Automated segmentation of metal stent and bioresorbable vascular scaffold in intravascular optical coherence tomography images using deep learning architectures

Percutaneous coronary intervention (PCI) with stent placement is a treatment effective for coronary artery diseases. Intravascular optical coherence tomography (OCT) with high resolution is used clinically to visualize stent deployment and restenosis, facilitating PCI operation and for complication inspection. Automated stent struts segmentation in OCT images is necessary as each pullback of OCT images could contain thousands of stent struts. In this paper, a deep learning framework is proposed and demonstrated for the automated segmentation of two major clinical stent types: metal stents and bioresorbable vascular scaffolds (BVS). U-Net, the current most prominent deep learning network in biomedical segmentation, was implemented for segmentation with cropped input. The architectures of MobileNetV2 and DenseNet121 were also adapted into U-Net for improvement in speed and accuracy. The results suggested that the proposed automated algorithm's segmentation performance approaches the level of independent human obsevers and is feasible for both types of stents despite their distinct appearance. U-Net with DenseNet121 encoder (U-Dense) performed best with Dice's coefficient of 0.86 for BVS segmentation, and precision/recall of 0.92/0.92 for metal stent segmentation under optimal crop window size of 256.

Influence of Diversity Nursing on Patients' Rehabilitation in Cardiology Treatment

With the improvement of living standards, people have more and more physical health problems. Among them, high-risk cardiovascular diseases such as hypertension, diabetes, and coronary heart disease are the most prominent. The number of cardiology patients is increasing year by year. Effectively improving the treatment of cardiology patients and speeding up the recovery of cardiology patients have become a social problem. This article aims to explore the impact of diverse nursing care on patients in cardiology treatment. This article first gives a detailed introduction to the treatment of diverse nursing and cardiology diseases, then takes 300 cardiology patients in our hospital as experimental subjects, and conducts a controlled experiment of nursing intervention, which is categorized into an experimental group of 150 cases (including 35 cases of hypertension, 46 cases of diabetes, 28 cases of coronary heart disease, 24 cases of angina pectoris, and 17 cases of multiple complications) and a control group of 150 cases (including 30 cases of hypertension, 47 cases of diabetes, 39 cases of coronary heart disease, 21 cases of angina pectoris, and 13 cases of multiple complications). The experimental results showed the following: the general information of the two groups of patients was not statistically different (P > 0.05); after the nursing intervention, the blood glucose levels of the two groups of patients decreased, but the experimental group decreased more significantly and the blood glucose control effect was more obvious; after the intervention, in the experimental group that implemented diversified nursing interventions, the patient's condition management effect was better and their scores were between 8 and 10; the mental state self-evaluation of the two groups of patients was significantly different from the domestic reference value (P < 0.05), and there is a very significant statistical difference between the two groups after nursing intervention (P < 0.01); after nursing intervention, compared with the control group, the quality of life of the experimental group improved more significantly and the highest score reached about 70; the overall satisfaction of the experimental group with nursing work reached 92%, while the satisfaction of the control group with nursing work was only 44.67%. Studies have shown that diversified care has a positive impact on the rehabilitation of patients in cardiology treatment.

Framed Recurrent Neural Network with Mutated Dragonfly Search Optimization (FRNN-MDSO) for hyperspectral image based face recognition in real time door locking applications

Face recognition is one of the best applications of computer recognition and recent smart house applications. Therefore, it draws considerable attention from researchers. Several face recognition algorithms have been proposed in the last decade, but these methods did not give the efficient outcome. Therefore, this work introduces a novel constructive training algorithm for smart face recognition in door locking applications. The proposed Framed Recurrent Neural Network with Mutated Dragonfly Search Optimization (FRNN-MDSO) Strategy is applied to face recognition application. The steady preparing system has been utilized where the training designs are adapted steadily and are divided into completely different modules. The facial feature process works on global and local features. After the feature extraction and selection process, employ the improved classifier followed by the Framed Recurrent Neural Network classification technique. Finally, the face image based on the feature library can be identified. The proposed Framed Recurrent Neural Network with Mutated Dragonfly Search Optimization starts with a single training pattern using Bidirectional Encoder Representations from Transformers (BERT) model. During network training, the Training Data (TD) decrease the Mean Square Error (MSE) while the matching process increases the algorithms generated which are trapped at the local minimum. The training data have been trained to increase the number of input forms (one after the other) until all the forms are selected and trained. An FRNN-MDSO based face recognition system is built, and face recognition is tested using hyperspectral Database parameters. The simulation results indicate that the proposed method acquires the associate grade optimum design of FRNN with MDSO methodology using the present constructive algorithm and prove the proposed FRNN-MDSO method’s effectiveness compared to the conventional architecture methods.

Application Research on Optimization Algorithm of sEMG Gesture Recognition Based on Light CNN+LSTM Model

The deep learning gesture recognition based on surface electromyography plays an increasingly important role in human-computer interaction. In order to ensure the high accuracy of deep learning in multistate muscle action recognition and ensure that the training model can be applied in the embedded chip with small storage space, this paper presents a feature model construction and optimization method based on multichannel sEMG amplification unit. The feature model is established by using multidimensional sequential sEMG images by combining convolutional neural network and long-term memory network to solve the problem of multistate sEMG signal recognition. The experimental results show that under the same network structure, the sEMG signal with fast Fourier transform and root mean square as feature data processing has a good recognition rate, and the recognition accuracy of complex gestures is 91.40%, with the size of 1 MB. The model can still control the artificial hand accurately when the model is small and the precision is high.

Extraction of a vascular function for a fully automated dynamic contrast-enhanced magnetic resonance brain image processing pipeline

PURPOSE

To develop a deep-learning model that leverages the spatial and temporal information from dynamic contrast-enhanced magnetic resonance (DCE MR) brain imaging in order to automatically estimate a vascular function (VF) for quantitative pharmacokinetic (PK) modeling.

METHODS

Patients with glioblastoma multiforme were scanned post-resection approximately every 2 months using a high spatial and temporal resolution DCE MR imaging sequence ( ≈ 5 s and ≈ 2 cm³ ). A region over the transverse sinus was manually drawn in the dynamic T1-weighted images to provide a ground truth VF. The manual regions and their resulting VF curves were used to train a deep-learning model based on a 3D U-net architecture. The model concurrently utilized the spatial and temporal information in DCE MR images to predict the VF. In order to analyze the contribution of the spatial and temporal terms, different weighted combinations were examined. The manual and deep-learning predicted regions and VF curves were compared.

RESULTS

Forty-three patients were enrolled in this study and 155 DCE MR scans were processed. The 3D U-net was trained using a loss function that combined the spatial and temporal information with different weightings. The best VF curves were obtained when both spatial and temporal information were considered. The predicted VF curve was similar to the manual ground truth VF curves.

CONCLUSION

The use of spatial and temporal information improved VF curve prediction relative to when only the spatial information is used. The method generalized well for unseen data and can be used to automatically estimate a VF curve suitable for quantitative PK modeling. This method allows for a more efficient clinical pipeline and may improve automation of permeability mapping.

Evaluation of the Effectiveness of Artificial Intelligence Chest CT Lung Nodule Detection Based on Deep Learning

Lung cancer is one of the most malignant tumors. If it can be detected early and treated actively, it can effectively improve a patient's survival rate. Therefore, early diagnosis of lung cancer is very important. Early-stage lung cancer usually appears as a solitary lung nodule on medical imaging. It usually appears as a round or nearly round dense shadow in the chest radiograph. It is difficult to distinguish lung nodules and lung soft tissues with the naked eye. Therefore, this article proposes a deep learning-based artificial intelligence chest CT lung nodule detection performance evaluation study, aiming to evaluate the value of chest CT imaging technology in the detection of noncalcified nodules and provide help for the detection and treatment of lung cancer. In this article, the Lung Medical Imaging Database Consortium (LIDC) was selected to obtain 536 usable cases based on inclusion criteria; 80 cases were selected for examination, artificial intelligence software, radiologists, and thoracic imaging specialists. Using 80 pulmonary nodules detection in each case, the pathological type of pulmonary nodules, nonlime tuberculous test results, detection sensitivity, false negative rate, false positive rate, and CT findings were individually analyzed, and the detection efficiency software of artificial intelligence was evaluated. Experiments have proved that the sensitivity of artificial intelligence software to detect noncalcified nodules in the pleural, peripheral, central, and hilar areas is higher than that of radiologists, indicating that the method proposed in this article has achieved good detection results. It has a better nodule detection sensitivity than a radiologist, reducing the complexity of the detection process.

Multiorgan segmentation from partially labeled datasets with conditional nnU-Net

Accurate and robust multiorgan abdominal CT segmentation plays a significant role in numerous clinical applications, such as therapy treatment planning and treatment delivery. Almost all existing segmentation networks rely on fully annotated data with strong supervision. However, annotating fully annotated multiorgan data in CT images is both laborious and time-consuming. In comparison, massive partially labeled datasets are usually easily accessible. In this paper, we propose conditional nnU-Net trained on the union of partially labeled datasets for multiorgan segmentation. The deep model employs the state-of-the-art nnU-Net as the backbone and introduces a conditioning strategy by feeding auxiliary information into the decoder architecture as an additional input layer. This model leverages the prior conditional information to identify the organ class at the pixel-wise level and encourages organs' spatial information recovery. Furthermore, we adopt a deep supervision mechanism to refine the outputs at different scales and apply the combination of Dice loss and Focal loss to optimize the training model. Our proposed method is evaluated on seven publicly available datasets of the liver, pancreas, spleen and kidney, in which promising segmentation performance has been achieved. The proposed conditional nnU-Net breaks down the barriers between nonoverlapping labeled datasets and further alleviates the problem of data hunger in multiorgan segmentation.

The Application and Development of Deep Learning in Radiotherapy: A Systematic Review

With the massive use of computers, the growth and explosion of data has greatly promoted the development of artificial intelligence (AI). The rise of deep learning (DL) algorithms, such as convolutional neural networks (CNN), has provided radiation oncologists with many promising tools that can simplify the complex radiotherapy process in the clinical work of radiation oncology, improve the accuracy and objectivity of diagnosis, and reduce the workload, thus enabling clinicians to spend more time on advanced decision-making tasks. As the development of DL gets closer to clinical practice, radiation oncologists will need to be more familiar with its principles to properly evaluate and use this powerful tool. In this paper, we explain the development and basic concepts of AI and discuss its application in radiation oncology based on different task categories of DL algorithms. This work clarifies the possibility of further development of DL in radiation oncology.

Personalized recommendation algorithm in social networks based on representation learning

Recommendation algorithm is not only widely used in entertainment media, but also plays an important role in national strategy, such as the recommendation algorithm of byte beating company. This paper studies the personalized recommendation algorithm based on representation learning. The data in social network is complex, and the data mainly exists in various platforms. This paper introduces AI (Artificial Intelligence) algorithm to guide the algorithm of representation learning, and integrates the algorithm steps of representation learning, to realize the implementation of personalized recommendation algorithm in social network, and compares the representation learning algorithm. Finally, this paper designs a method based on heat conduction and text mining to provide users with webpage recommendations and help users better mine interesting popular webpages. Research shows that the performance of IMF is better than that of PMF because it overcomes the sparsity of data by pre-filling. The accuracy of IMF is 3.69% higher than that of PMF on the epinions data set, and 6.24% higher than that of PMF on the double data set. Rtcf, socialmf, tcars, CSIT, isrec, and hesmf have better performance than PMF and IMF. Among them, rtcf, socialmf, tcars, CSIT, isrec, and hesmf improve the MAE performance of PMF by 7.6%, 6.3%, 8.8%, 7.9%, 9.5% and 14.2%, respectively.

Anatomy-aided deep learning for medical image segmentation: a review

Deep learning (DL) has become widely used for medical image segmentation in recent years. However, despite these advances, there are still problems for which DL-based segmentation fails. Recently, some DL approaches had a breakthrough by using anatomical information which is the crucial cue for manual segmentation. In this paper, we provide a review of anatomy-aided DL for medical image segmentation which covers systematically summarized anatomical information categories and corresponding representation methods. We address known and potentially solvable challenges in anatomy-aided DL and present a categorized methodology overview on using anatomical information with DL from over 70 papers. Finally, we discuss the strengths and limitations of the current anatomy-aided DL approaches and suggest potential future work.

Automatic Pancreas Segmentation in CT Images With Distance-Based Saliency-Aware DenseASPP Network

Pancreas identification and segmentation is an essential task in the diagnosis and prognosis of pancreas disease. Although deep neural networks have been widely applied in abdominal organ segmentation, it is still challenging for small organs (e.g. pancreas) that present low contrast, highly flexible anatomical structure and relatively small region. In recent years, coarse-to-fine methods have improved pancreas segmentation accuracy by using coarse predictions in the fine stage, but only object location is utilized and rich image context is neglected. In this paper, we propose a novel distance-based saliency-aware model, namely DSD-ASPP-Net, to fully use coarse segmentation to highlight the pancreas feature and boost accuracy in the fine segmentation stage. Specifically, a DenseASPP (Dense Atrous Spatial Pyramid Pooling) model is trained to learn the pancreas location and probability map, which is then transformed into saliency map through geodesic distance-based saliency transformation. In the fine stage, saliency-aware modules that combine saliency map and image context are introduced into DenseASPP to develop the DSD-ASPP-Net. The architecture of DenseASPP brings multi-scale feature representation and achieves larger receptive field in a denser way, which overcome the difficulties brought by variable object sizes and locations. Our method was evaluated on both public NIH pancreas dataset and local hospital dataset, and achieved an average Dice-Sørensen Coefficient (DSC) value of 85.49±4.77% on the NIH dataset, outperforming former coarse-to-fine methods.

Automatic Detection of Atherosclerotic Plaque and Calcification From Intravascular Ultrasound Images by Using Deep Convolutional Neural Networks

Atherosclerosis is the major cause of cardiovascular diseases (CVDs). Intravascular ultrasound (IVUS) is a common imaging modality for diagnosing CVDs. However, an efficient analyzer for IVUS image segmentation is required for assisting cardiologists. In this study, an end-to-end deep-learning convolutional neural network was developed for automatically detecting media-adventitia borders, luminal regions, and calcified plaque in IVUS images. A total of 713 grayscale IVUS images from 18 patients were used as training data for the proposed deep-learning model. The model is constructed using the three modified U-Nets and combined with the concept of cascaded networks to prevent errors in the detection of calcification owing to the interference of pixels outside the plaque regions. Three loss functions (Dice, Tversky, and focal loss) with various characteristics were tested to determine the best setting for the proposed model. The efficacy of the deep-learning model was evaluated by analyzing precision-recall curve. The average precision (AP), Dice score coefficient, precision, sensitivity, and specificity of the predicted and ground truth results were then compared. All training processes were validated using leave-one-subject-out cross-validation. The experimental results showed that the proposed deep-learning model exhibits high performance in segmenting the media-adventitia layers and luminal regions for all loss functions, with all tested metrics being higher than 0.90. For locating calcified tissues, the best result was obtained when the focal loss function was applied to the proposed model, with an AP of 0.73; however, the prediction efficacy was affected by the proportion of calcified tissues within the plaque region when the focal loss function was employed. Compared with commercial software, the proposed method exhibited high accuracy in segmenting IVUS images in some special cases, such as when shadow artifacts or side vessels surrounded the target vessel.

Mathematical modeling method of cell tension and compression based on multi-modal mechanical signals

Mechanical biology is the study of the influence of the mechanical environment on human health, disease, or injury. To study the mechanism of the organism’s perception and response to mechanical signals can promote the development of biomedical basic and clinical research, and promote human health. The purpose of this paper is to study the mathematical modeling method of the effect of multimodal mechanical signals on cell stretching and compression. This article first established a cell mechanics model based on the generalization of membrane theory, introduced the micro-manipulation techniques used to characterize cell mechanics and the method of cell mechanics loading, and then explained why mathematical modeling was established. Finally, according to the multi-modality During the mechanical preparation process, the effects of multi-modal mechanical signals on the stretching and compression of annulus fibrosus stem cells were studied. The experimental results in this paper show that after planting fibrous stem cells with different elastic modulus, the cell proliferation is obvious after the tensile mechanical stimulation of different conditions, and the different elastic modulus scaffolds are stimulated by the tensile mechanical stimulation of 2% tensile amplitude. The cell morphology is different. The low elastic modulus is round-like, and the high elastic modulus is fusiform-like. After 5% and 12% stretch amplitude, the cells are oriented at different elastic modulus. Arranged, there is no obvious difference in cell morphology.

Application of Laparoscopy in Comprehensive Staging Operation of Ovarian Cancer Based on Electronic Medical Blockchain Technology

Ovarian cancer has always entangled most women. Studies have shown that the prevalence of ovarian cancer ranks third in female reproductive malignancies, and the mortality rate has always been the highest. The reason is mainly because the diagnosis and treatment of preovarian cancer has always been a big problem. However, the emergence of laparoscopy can well solve this problem, especially laparoscopy assisted by blockchain technology, which plays a huge role in the overall staging of ovarian cancer. This article proposes the application research of laparoscopy in the comprehensive staging of ovarian cancer based on electronic medical blockchain technology. First of all, this article uses the literature method to study the clinical characteristics and surgical classification of ovarian cancer, as well as the application status of blockchain technology and laparoscopic technology. Secondly, it designed an application experiment based on electronic medical blockchain technology to assist laparoscopy in the comprehensive staging of ovarian cancer and analyzed the comparison of the laparoscopic group and the control group in the comprehensive staging of ovarian cancer. The results of the study showed that the amount of bleeding in the laparoscopic group was 103.5 ml, while the amount of bleeding in the control group was 141.1 ml; the proportion of tertiary pain in the laparoscopic group was 11.37%, and the proportion of tertiary pain in the control group was 31.82%. From this, it can be seen that, in the comprehensive staging operation for ovarian cancer, the laparoscopic group has less intraoperative blood loss than the control group and lower pain, and the treatment effect is better.

Research on the expression of new visual intelligence system based on machine learning technology

With the continuous progress of society, the level of science and technology of the country has made a leap forward development, the research energy of various industries on new science and technology continues to deepen, greatly promoting the promotion of science and technology. At the same time, with the increase in social pressure, more and more people pursue spiritual relaxation, and appropriate leisure and entertainment activities have gradually become a part of people’s life. Film plays an irreplaceable role in leisure and entertainment. Mainly from the background of the development of the film industry towards intelligent direction, and then use machine learning technology to study the application of film animation production and film virtual assets analysis and investigation. Based on the Internet of things technology, we also vigorously develop the ways and methods of visual expression of movies, and at the same time introduce new expression modes to promote the expression effect of the intelligent system. Finally, by comparing various algorithms in machine learning technology, the results of intelligent expression of random number forest algorithm in machine learning technology are more accurate. The system is also applied to 3D animation production to observe the measurement error of 3D motion data and facial expression data.

Pose recognition in sports scenes based on deep learning skeleton sequence model

Human skeleton extraction is a basic problem in the field of computer vision. With the rapid progress of science and technology, it has become a hot issue in the field of target detection such as pedestrian recognition, behavior monitoring, and pedestrian gesture recognition. In recent years, due to the development of deep neural networks, modeling of human joints in acquired images has made progress in skeleton extraction. However, most models have low modeling accuracy, poor real-time performance, and poor model availability. problem. Aiming at the above-mentioned human target detection problem, this paper uses the deep learning skeleton sequence model gesture recognition method in sports scenes to study, aiming to provide a gesture recognition method with strong noise resistance, good real-time performance and accurate model. This article uses motion video frame images to train the VGG16 network. Using the network to extract skeleton information can strengthen the posture feature expression, and use HOG for feature extraction, and use the Adam algorithm to optimize the network to extract more posture features, thereby improving the posture of the network Recognition accuracy. Then adjust the hyperparameters and network structure of the basic network according to the training results, and obtain the key poses in the sports scene through the final classifier.

Deep Reinforcement Learning for Weakly-Supervised Lymph Node Segmentation in CT Images

Accurate and automated lymph node segmentation is pivotal for quantitatively accessing disease progression and potential therapeutics. The complex variation of lymph node morphology and the difficulty of acquiring voxel-wise manual annotations make lymph node segmentation a challenging task. Since the Response Evaluation Criteria in Solid Tumors (RECIST) annotation, which indicates the location, length, and width of a lymph node, is commonly available in hospital data archives, we advocate to use RECIST annotations as the supervision, and thus formulate this segmentation task into a weakly-supervised learning problem. In this paper, we propose a deep reinforcement learning-based lymph node segmentation (DRL-LNS) model. Based on RECIST annotations, we segment RECIST-slices in an unsupervised way to produce pseudo ground truths, which are then used to train U-Net as a segmentation network. Next, we train a DRL model, in which the segmentation network interacts with the policy network to optimize the lymph node bounding boxes and segmentation results simultaneously. The proposed DRL-LNS model was evaluated against three widely used image segmentation networks on a public thoracoabdominal Computed Tomography (CT) dataset that contains 984 3D lymph nodes, and achieves the mean Dice similarity coefficient (DSC) of 77.17% and the mean Intersection over Union (IoU) of 64.78% in the four-fold cross-validation. Our results suggest that the DRL-based bounding box prediction strategy outperforms the label propagation strategy and the proposed DRL-LNS model is able to achieve the state-of-the-art performance on this weakly-supervised lymph node segmentation task.

CAMPO Precision128 Max ENERGY Spectrum CT Combined with Multiple Parameters to Evaluate the Benign and Malignant Pleural Effusion

The emergence of energy spectrum CT provides greater diagnostic value for clinical practice. Its advantage is that it can provide more functional imaging parameters and accurate image information for clinical practice, which represents a mainstream direction of CT technology development at present. This paper mainly studies the clinical trial of CAMPO Precision128 Max ENERGY spectrum CT combined with multiple parameters to evaluate the benign and malignant pleural effusion. This paper analyzes the principle and key performance parameters of energy spectrum CT imaging, the etiology of pleural effusion, and its conventional diagnostic methods and uses energy spectrum CT to detect the benign and malignant pleural effusion. In this paper, two groups of patients with different types of pleural effusions were scanned by line spectrum chest CT scans, and energy spectrum analysis software was used to measure and calculate the CT values of conventional mixed energy values of ROI of patients with pleural effusions. For the CT value and energy curve slope measurement value of different single energy keV, independent sample t-test was used to analyze and compare the two sets of data, and finally it has been found out that the two sets of data were similar. According to the experimental results, the curves of energy spectrum of the two groups of data are similar in the descending curve of bow-back. The slope of energy spectrum curve in the leakage group was lower than that in the exudate group, showing statistical significance (P < 0.05). The slope of energy spectrum curve K in the malignant pleural effusion group was significantly higher than that in the benign pleural effusion group, and the difference was statistically significant (P < 0.05). The trend of energy spectrum curves of the two is roughly the same, while at the high energy level, part of the energy spectrum curves of the two are overlapped. The above conclusion indicates that energy spectrum CT plays a certain role in the differential diagnosis of pleural effusion. At the same time, energy spectrum CT also provides a noninvasive and rapid examination method for clinical differentiation of pleural effusion, which has certain clinical application value and prospect.

Iris-Fingerprint multimodal biometric system based on optimal feature level fusion model

<abstract><p>For reliable and accurate multimodal biometric based person verification, demands an effective discriminant feature representation and fusion of the extracted relevant information across multiple biometric modalities. In this paper, we propose feature level fusion by adopting the concept of canonical correlation analysis (CCA) to fuse Iris and Fingerprint feature sets of the same person. The uniqueness of this approach is that it extracts maximized correlated features from feature sets of both modalities as effective discriminant information within the features sets. CCA is, therefore, suitable to analyze the underlying relationship between two feature spaces and generates more powerful feature vectors by removing redundant information. We demonstrate that an efficient multimodal recognition can be achieved with a significant reduction in feature dimensions with less computational complexity and recognition time less than one second by exploiting CCA based joint feature fusion and optimization. To evaluate the performance of the proposed system, Left and Right Iris, and thumb Fingerprints from both hands of the SDUMLA-HMT multimodal dataset are considered in this experiment. We show that our proposed approach significantly outperforms in terms of equal error rate (EER) than unimodal system recognition performance. We also demonstrate that CCA based feature fusion excels than the match score level fusion. Further, an exploration of the correlation between Right Iris and Left Fingerprint images (EER of 0.1050%), and Left Iris and Right Fingerprint images (EER of 1.4286%) are also presented to consider the effect of feature dominance and laterality of the selected modalities for the robust multimodal biometric system.</p></abstract>

Fast parallel vessel segmentation

BACKGROUND AND OBJECTIVE

Accurate and fast vessel segmentation from liver slices remain challenging and important tasks for clinicians. The algorithms from the literature are slow and less accurate. We propose fast parallel gradient based seeded region growing for vessel segmentation. Seeded region growing is tedious when the inter connectivity between the elements is unavoidable. Parallelizing region growing algorithms are essential towards achieving real time performance for the overall process of accurate vessel segmentation.

METHODS

The parallel implementation of seeded region growing for vessel segmentation is iterative and hence time consuming process. Seeded region growing is implemented as kernel termination and relaunch on GPU due to its iterative mechanism. The iterative or recursive process in region growing is time consuming due to intermediate memory transfers between CPU and GPU. We propose persistent and grid-stride loop based parallel approach for region growing on GPU. We analyze static region of interest of tiles on GPU for the acceleration of seeded region growing.

RESULTS

We aim fast parallel gradient based seeded region growing for vessel segmentation from CT liver slices. The proposed parallel approach is 1.9x faster compared to the state-of-the-art.

CONCLUSION

We discuss gradient based seeded region growing and its parallel implementation on GPU. The proposed parallel seeded region growing is fast compared to kernel termination and relaunch and accurate in comparison to Chan-Vese and Snake model for vessel segmentation.

Multiple Discrimination and Pairwise CNN for view-based 3D object retrieval

With the rapid development and wide application of computer, camera device, network and hardware technology, 3D object (or model) retrieval has attracted widespread attention and it has become a hot research topic in the computer vision domain. Deep learning features already available in 3D object retrieval have been proven to be better than the retrieval performance of hand-crafted features. However, most existing networks do not take into account the impact of multi-view image selection on network training, and the use of contrastive loss alone only forcing the same-class samples to be as close as possible. In this work, a novel solution named Multi-view Discrimination and Pairwise CNN (MDPCNN) for 3D object retrieval is proposed to tackle these issues. It can simultaneously input multiple batches and multiple views by adding the Slice layer and the Concat layer. Furthermore, a highly discriminative network is obtained by training samples that are not easy to be classified by clustering. Lastly, we deploy the contrastive-center loss and contrastive loss as the optimization objective that has better intra-class compactness and inter-class separability. Large-scale experiments show that the proposed MDPCNN can achieve a significant performance over the state-of-the-art algorithms in 3D object retrieval.

GPU acceleration of liver enhancement for tumor segmentation

BACKGROUND AND OBJECTIVE

Medical image segmentation plays a vital role in medical image analysis. There are many algorithms developed for medical image segmentation which are based on edge or region characteristics. These are dependent on the quality of the image. The contrast of a CT or MRI image plays an important role in identifying region of interest i.e. lesion(s). In order to enhance the contrast of image, clinicians generally use manual histogram adjustment technique which is based on 1D histogram specification. This is time consuming and results in poor distribution of pixels over the image. Cross modality based contrast enhancement is 2D histogram specification technique. This is robust and provides a more uniform distribution of pixels over CT image by exploiting the inner structure information from MRI image. This helps in increasing the sensitivity and accuracy of lesion segmentation from enhanced CT image. The sequential implementation of cross modality based contrast enhancement is slow. Hence we propose GPU acceleration of cross modality based contrast enhancement for tumor segmentation.

METHODS

The aim of this study is fast parallel cross modality based contrast enhancement for CT liver images. This includes pairwise 2D histogram, histogram equalization and histogram matching. The sequential implementation of the cross modality based contrast enhancement is computationally expensive and hence time consuming. We propose persistence and grid-stride loop based fast parallel contrast enhancement for CT liver images. We use enhanced CT liver image for the lesion or tumor segmentation. We implement the fast parallel gradient based dynamic seeded region growing for lesion segmentation.

RESULTS

The proposed parallel approach is 104.416 ( ±  5.166) times faster compared to the sequential implementation and increases the sensitivity and specificity of tumor segmentation.

CONCLUSION

The cross modality approach is inspired by 2D histogram specification which incorporates spatial information existing in both guidance and input images for remapping the input image intensity values. The cross modality based liver contrast enhancement improves the quality of tumor segmentation.