Clinical target segmentation using a novel deep neural network: double attention Res-U-Net

Abstract

We introduced Double Attention Res-U-Net architecture to address medical image segmentation problem in different medical imaging system. Accurate medical image segmentation suffers from some challenges including, difficulty of different interest object modeling, presence of noise, and signal dropout throughout the measurement. The base line image segmentation approaches are not sufficient for complex target segmentation throughout the various medical image types. To overcome the issues, a novel U-Net-based model proposed that consists of two consecutive networks with five and four encoding and decoding levels respectively. In each of networks, there are four residual blocks between the encoder-decoder path and skip connections that help the networks to tackle the vanishing gradient problem, followed by the multi-scale attention gates to generate richer contextual information. To evaluate our architecture, we investigated three distinct data-sets, (i.e., CVC-ClinicDB dataset, Multi-site MRI dataset, and a collected ultrasound dataset). The proposed algorithm achieved Dice and Jaccard coefficients of 95.79%, 91.62%, respectively for CRL, and 93.84% and 89.08% for fetal foot segmentation. Moreover, the proposed model outperformed the state-of-the-art U-Net based model on the external CVC-ClinicDB, and multi-site MRI datasets with Dice and Jaccard coefficients of 83%, 75.31% for CVC-ClinicDB, and 92.07% and 87.14% for multi-site MRI dataset, respectively.

Bionnica: A Deep Neural Network Architecture for Colorectal Polyps’ Premalignancy Risk Evaluation

“…They can group images into precise subcategories, which improves diagnostic precision. This architecture is used for object detection [36], particularly for the location of polyps in medical images [37], but also for bioimage Table 1 The performance evaluation metrics, obtained by authors from [27], applied on the CVC-ClinicDB.…”

“…The authors of [27] gained a 95.79% F1-score and 91.62% Jaccard Index (Table 1), while the authors of [30] received the F1-score of 91.57%, an accuracy of 98.82%, a sensitivity of 98.28%, and a specificity of 98.68% (Table 2), from a dataset [31] known as The Cancer Imaging Archive in the context of CT Colonography [32].…”

IRv2-Net: A Deep Learning Framework for Enhanced Polyp Segmentation Performance Integrating InceptionResNetV2 and UNet Architecture with Test Time Augmentation Techniques

“…However, contemporary methods increasingly make use of CNN and pre-trained networks. The Kvasir-SEG dataset [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 ], the CVC-ClinicDB dataset [ 24 , 25 , 26 , 27 , 28 , 29 , 30 ], and the TTA [ 31 , 32 , 33 , 34 , 35 , 36 , 37 ] are frequently used datasets in analyses of polyp detection and segmentation networks.…”

“…To solve the vanishing gradient problem and provide contextual data, Vahid et al [ 27 ] suggested a Res-UNet architectural model that used successive networks with multi-scale attention gates, residual blocks, and skip connections. The CVC-ClinicDB dataset showed remarkable segmentation findings with a Dice of 83% and Jaccard of 75.31%.…”

Deep Learning-based Non-rigid Image Registration for High-dose Rate Brachytherapy in Inter-fraction Cervical Cancer

“…Recently, deep learning approaches, including supervised, unsupervised, and semi-supervised methods, are employed in different medical image analysis tasks [18][19][20][21][22][23][24][25][26][27][28], different BT tasks [29][30][31], and learn and carry out spatial alignment/ transformation between images [32][33][34][35][36][37][38][39][40][41]. These methods usually used convolutional neural networks (CNNs) to extract informative features automatically to perform this task [32][33][34][35][36][37][38][39][40][41].…”