A multi-phase semi-automatic approach for multisequence brain tumor image segmentation

Lim, Khai Yin; Rajeswari, M.

doi:10.1016/j.eswa.2018.06.041

Cited by 31 publications

(14 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Panigrahi et al 110 (2018) used in 2018 a fuzzy c-means clustering combined with multi-scale vector field convolution in order to segment breast tumors in ultrasound images. Lim & Mandava (2018) suggested a semi-automatic approach for segmentation of BRATS images but only with an average Dice accuracy of 0.7. Considering MRI brain images have more contrasts and less noise than CT-Scans, this 115 approach does not seem relevant in our context.…”

Section: Related Workmentioning

confidence: 99%

Segmentation of deformed kidneys and nephroblastoma using Case-Based Reasoning and Convolutional Neural Network

Florent

Corbat

Chaussy

et al. 2019

Expert Systems with Applications

View full text Add to dashboard Cite

Most often, image segmentation is not fully automated and a user is required to lead the process in order to obtain correct results. In a medical context, segmentation can furnish a lot of information to surgeons, but this task is rarely executed. Artificial Intelligence (AI) is a powerful approach for devising a viable solution to fully automated treatment. In this paper, we have focused on kidneys deformed by nephroblastoma. Yet, a frequent medical constraint is encountered which is a lack of data with which to train our system. In function of this constraint, two AI approaches were used to segment these structures.First, a Case Based Reasoning (CBR) approach was defined which can enhance the growth of regions for segmentation of deformed kidneys with an adaptation phase to modify coordinates of recovered seeds. This CBR approach was confronted with manual region growing and a Convolutional Neural Network (CNN). The CBR system succeeded in performing the best segmentation for the kidney with a mean Dice of 0.83. Deep Learning was then examined as a possible solution, using the latest performing networks for image segmenta- tion. However, for relevant efficiency, this method requires a large data set. An option would be to manually segment only certain representative slices from a patient and, using them, to train a Convolutional Neural Network how to segment. In this article the authors propose an evaluation of a CNN for medical image segmentation following different training sets with avariable number of manual segmentations. To choose slices to train the CNN, an Overlearning Vector for Valid Sparsed SegmentatIONs (OV 2 ASSION) was used, with the notion of gap between two slices from the training set. This protocol made it possible to obtain reliable segmentations of tumor per patient with a low data set and to determine that only 26% of initial segmented slices are required to obtain a complete segmentation of a patient with a mean Dice of 0.897.

show abstract

Section: Related Workmentioning

confidence: 99%

Segmentation of deformed kidneys and nephroblastoma using Case-Based Reasoning and Convolutional Neural Network

Florent

Corbat

Chaussy

et al. 2019

Expert Systems with Applications

View full text Add to dashboard Cite

show abstract

“…Khai Yin Lim, et.al [19] discussed the approach that includes three important phases. The improved random walk algorithm was used for segmentation in the initial step.…”

Section: E Classificationmentioning

confidence: 99%

Localization and Classification of Brain Tumor using Machine Learning & Deep Learning Techniques

2019

IJITEE

View full text Add to dashboard Cite

Digital image processing is a rising field for the investigation of complicated diseases such as brain tumor, breast cancer, kidney stones, lung cancer, ovarian cancer, and cervix cancer and so on. The recognition of the brain tumor is considered to be a very critical task. A number of approaches are used for the scanning of a particular body part like CT scan, X-rays, and Magnetic Resonance Image (MRI). These pictures are then examined by the surgeons for the removal of the problem. The main objective of examining these MRI images (mainly) is to extract the meaningful information with high accuracy. Machine Learning and Deep Learning algorithms are mainly used for analysing the medical images which can identify, localize and classify the brain tumor into sub categories, according to which the diagnosis would be done by the professionals. In this paper, we have discussed the different techniques that are used for tumor pre-processing, segmentation, localization, extraction of features and classification and summarize more than 30 contributions to this field. Also, we discussed the existing state-of-the-art, literature gaps, open challenges and future scope in this area.

show abstract

“…In order to verify the rapidity of the algorithm, this paper compares with the following algorithms, e.g. BMAP [13], FCM [37], MAP [15], MLC [21], TSKMEANS [28], FLGMN [31]. The training is aimed at gray matter and white matter in brain MR images, and the training speed is shown as Figure 9.…”

Section: Comparison Of Training Coefficientmentioning

confidence: 99%

“…Furthermore, Thchumperla proposes a directional diffusion filtering method with curvature preservation ability based on line integral convolution, which achieves effective curvature preservation. Hermosillo, Weickert and other researchers use diffusion filter function to define the regular terms in the registration model, and achieve the effective preservation of image features in the registration process [21]. Snake model is a classical application of partial differential equation in image segmentation [22].…”

Section: Introductionmentioning

confidence: 99%

Application of Partial Differential Equation in Brain Image Segmentation and Registration

Tian¹,

Zhang²

2019

IEEE Access

View full text Add to dashboard Cite

Due to the problem that local binary fitting model is easy to fall into local extremum, based on the hypothesis that the brightness of the image on both sides of the target boundary has contrast, an improved method of adding contrast constraints to the fitting function is proposed in this paper. The contrast-constrained model can effectively overcome the problem that energy is easily trapped in local extremum, but there is still a problem that requires the initialization curve to be close to the edge of the target. Therefore, balloon force is added to the contrast-constrained model, which enlarges the scope of external force and enhances the robustness of the model to initialization. Meanwhile, the diffusion filtering method with adaptive image structure is introduced into image registration, and the anisotropic diffusion function with the ability of feature preservation and consistency enhancement is defined as the regularization term of the model. The improved model can effectively maintain image features and achieve accurate registration of complex images. In this paper, brain image (MRI) in Mcgill database is taken as an example to verify the simulation results. The simulation results verify the reliability and validity of the partial differential equation-based algorithm. INDEX TERMS Brain image segmentation, partial differential equation (PDE), LBF model, diffusion filtering method, image registration.

show abstract

A multi-phase semi-automatic approach for multisequence brain tumor image segmentation

Cited by 31 publications

References 24 publications

Segmentation of deformed kidneys and nephroblastoma using Case-Based Reasoning and Convolutional Neural Network

Segmentation of deformed kidneys and nephroblastoma using Case-Based Reasoning and Convolutional Neural Network

Localization and Classification of Brain Tumor using Machine Learning & Deep Learning Techniques

Application of Partial Differential Equation in Brain Image Segmentation and Registration

Contact Info

Product

Resources

About