MS Lesions Detection in MRI Using Grouping Artificial Immune Networks

Younis, Akmal A.; Soliman, Ahmed; Kabuka, Mansur R.; John, Nigel

doi:10.1109/bibe.2007.4375704

Cited by 6 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When it comes to automatic methods, which on the other side aim to exclude human interaction (and consequently reduce the variability), the biggest obstacle is the validation of a certain method prior to its integration in clinical practice ( García-Lorenzo et al, 2013 ). Numerous studies used either synthetic data, which does not provide sufficient proof of robustness in real-world scenarios ( Bricq et al, 2008 , Forbes et al, 2010 , Shiee et al, 2010 , Younis et al, 2007 ). Others considered data that lacks in variability, mainly due to the fact that it comes from a single scanner ( Parry et al, 2002 , Vrenken et al, 2010 , Alfano et al, 2000 ).…”

Section: Discussionmentioning

confidence: 99%

icobrain ms 5.1: Combining unsupervised and supervised approaches for improving the detection of multiple sclerosis lesions

Rakić

Vercruyssen

Eyndhoven

et al. 2021

NeuroImage: Clinical

View full text Add to dashboard Cite

Highlights We present a method for multiple sclerosis lesion detection and segmentation. The method (ico brain ms 5.1) combines unsupervised and supervised learning techniques. We highlight the importance of training and validation on a highly diverse dataset. Unsupervised method ico brain ms 5.0 was used as a baseline to evaluate performance. The detection of infratentorial and juxtacortical lesions improves by 14% and 31%.

show abstract

Section: Discussionmentioning

confidence: 99%

icobrain ms 5.1: Combining unsupervised and supervised approaches for improving the detection of multiple sclerosis lesions

Rakić

Vercruyssen

Eyndhoven

et al. 2021

NeuroImage: Clinical

View full text Add to dashboard Cite

show abstract

“…Hadjiprocopis and Tofts (Hadjiprocopis and Tofts, 2003) included the Cartesian and polar coordinates of the voxel in addition to its intensity information. Younis et al (Younis et al, 2007) added the intensity of the six neighbor voxels into the ANN to include local spatial information in the segmentation, thus avoiding a noisy segmentation. They employed the ANN twice, first to classify the brain tissue in the T1w image and, after removing the CSF, to segment the T2w image and obtain the WML.…”

Section: Supervised Learning Methodsmentioning

confidence: 99%

“…This method prevents direct comparison of results with those of other methods and overestimates generalization, as the exact same image is used for both testing and training. o Use a noiseless image for training: The noiseless BrainWeb image was employed as the training data (Younis et al, 2007), then the same image with noise was used for testing. This validation also employs the same image to test and train, thus limiting the comparison of their results to those of other methods.…”

Section: Reduced Lesion Load Rangementioning

confidence: 99%

Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging

García-Lorenzo

Francis

Narayanan

et al. 2013

Medical Image Analysis

314

268

View full text Add to dashboard Cite

Magnetic resonance (MR) imaging is often used to characterize and quantify multiple sclerosis (MS) lesions in the brain and spinal cord. The number and volume of lesions have been used to evaluate MS disease burden, to track the progression of the disease and to evaluate the effect of new pharmaceuticals in clinical trials. Accurate identification of MS lesions in MR images is extremely difficult due to variability in lesion location, size and shape in addition to anatomical variability between subjects. Since manual segmentation requires expert knowledge, is time consuming and is subject to intra-and interexpert variability, many methods have been proposed to automatically segment lesions.The objective of this study was to carry out a systematic review of the literature to evaluate the state of the art in automated multiple sclerosis lesion segmentation. From 1240 hits found initially with PubMed and Google scholar, our selection criteria identified 80 papers that described an automatic lesion segmentation procedure applied to MS. Only 47 of these included quantitative validation with at least one realistic image. In this paper, we describe the complexity of lesion segmentation, classify the automatic MS lesion segmentation methods found, and review the validation methods applied in each of the papers reviewed. Although many segmentation solutions have been proposed, including some with promising results using MRI data obtained on small groups of patients, no single method is widely employed due to performance issues related to the high variability of MS lesion appearance and differences in image acquisition. The challenge remains to provide segmentation techniques that work in all cases regardless of the type of MS, duration of the disease, or MRI protocol, and this within a comprehensive, standardized validation framework. MS lesion segmentation remains an open problem.

show abstract

“…Zijdenbos et al (2002) used an MRI intensity feature space that was extended by the tissue probability of the given voxel based in a probabilistic tissue atlas. An alternative way to encode spatial information was proposed by Younis, Soliman, Kabuka, and John (2007), where local neighboring information was included by extending the voxel intensity feature by including the MRI intensity of the six neighboring voxels. An alternative way to encode spatial information was proposed by Younis, Soliman, Kabuka, and John (2007), where local neighboring information was included by extending the voxel intensity feature by including the MRI intensity of the six neighboring voxels.…”

Section: Automated Lesion Segmentation Algorithmsmentioning

confidence: 99%