Brain MR images segmentation using 3D CNN with features recalibration mechanism for segmented CT generation

Mecheter, Imene; Abbod, Maysam F.; Zaidi, Habib; Amira, Abbes

doi:10.1016/j.neucom.2022.03.039

Cited by 8 publications

(6 citation statements)

References 37 publications

(53 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The novel idea arises from sequentially combining the two excitation blocks to improve segmentation performance and reduce model complexity. This segmentation network architecture shows promising and competitive results compared to other methods in the literature and reduces the complexity of the model thanks to the sequential combination of the two excitation blocks [87]. For adversarial learning, a novel asymmetric semi-supervised GAN (ASSGAN) with two generators and a discriminator is suggested.…”

Section: Neurosciencementioning

confidence: 97%

Image Segmentation Using Hybrid Optimization Algorithms: Review

Hemeida¹,

Yahya

Al-Sanary

2022

Aswan University Journal of Sciences and Technology

View full text Add to dashboard Cite

Image segmentation, often based on the properties of image pixels, is a widely used method in digital image processing and analysis to divide an image into multiple parts or areas. The main goal of image segmentation is to simplify the image for easier analysis. It would be very difficult to implement computer vision without performing image segmentation. An important component of computer vision is image segmentation, which has numerous commercial applications. Google and others Image-based search engines use image segmentation algorithms to identify the objects in your image and match their results to the relevant images they find to present you with search results.

show abstract

Section: Neurosciencementioning

confidence: 97%

Image Segmentation Using Hybrid Optimization Algorithms: Review

Hemeida¹,

Yahya

Al-Sanary

2022

Aswan University Journal of Sciences and Technology

View full text Add to dashboard Cite

show abstract

“…Later, the SVM classifier was replaced by CNN architectures. Besides this, on CT images some other segmentation works exist that utilize the power of deep models on brain image slices 13 …”

Section: Introductionmentioning

confidence: 99%

“…Besides this, on CT images some other segmentation works exist that utilize the power of deep models on brain image slices. 13 Convolution neural network (CNN) was not only limited to a single application but also used for several other medical problems like segmentation of brain glioma tumors 14 and Human-Computer Interaction for humans with disability. 15 Using CNN, some experiments were performed on CT images for the binary classification of acute hemorrhage images.…”

Section: Introductionmentioning

confidence: 99%

Stacked deep model‐based classification of the multiclass brain hemorrhages in CT scans

Malik,

Vidyarthi

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Intracranial hemorrhage (ICH) brain hemorrhages are life‐threatening medical conditions that require a prompt and accurate diagnosis to ensure timely intervention and treatment. Computed tomography (CT) scans play a vital role in diagnosing brain hemorrhages due to their ability to provide detailed cross‐sectional images of the brain. However, the manual interpretation of CT scans is time‐consuming and subjects to human error. To address these challenges, this study proposes a stacked deep model‐based classification approach for the automated identification and differentiation of multiclass brain hemorrhages in CT scans. The stacked deep model architecture consists of multiple layers of neural networks, each designed to learn specific levels of features, leading to enhanced representation learning and improved classification performance. Later, the stacked models are connected with the weighted approach that helps to get the ensemble models in the proper order based on their power of predicting correct samples. To build the stacked deep model, a large dataset of 7,52 000 annotated CT scans is utilized to train the model and optimize its parameters. Furthermore, transfer learning is employed to leverage pretrained models for feature extraction, which aids in handling limited data scenarios. Experimental results demonstrate that the stacked deep model achieves superior accuracy (98.56%), and other classification performance measures such as sensitivity (95.89%), specificity (99.58%), AUC‐ROC (98.47%), and weighted log loss (0.04967) as compared to conventional methods. The automated classification of multiclass brain hemorrhages in CT scans using the proposed model not only significantly reduces the time required for diagnosis but also enhances the reliability and consistency of results. This model has the potential to assist medical professionals in making more informed decisions and improve patient outcomes by enabling rapid and accurate diagnosis of brain hemorrhages.

show abstract

“…The segmentation of magnetic resonance (MR) images has various applications in the process of disease's diagnosis [1], treatment planning [2], and quantification of image-derived metrics [3]. One of these applications is the generation of pseudo computed tomography (CT) images for positron emission tomography (PET) attenuation correction [4].…”

Section: Introductionmentioning

confidence: 99%

“…One of the limitations of the segmentation methods is the accurate delineation of bone tissue which is the objective of the proposed method herein. https://doi.org/10.1016/j.artmed.2022.102365 Received 28 November 2021; Received in revised form 28 June 2022; Accepted 9 July 2022 Convolutional neural networks (CNNs) have been widely applied to segment different medical images including MR images [4,9]. The process of automatic extraction of features using CNN has shown its superiority in various applications to perform different tasks.…”

Section: Introductionmentioning

confidence: 99%