Objective To propose an innovative self-supervised learning method for vascular segmentation in computed tomography angiography (CTA) images by integrating feature reconstruction with masked autoencoding. Methods A 3D masked autoencoder-based framework was developed, where in 3D histogram of oriented gradients (HOG) was utilized for multi-scale vascular feature extraction. During pre-training, random masking was applied to local patches of CTA images, and the model was trained to jointly reconstruct original voxels and HOG features of masked regions. The pre-trained model was further fine-tuned on two annotated datasets for clinical-level vessel segmentation. Results Evaluated on two independent datasets (30 labeled CTA images each), our method achieved superior segmentation accuracy to the supervised neural network U-Net (nnU-Net) baseline, with Dice similarity coefficients of 91.2% vs. 89.7% (aorta) and 84.8% vs. 83.2% (coronary arteries). Conclusion The proposed self-supervised model significantly reduces manual annotation costs without compromising segmentation precision, showing substantial potential for enhancing clinical workflows in vascular disease management.
Magnetic resonance imaging (MRI) plays a crucial role in the diagnosis of ischemic stroke. Accurate segmentation of the infarct is of great significance for selecting intervention treatment methods and evaluating the prognosis of patients. To address the issue of poor segmentation accuracy of existing methods for multiscale stroke lesions, a novel encoder-decoder architecture network based on depthwise separable convolution is proposed. Firstly, this network replaces the convolutional layer modules of the U-Net with redesigned depthwise separable convolution modules. Secondly, an modified Atrous spatial pyramid pooling (MASPP) is introduced to enlarge the receptive field and enhance the extraction of multiscale features. Thirdly, an attention gate (AG) structure is incorporated at the skip connections of the network to further enhance the segmentation accuracy of multiscale targets. Finally, Experimental evaluations are conducted using the ischemic stroke lesion segmentation 2022 challenge (ISLES2022) dataset. The proposed algorithm in this paper achieves Dice similarity coefficient (DSC), Hausdorff distance (HD), sensitivity (SEN), and precision (PRE) scores of 0.816 5, 3.668 1, 0.889 2, and 0.894 6, respectively, outperforming other mainstream segmentation algorithms. The experimental results demonstrate that the method in this paper effectively improves the segmentation of infarct lesions, and is expected to provide a reliable support for clinical diagnosis and treatment.
The detection of electrocardiogram (ECG) characteristic wave is the basis of cardiovascular disease analysis and heart rate variability analysis. In order to solve the problems of low detection accuracy and poor real-time performance of ECG signal in the state of motion, this paper proposes a detection algorithm based on segmentation energy and stationary wavelet transform (SWT). Firstly, the energy of ECG signal is calculated by segmenting, and the energy candidate peak is obtained after moving average to detect QRS complex. Secondly, the QRS amplitude is set to zero and the fifth component of SWT is used to locate P wave and T wave. The experimental results show that compared with other algorithms, the algorithm in this paper has high accuracy in detecting QRS complex in different motion states. It only takes 0.22 s to detect QSR complex of a 30-minute ECG record, and the real-time performance is improved obviously. On the basis of QRS complex detection, the accuracy of P wave and T wave detection is higher than 95%. The results show that this method can improve the efficiency of ECG signal detection, and provide a new method for real-time ECG signal classification and cardiovascular disease diagnosis.
The existing retinal vessels segmentation algorithms have various problems that the end of main vessels are easy to break, and the central macula and the optic disc boundary are likely to be mistakenly segmented. To solve the above problems, a novel retinal vessels segmentation algorithm is proposed in this paper. The algorithm merged together vessels contour information and conditional generative adversarial nets. Firstly, non-uniform light removal and principal component analysis were used to process the fundus images. Therefore, it enhanced the contrast between the blood vessels and the background, and obtained the single-scale gray images with rich feature information. Secondly, the dense blocks integrated with the deep separable convolution with offset and squeeze-and-exception (SE) block were applied to the encoder and decoder to alleviate the gradient disappearance or explosion. Simultaneously, the network focused on the feature information of the learning target. Thirdly, the contour loss function was added to improve the identification ability of the blood vessels information and contour information of the network. Finally, experiments were carried out on the DRIVE and STARE datasets respectively. The value of area under the receiver operating characteristic reached 0.982 5 and 0.987 4, respectively, and the accuracy reached 0.967 7 and 0.975 6, respectively. Experimental results show that the algorithm can accurately distinguish contours and blood vessels, and reduce blood vessel rupture. The algorithm has certain application value in the diagnosis of clinical ophthalmic diseases.
Compared with the previous automatic segmentation neural network for the target area which considered the target area as an independent area, a stacked neural network which uses the position and shape information of the organs around the target area to regulate the shape and position of the target area through the superposition of multiple networks and fusion of spatial position information to improve the segmentation accuracy on medical images was proposed in this paper. Taking the Graves’ ophthalmopathy disease as an example, the left and right radiotherapy target areas were segmented by the stacked neural network based on the fully convolutional neural network. The volume Dice similarity coefficient (DSC) and bidirectional Hausdorff distance (HD) were calculated based on the target area manually drawn by the doctor. Compared with the full convolutional neural network, the stacked neural network segmentation results can increase the volume DSC on the left and right sides by 1.7% and 3.4% respectively, while the two-way HD on the left and right sides decrease by 0.6. The results show that the stacked neural network improves the degree of coincidence between the automatic segmentation result and the doctor's delineation of the target area, while reducing the segmentation error of small areas. The stacked neural network can effectively improve the accuracy of the automatic delineation of the radiotherapy target area of Graves' ophthalmopathy.
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.
ObjectiveTo systematically summarize recent advancements in the application of artificial intelligence (AI) in key components of radiotherapy (RT), explore the integration of technical innovations with clinical practice, and identify current limitations in real-world implementation. MethodsA comprehensive analysis of representative studies from recent years was conducted, focusing on the technical implementation and clinical effectiveness of AI in image reconstruction, automatic delineation of target volumes and organs at risk, intelligent treatment planning, and prediction of RT-related toxicities. Particular attention was given to deep learning models, multimodal data integration, and their roles in enhancing decision-making processes. ResultsAI-based low-dose image enhancement techniques had significantly improved image quality. Automated segmentation methods had increased the efficiency and consistency of contouring. Both knowledge-driven and data-driven planning systems had addressed the limitations of traditional experience-dependent approaches, contributing to higher quality and reproducibility in treatment plans. Additionally, toxicity prediction models that incorporated multimodal data enabled more accurate, personalized risk assessment, supporting safer and more effective individualized RT. ConclusionsRT is a fundamental modality in cancer treatment. However, achieving precise tumor ablation while minimizing damage to surrounding healthy tissues remains a significant challenge. AI has demonstrated considerable value across multiple technical stages of RT, enhancing precision, efficiency, and personalization. Nevertheless, challenges such as limited model generalizability, lack of data standardization, and insufficient clinical validation persist. Future work should emphasize the alignment of algorithmic development with clinical demands to facilitate the standardized, reliable, and practical application of AI in RT.
Glaucoma is one of blind causing diseases. The cup-to-disc ratio is the main basis for glaucoma screening. Therefore, it is of great significance to precisely segment the optic cup and disc. In this article, an optic cup and disc segmentation model based on the linear attention and dual attention is proposed. Firstly, the region of interest is located and cropped according to the characteristics of the optic disc. Secondly, linear attention residual network-34 (ResNet-34) is introduced as a feature extraction network. Finally, channel and spatial dual attention weights are generated by the linear attention output features, which are used to calibrate feature map in the decoder to obtain the optic cup and disc segmentation image. Experimental results show that the intersection over union of the optic disc and cup in Retinal Image Dataset for Optic Nerve Head Segmentation (DRISHTI-GS) dataset are 0.962 3 and 0.856 4, respectively, and the intersection over union of the optic disc and cup in retinal image database for optic nerve evaluation (RIM-ONE-V3) are 0.956 3 and 0.784 4, respectively. The proposed model is better than the comparison algorithm and has certain medical value in the early screening of glaucoma. In addition, this article uses knowledge distillation technology to generate two smaller models, which is beneficial to apply the models to embedded device.
In the segmentation of aortic dissection, there are issues such as low contrast between the aortic dissection and surrounding organs and vessels, significant differences in dissection morphology, and high background noise. To address these issues, this paper proposed a reinforcement learning-based method for type B aortic dissection localization. With the assistance of a two-stage segmentation model, the deep reinforcement learning was utilized to perform the first-stage aortic dissection localization task, ensuring the integrity of the localization target. In the second stage, the coarse segmentation results from the first stage were used as input to obtain refined segmentation results. To improve the recall rate of the first-stage segmentation results and include the segmentation target more completely in the localization results, this paper designed a reinforcement learning reward function based on the direction of recall changes. Additionally, the localization window was separated from the field of view window to reduce the occurrence of segmentation target loss. Unet, TransUnet, SwinUnet, and MT-Unet were selected as benchmark segmentation models. Through experiments, it was verified that the majority of the metrics in the two-stage segmentation process of this paper performed better than the benchmark results. Specifically, the Dice index improved by 1.34%, 0.89%, 27.66%, and 7.37% for each respective model. In conclusion, by incorporating the type B aortic dissection localization method proposed in this paper into the segmentation process, the overall segmentation accuracy is improved compared to the benchmark models. The improvement is particularly significant for models with poorer segmentation performance.
As an important basis for lesion determination and diagnosis, medical image segmentation has become one of the most important and hot research fields in the biomedical field, among which medical image segmentation algorithms based on full convolutional neural network and U-Net neural network have attracted more and more attention by researchers. At present, there are few reports on the application of medical image segmentation algorithms in the diagnosis of rectal cancer, and the accuracy of the segmentation results of rectal cancer is not high. In this paper, a convolutional network model of encoding and decoding combined with image clipping and pre-processing is proposed. On the basis of U-Net, this model replaced the traditional convolution block with the residual block, which effectively avoided the problem of gradient disappearance. In addition, the image enlargement method is also used to improve the generalization ability of the model. The test results on the data set provided by the "Teddy Cup" Data Mining Challenge showed that the residual block-based improved U-Net model proposed in this paper, combined with image clipping and preprocessing, could greatly improve the segmentation accuracy of rectal cancer, and the Dice coefficient obtained reached 0.97 on the verification set.