Due to its irregular shape and varying contour, pancreas segmentation is a recognized challenge in medical image segmentation. Convolutional neural network (CNN) and Transformer-based networks perform well but have limitations: CNN have constrained receptive fields, and Transformer underutilize image features. This work proposes an improved pancreas segmentation method by combining CNN and Transformer. Point-wise separable convolution was introduced in a stage-wise encoder to extract more features with fewer parameters. A densely connected ensemble decoder enabled multi-scale feature fusion, addressing the structural constraints of skip connections. Consistency terms and contrastive loss were integrated into deep supervision to ensure model accuracy. Extensive experiments on the Changhai and National Institute of Health (NIH) pancreas datasets achieved the highest Dice similarity coefficient (DSC) values of 76.32% and 86.78%, with superiority in other metrics. Ablation studies validated each component’s contributions to performance and parameter reduction. Results demonstrate that the proposed loss function smooths training and optimizes performance. Overall, the method outperforms other advanced methods, enhances pancreas segmentation performance, supports physician diagnosis, and provides a reliable reference for future research.
In order to overcome the difficulty in lung parenchymal segmentation due to the factors such as lung disease and bronchial interference, a segmentation algorithm for three-dimensional lung parenchymal is presented based on the integration of surfacelet transform and pulse coupled neural network (PCNN). First, the three-dimensional computed tomography of lungs is decomposed into surfacelet transform domain to obtain multi-scale and multi-directional sub-band information. The edge features are then enhanced by filtering sub-band coefficients using local modified Laplacian operator. Second, surfacelet inverse transform is implemented and the reconstructed image is fed back to the input of PCNN. Finally, iteration process of the PCNN is carried out to obtain final segmentation result. The proposed algorithm is validated on the samples of public dataset. The experimental results demonstrate that the proposed algorithm has superior performance over that of the three-dimensional surfacelet transform edge detection algorithm, the three-dimensional region growing algorithm, and the three-dimensional U-NET algorithm. It can effectively suppress the interference coming from lung lesions and bronchial, and obtain a complete structure of lung parenchyma.
Heart sound segmentation is a key step before heart sound classification. It refers to the processing of the acquired heart sound signal that separates the cardiac cycle into systolic and diastolic, etc. To solve the accuracy limitation of heart sound segmentation without relying on electrocardiogram, an algorithm based on the duration hidden Markov model (DHMM) was proposed. Firstly, the heart sound samples were positionally labeled. Then autocorrelation estimation method was used to estimate cardiac cycle duration, and Gaussian mixture distribution was used to model the duration of sample-state. Next, the hidden Markov model (HMM) was optimized in the training set and the DHMM was established. Finally, the Viterbi algorithm was used to track back the state of heart sounds to obtain S1, systole, S2 and diastole. 500 heart sound samples were used to test the performance of our algorithm. The average evaluation accuracy score (F1) was 0.933, the average sensitivity was 0.930, and the average accuracy rate was 0.936. Compared with other algorithms, the performance of our algorithm was more superior. It is proved that the algorithm has high robustness and anti-noise performance, which might provide a novel method for the feature extraction and analysis of heart sound signals collected in clinical environments.
Objective To develop a neural network architecture based on deep learning to assist knee CT images automatic segmentation, and validate its accuracy. Methods A knee CT scans database was established, and the bony structure was manually annotated. A deep learning neural network architecture was developed independently, and the labeled database was used to train and test the neural network. Metrics of Dice coefficient, average surface distance (ASD), and Hausdorff distance (HD) were calculated to evaluate the accuracy of the neural network. The time of automatic segmentation and manual segmentation was compared. Five orthopedic experts were invited to score the automatic and manual segmentation results using Likert scale and the scores of the two methods were compared. Results The automatic segmentation achieved a high accuracy. The Dice coefficient, ASD, and HD of the femur were 0.953±0.037, (0.076±0.048) mm, and (3.101±0.726) mm, respectively; and those of the tibia were 0.950±0.092, (0.083±0.101) mm, and (2.984±0.740) mm, respectively. The time of automatic segmentation was significantly shorter than that of manual segmentation [(2.46±0.45) minutes vs. (64.73±17.07) minutes; t=36.474, P<0.001). The clinical scores of the femur were 4.3±0.3 in the automatic segmentation group and 4.4±0.2 in the manual segmentation group, and the scores of the tibia were 4.5±0.2 and 4.5±0.3, respectively. There was no significant difference between the two groups (t=1.753, P=0.085; t=0.318, P=0.752). Conclusion The automatic segmentation of knee CT images based on deep learning has high accuracy and can achieve rapid segmentation and three-dimensional reconstruction. This method will promote the development of new technology-assisted techniques in total knee arthroplasty.
Precise segmentation of lung field is a crucial step in chest radiographic computer-aided diagnosis system. With the development of deep learning, fully convolutional network based models for lung field segmentation have achieved great effect but are poor at accurate identification of the boundary and preserving lung field consistency. To solve this problem, this paper proposed a lung segmentation algorithm based on non-local attention and multi-task learning. Firstly, an encoder-decoder convolutional network based on residual connection was used to extract multi-scale context and predict the boundary of lung. Secondly, a non-local attention mechanism to capture the long-range dependencies between pixels in the boundary regions and global context was proposed to enrich feature of inconsistent region. Thirdly, a multi-task learning to predict lung field based on the enriched feature was conducted. Finally, experiments to evaluate this algorithm were performed on JSRT and Montgomery dataset. The maximum improvement of Dice coefficient and accuracy were 1.99% and 2.27%, respectively, comparing with other representative algorithms. Results show that by enhancing the attention of boundary, this algorithm can improve the accuracy and reduce false segmentation.
Objective To investigate an artificial intelligence (AI) automatic segmentation and modeling method for knee joints, aiming to improve the efficiency of knee joint modeling. Methods Knee CT images of 3 volunteers were randomly selected. AI automatic segmentation and manual segmentation of images and modeling were performed in Mimics software. The AI-automated modeling time was recorded. The anatomical landmarks of the distal femur and proximal tibia were selected with reference to previous literature, and the indexes related to the surgical design were calculated. Pearson correlation coefficient (r) was used to judge the correlation of the modeling results of the two methods; the consistency of the modeling results of the two methods were analyzed by DICE coefficient. Results The three-dimensional model of the knee joint was successfully constructed by both automatic modeling and manual modeling. The time required for AI to reconstruct each knee model was 10.45, 9.50, and 10.20 minutes, respectively, which was shorter than the manual modeling [(64.73±17.07) minutes] in the previous literature. Pearson correlation analysis showed that there was a strong correlation between the models generated by manual and automatic segmentation (r=0.999, P<0.001). The DICE coefficients of the 3 knee models were 0.990, 0.996, and 0.944 for the femur and 0.943, 0.978, and 0.981 for the tibia, respectively, verifying a high degree of consistency between automatic modeling and manual modeling. Conclusion The AI segmentation method in Mimics software can be used to quickly reconstruct a valid knee model.
Aiming at the problems of low accuracy and large difference of segmentation boundary distance in anterior cruciate ligament (ACL) image segmentation of knee joint, this paper proposes an ACL image segmentation model by fusing dilated convolution and residual hybrid attention U-shaped network (DRH-UNet). The proposed model builds upon the U-shaped network (U-Net) by incorporating dilated convolutions to expand the receptive field, enabling a better understanding of the contextual relationships within the image. Additionally, a residual hybrid attention block is designed in the skip connections to enhance the expression of critical features in key regions and reduce the semantic gap, thereby improving the representation capability for the ACL area. This study constructs an enhanced annotated ACL dataset based on the publicly available Magnetic Resonance Imaging Network (MRNet) dataset. The proposed method is validated on this dataset, and the experimental results demonstrate that the DRH-UNet model achieves a Dice similarity coefficient (DSC) of (88.01±1.57)% and a Hausdorff distance (HD) of 5.16±0.85, outperforming other ACL segmentation methods. The proposed approach further enhances the segmentation accuracy of ACL, providing valuable assistance for subsequent clinical diagnosis by physicians.
In this paper, we propose a new active contour algorithm, i.e. hierarchical contextual active contour (HCAC), and apply it to automatic liver segmentation from three-dimensional CT (3D-CT) images. HCAC is a learning-based method and can be divided into two stages. At the first stage, i.e. the training stage, given a set of abdominal 3D-CT training images and the corresponding manual liver labels, we tried to establish a mapping between automatic segmentations (in each round) and manual reference segmentations via context features, and obtained a series of self-correcting classifiers. At the second stage, i.e. the segmentation stage, we firstly used the basic active contour to segment the image and subsequently used the contextual active contour (CAC) iteratively, which combines the image information and the current shape model, to improve the segmentation result. The current shape model is produced by the corresponding self-correcting classifier (the input is the previous automatic segmentation result). The proposed method was evaluated on the datasets of MICCAI 2007 liver segmentation challenge. The experimental results showed that we would get more and more accurate segmentation results by the iterative steps and the satisfied results would be obtained after about six rounds of iterations.
Objective To explore the histochemical staining for distinguishing and local izing nerve fibers and fascicles at histological level in three-dimensional reconstruction of peri pheral nerves. Methods The right median nerve was harvested from one fresh cadaver and embedded in OCT compound. The sample was serially horizontally sl iced with 6 μm thickness. All sections were stained with Karnovsky-Roots method (group A, n=30) firstly and then stained with toluidine blue (group B, =28) and Ponceau 2R (group C, n=21) in proper sequence. The results of each step were taken photos (× 100). After successfully stitching, the two-dimensional panorama images were compared, including texture feature, the number and aver gray level of area showing acetylchol inesterase (AchE) activity, and result of auto microscopic medical image segmentation. Results In groups A, B, and C, the number of AchE-positive area was (21.63 ± 4.06)× 102, (20.64 ± 3.51)× 102, and (20.54 ± 5.71)× 102, respectively, showing no significant difference among 3 groups (F=0.64, P=0.54); the mean gray level was (1.41 ± 0.06)× 102, (1.10 ± 0.05)× 102, and (1.14 ± 0.07)× 102, respectively, showing significant differences between group A and groups B and C (P lt; 0.001). In the image of group A, only AchE-positive area was stained; in the image of group B, myelin sheath was obscure; and in the image of group C, axons and myelin sheath could be indentified, the character of nerve fibers could be distinguished clearly and accurately, and the image segmentation of fascicles could be achieved easier than other 2 images. Conclusion The image of Karnovsky-Roots-toluidine blue-Ponceau 2R staining has no effect on the AchE-positive area in the image of Karnovsky-Roots staining and shows better texture feature. This improved histochemical process may provide ideal image for the three-dimensional reconstruction of peri pheral nerves.
As one of the non-invasive imaging techniques, myocardial perfusion imaging provides a basis for the diagnosis of myocardial ischemia in coronary heart disease. Aiming at the bull-eye image in myocardial perfusion imaging, this paper proposed a branching structure, which included multi-layer transposed convolution up-sampling concatenate module and four-channel weighted channels attention module, and the output results of the branch structure were fused with the output results of trunk U-Net, to achieve accurate segmentation of the cardiac ischemia missing degree in myocardial perfusion bull-eye image. The experimental results show that the multi-layer transposed convolution up-sampling concatenate module realizes the fusion of different depth feature maps, and effectively reduces the interference of the severe sparse degree which is similar to the missing degree on the segmentation. Four-channel weighted attention module can further improve the ability to distinguish between the two similar degrees and the ability to learn edge details of the targets, and retain more abundant edge details features. The experimental data came from Tianjin Medical University General Hospital, Tianjin TEDA Hospital, Tianjin First Central Hospital and Third Central Hospital. The Jaccard scores in the self-built dataset was 5.00% higher than that of U-Net. The model presented in this paper is superior to other optimized models based on U-Net, and the subjective evaluation meets the accuracy requirements for clinical diagnosis.
