Segmentation and detection of brain tumor through optimal selection of integrated features using transfer learning

Swaraja, K.; Meenakshi, K.; Valiveti, Hima Bindu; Karuna, G.

doi:10.1007/s11042-022-12414-0

Cited by 15 publications

(11 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several existing DL models used for segmentation of brain tumors need high computation, and the provided results are affected by the batch size. 37 Automatic CNN models, which have automatic segmentation, are suggested as a way to get around these problems.…”

Section: Discussionmentioning

confidence: 99%

A deep conventional neural network model for glioma tumor segmentation

Ayadi

Elhamzi

Atri

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Glioma represents one of the most aggressive cancers, which can develop in the brain. The automatic tumor segmentation and its sub‐regions represent a challenging task owing to their considerable structural variation. It can appear in different ways and with several shapes, which makes tissue identification a crucial task. Reliable and accurate segmentation presents an important component in tumor treatment and diagnosis planning. To overcome these drawbacks, various Deep Learning (DL) schemes are proposed to aid doctors. In this paper, a novel Convolutional Neural Network (CNN) scheme for glioma segmentation was proposed. Our suggested technique consists of three phases. First, we use intensity normalization to ameliorate the image quality as a preprocessing step. Second, an automatic segmentation technique based on CNN has been proposed. The new model has several layers. Finally, and with the goal of refining the segmentation results, we employ a post‐processing approach. We use the public benchmark BraTS datasets from 2018 and 2020 with low‐grade and high‐grade glioma tumors to test the suggested framework. These datasets contain about 285 and 369 patients, respectively. The four modalities are exploited. Each patient has about 155 2D images from every modality. All the images have the same size (240 × 240 pixels). Our technique performs well compared to new methods, with Dice scores of 0.88 for the Whole Tumor, 0.84 for Tumor Core, and 0.71 for Enhancing Tumor based on the first dataset. According to the second dataset, the three regions had an average of 0.88, 0.9, and 0.75, respectively. The Jaccard indexes for the first data set are 0.8, 0.73, and 0.56 for the three regions, respectively. The second data set attains 0.8, 0.82, and 0.6 for the three regions. The results show that the suggested framework is an excellent way to segment data, especially compared to other methods.

show abstract

Section: Discussionmentioning

confidence: 99%

A deep conventional neural network model for glioma tumor segmentation

Ayadi

Elhamzi

Atri

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…Swaraja [21] enhanced the accuracy through pre-trained techniques. However, this model was unable to augment certain features of brain images, therefore leading to misclassification.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The accuracy, precision, sensitivity, specificity, f1-score, Dice Score Coefficient (DSC), Structural Similarity Index (SSIM), Intersection-Over-Union (IoU) and Mean IoU (MIoU) are employed to estimate the GSAM based CNN performance which is mathematically shown in Eqs. (18)(19)(20)(21)(22)(23)(24)(25)(26).…”

Section: Experimental Analysismentioning

confidence: 99%

Global Self-Attention Module Based Convolutional Neural Network for Brain Tumor Classification

2024

IJIES

View full text Add to dashboard Cite

The early analysis of brain tumors plays a significant role in enhancing treatment and patient survival rates. The accurate diagnosis is critical to creating treatment plans that extend the lifetime of affected individuals. The existing methods have limitations like less accuracy and not able to learn global features. To overcome this issue, Global Self-Attention Module based Convolutional Neural Network (GSAM based CNN) is proposed for brain tumor classification. The Figshare dataset is preprocessed through data augmentation which is utilized to enhance the data size, and the median filter is employed to eliminate noise in an input image. The preprocessed images are given to Adaptive Kernel Fuzzy C Means (AKFCM) with Otsu thresholding for segmentation process. After segmentation, the Sine Cosine Reptile Search Algorithm (SCRSA) is employed for feature selection. Then, the features are provided to GSAM based CNN for brain tumor classification. The proposed model achieves better result on Figshare dataset on the metrics of accuracy, precision, sensitivity, specificity and f1-score, with values of 99.83%, 99.65%, 99.79%, 99.78% and 99.71%, correspondingly, when compared to the existing methods like Comprehensive Learning Elephant Herding Optimization based K-Nearest Neighbor (CLEHO based KNN) and Parallel Deep CNN (PDCNN) with data augmentation.

show abstract

“…Deepak and Ameer utilized TL to extract features from brain MRI images for the purpose of classifying brain tumors [ 9 ]. Swaraja et al proposed a framework that provides both brain tumor segmentation and classification, which can be efficiently processed on all four basic image pulse sequences [ 10 ]. Singh et al developed a modified TL method to improve brain tumor segmentation accuracy [ 11 ].…”

Section: Introductionmentioning

confidence: 99%

In-Domain Transfer Learning Strategy for Tumor Detection on Brain MRI

Terzi

Azginoglu

2023

Diagnostics

View full text Add to dashboard Cite

Transfer learning has gained importance in areas where there is a labeled data shortage. However, it is still controversial as to what extent natural image datasets as pre-training sources contribute scientifically to success in different fields, such as medical imaging. In this study, the effect of transfer learning for medical object detection was quantitatively compared using natural and medical image datasets. Within the scope of this study, transfer learning strategies based on five different weight initialization methods were discussed. A natural image dataset MS COCO and brain tumor dataset BraTS 2020 were used as the transfer learning source, and Gazi Brains 2020 was used for the target. Mask R-CNN was adopted as a deep learning architecture for its capability to effectively handle both object detection and segmentation tasks. The experimental results show that transfer learning from the medical image dataset was found to be 10% more successful and showed 24% better convergence performance than the MS COCO pre-trained model, although it contains fewer data. While the effect of data augmentation on the natural image pre-trained model was 5%, the same domain pre-trained model was measured as 2%. According to the most widely used object detection metric, transfer learning strategies using MS COCO weights and random weights showed the same object detection performance as data augmentation. The performance of the most effective strategies identified in the Mask R-CNN model was also tested with YOLOv8. Results showed that even if the amount of data is less than the natural dataset, in-domain transfer learning is more efficient than cross-domain transfer learning. Moreover, this study demonstrates the first use of the Gazi Brains 2020 dataset, which was generated to address the lack of labeled and qualified brain MRI data in the medical field for in-domain transfer learning. Thus, knowledge transfer was carried out from the deep neural network, which was trained with brain tumor data and tested on a different brain tumor dataset.

show abstract

Segmentation and detection of brain tumor through optimal selection of integrated features using transfer learning

Cited by 15 publications

References 54 publications

A deep conventional neural network model for glioma tumor segmentation

A deep conventional neural network model for glioma tumor segmentation

Global Self-Attention Module Based Convolutional Neural Network for Brain Tumor Classification

In-Domain Transfer Learning Strategy for Tumor Detection on Brain MRI

Contact Info

Product

Resources

About