi Al-Turjman, Cancer Prediction for Industrial IoT 4.0: A Machine Learning Perspective, 2021
Gurmanik Kaur, Ajat Shatru Arora
Mehmood et al. [36] created a semi-automated and adaptive threshold selection approach for brain MRIs. Following segmentation, the tumor was indeed categorized as malignant or benign, employing a robust SVM classification model driven using Bag of Words (BoW). The BoW methodology for feature extraction was also enhanced further by the use of Speeded Up Robust Features (SURF). The BoW feature extraction method was even further enhanced by SURF, which incorporated its process for selecting interest points. Ultimately, the volume marching cube algorithm, that was used to render medical information, was used to create 3D visualizations of the brain and tumor. The proposed system’s effectiveness was validated using a dataset of 30 patients, and it attained 99% accuracy. In addition, a comparative study was conducted between the developed method and two cutting-edge technologies, ITK-SNAP as well as 3D-Doctor. The findings indicate that the suggested framework outperformed current frameworks in terms of assisting radiologists in assessing the size, structure, and position of the brain tumor.
Level Set Methods for Cardiac Segmentation in MSCT Images
Published in Ayman El-Baz, Jasjit S. Suri, Level Set Method in Medical Imaging Segmentation, 2019
Ruben Medina, Sebastian Bautista, Villie Morocho, Alexandra La Cruz
Other approaches are based on level set contour deformation techniques. In Lynch et al. [14] a cardiac motion model based on the volume is used in a level set framework for performing the 4-D segmentation in MRI data. A user–guided 3-D active contour segmentation approach based on level set has been developed by Yushkevich et al. [15] as an open–software platform known as ITK–SNAP. This platform was initially conceived for neuroimaging modalities but their application can be extended to cardiac images. A deformable contour based approach has been used by Heiberg et al. [16] for developing a software platform that is freely available for non–commercial research. A level set approach for left ventricle segmentation in echocardiographic images has been proposed by Dydenko et al. [17] where a prior shape for the left ventricle is registered to the image, and a level set approach is used for performing the contour deformation that allows the recovering of the ventricle shape.
Quantitative imaging using CT
Published in Ruijiang Li, Lei Xing, Sandy Napel, Daniel L. Rubin, Radiomics and Radiogenomics, 2019
Lin Lu, Lawrence H. Schwartz, Binsheng Zhao
3D Slicer and ITK-SNAP are two popular segmentation software packages that are publicly available. 3D Slicer is an open source software available for multiple operating systems, including Linux, MacOSX, and Windows. It is a platform that provides image analysis (e.g., registration and interactive segmentation) and visualization (e.g., volume rendering) of medical images including CT, MRI, and positron emission tomography (PET). ITK-SNAP [43] is an easy-to-use software tool that provides semi-automatic segmentation using active contour methods, as well as manual delineation and image navigation.
Comparison of machine learning methods for the detection of focal cortical dysplasia lesions: decision tree, support vector machine and artificial neural network
Published in Neurological Research, 2022
Zohreh Ganji, Mohsen Aghaee Hakak, Hoda Zare
In the preprocessing stage, the image quality is improved for subsequent processing to make it easier and faster to execute. During this stage, depending on the type of processing, several steps are performed, including removing the skull, removing the noise and bias field correction. In this study, Freesurfer software (version 6.0; Athinoula A. Martinos Center for Biomedical Imaging at Massachusetts General Hospital, Boston USA) [14–16] was used for pre-processing and processing stages. Cortical reconstruction performed by the Recon-all pipeline had several steps as follows: transferring raw image voxels to the isotropic space, normalizing the images for bias field correction, skull stripping, automatic subcortical segmentation, white matter segmentation, and determining the WM/GM interface. Then, two experienced neurologists labeled FCD lesions on reconstructed images using ITK-SNAP software (version 3.6.0; Penn Image Computing and Science Laboratory – PICSL -, and Scientific Computing and Imaging Institute – SCI, USA).
Cross-sectional changes of the distal carpal tunnel with simulated carpal bone rotation
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2022
Ten cadaveric hands were used for the study. The palmar surface of each specimen was dissected to expose the TCL. The carpal tunnel contents were evacuated, leaving only the carpal bones and intact TCL. A medical balloon was inserted into the tunnel and pressurized to 10 mm Hg using a solution of water and computed tomography contrast agent. A clinical computed tomography scanner (InReach, CurveBeam, Hatfield, PA) was used to scan each specimen from the distal radius to the distal phalanges. The DICOM image files were uploaded into an image segmentation software platform ITK-SNAP (Yushkevich et al. 2006), where a single distal slice containing the hook of hamate and ridge of trapezium was identified for analysis (Figure 1). In Figure 1, the total carpal tunnel cross-sectional area is outlined in black.
The use of augmented reality in transsphenoidal surgery: A systematic review
Published in British Journal of Neurosurgery, 2022
Santhosh G. Thavarajasingam, Robert Vardanyan, Arian Arjomandi Rad, Ahkash Thavarajasingam, Artur Khachikyan, Nigel Mendoza, Ramesh Nair, Peter Vajkoczy
Four of the included studies, all non-in vivo studies, utilised manual segmentation as the sole segmentation technique, employing a variety of software.29,31,33,38 Dixon et al.29 and Dixon et al.31 used the ITK-SNAP 2.0 software,38 whereas Lai et al. used a self-developed AR surgical navigation system. Four studies used purely automated segmentation,23,25,26,32 with Lapeer et al. using the ARView system to achieve this.26 Two studies employed a combined automated and manual segmentation.24,39 Using the Brainlab software, Carl et al. used manual segmentation for vascular risk structures, but automated segmentation to visualise nerve risk structures. Being the only studies to employ each, Pennacchietti et al. used complex pixel volume segmentation and Dixon et al.30 used contour-based segmentation.