Glomerular lesions are the main injuries of diabetic nephropathy (DN) and are used as a crucial index for pathologic classification. Manual quantification of these morphologic features currently used ...is semi-quantitative and time-consuming. Automatically quantifying glomerular morphologic features is urgently needed.
A series of convolutional neural networks (CNN) were designed to identify and classify glomerular morphologic features in DN patients. Associations of these digital features with pathologic classification and prognosis were further analyzed.
Our CNN-based model achieved a 0.928 F1-score for global glomerulosclerosis and 0.953 F1-score for Kimmelstiel-Wilson lesion, further obtained a dice of 0.870 for the mesangial area and F1-score beyond 0.839 for three glomerular intrinsic cells. As the pathologic classes increased, mesangial cell numbers and mesangial area increased, and podocyte numbers decreased (p for all < 0.001), while endothelial cell numbers remained stable (p = 0.431). Glomeruli with Kimmelstiel-Wilson lesion showed more severe podocyte deletion compared to those without (p < 0.001). Furthermore, CNN-based classifications showed moderate agreement with pathologists-based classification, the kappa value between the CNN model 3 and pathologists reached 0.624 (ranging from 0.529 to 0.688, p < 0.001). Notably, CNN-based classifications obtained equivalent performance to pathologists-based classifications on predicting baseline and long-term renal function.
Our CNN-based model is promising in assisting the identification and pathologic classification of glomerular lesions in DN patients.
•The KiTS19 Challenge measured the state of the art in kidney and tumor segmentation.•Deep 3D CNNs were by far the most popular method used by submissions.•The nnU-Net won with a kidney Dice of 0.974 ...and a tumor Dice of 0.851.•The challenge remains open as a challenging benchmark in 3D semantic segmentation.
Display omitted
There is a large body of literature linking anatomic and geometric characteristics of kidney tumors to perioperative and oncologic outcomes. Semantic segmentation of these tumors and their host kidneys is a promising tool for quantitatively characterizing these lesions, but its adoption is limited due to the manual effort required to produce high-quality 3D segmentations of these structures. Recently, methods based on deep learning have shown excellent results in automatic 3D segmentation, but they require large datasets for training, and there remains little consensus on which methods perform best. The 2019 Kidney and Kidney Tumor Segmentation challenge (KiTS19) was a competition held in conjunction with the 2019 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) which sought to address these issues and stimulate progress on this automatic segmentation problem. A training set of 210 cross sectional CT images with kidney tumors was publicly released with corresponding semantic segmentation masks. 106 teams from five continents used this data to develop automated systems to predict the true segmentation masks on a test set of 90 CT images for which the corresponding ground truth segmentations were kept private. These predictions were scored and ranked according to their average Sørensen-Dice coefficient between the kidney and tumor across all 90 cases. The winning team achieved a Dice of 0.974 for kidney and 0.851 for tumor, approaching the inter-annotator performance on kidney (0.983) but falling short on tumor (0.923). This challenge has now entered an “open leaderboard” phase where it serves as a challenging benchmark in 3D semantic segmentation.
The morphological characteristics of kidney tumor is crucial factor for radiologists to make accurate diagnosis and treatment. Unfortunately, performing quantitative study of the relationship between ...kidney tumor morphology and clinical outcomes is very difficult because kidney tumor varies dramatically in its size, shape, location, etc. Automatic semantic segmentation of kidney and tumor is a promising tool towards developing advanced surgical planning techniques. In this work, we present a triple-stage self-guided network for kidney tumor segmentation task. The low-resolution net can roughly locate the volume of interest (VOI) from down-sampled CT images, while the full-resolution net and tumor refine net can extract accurate boundaries of kidney and tumor within VOI from full resolution CT images. We innovatively propose dilated convolution blocks (DCB) to replace the traditional pooling operations in deeper layers of U-Net architecture to retain detailed semantic information better. Besides, a hybrid loss of dice and weighted cross entropy is used to guide the model to focus on voxels close to the boundary and hard to be distinguished. We evaluate our method on the KiTS19 (MICCAI 2019 Kidney Tumor Segmentation Challenge) test dataset and achieve 0.9674, 0.8454 average dice for kidney and tumor respectively, which ranked the 2nd place in the KiTS19 challenge.
Study on the evaluation of cylinder’s global sizes
In ISO 14405-1, the global sizes, such as least-squares diameter, minimum circumscribed diameter and maximum inscribed diameter are defined. The ...diameters above can be measured by using cylindrical coordinate measuring method like the circular section measuring method of cylindricity error. The determination method of the least-squares diameter was firstly given based on the cylindrical measuring system, and the optimization models of the minimum circumscribed diameter and the maximum inscribed diameter were built, respectively. The corresponding objective functions were unified as “minimax” expressions. For the four axis parameters of the cylinder with the minimum circumscribed diameter or the maximum inscribed diameter, the searching ranges of cylinder’s axis parameters for their optimal solutions were defined numerically. Thereafter, the genetic, steepest decent and BFGS-0.618 algorithms were introduced, and the optimization evaluation algorithms of two kinds of diameters mentioned above were given. Based on many cylinders’ profiles obtained by the circular section measuring method on a measuring instrument of cylinder’s global sizes which was developed by Zhongyuan University of Technology, Zhengzhou, China. The accuracy, efficiency and suitability of three optimization algorithms were investigated through the evaluation of a lot of the minimum circumscribed diameters and the maximum inscribed diameters. The measurement uncertainty of the global sizes for the cylindrical specimen was analyzed, and the measurement uncertainties of the sizes in the radial and z directions are ±0.95μm and ±0.5μm, respectively. The total measurement uncertainties of the global sizes of the cylindrical specimens with the specifications of ϕ10×120mm and ϕ100×300mm are ±3.8μm and ±5.7μm, respectively. The investigation results showed that for the evaluation of the globe sizes, any one of three algorithms above is not absolutely prior to the other two algorithms while considering both evaluation accuracy and efficiency, and the difference of their evaluation results do not exceed 0.5μm. On the other hand, many points between the maximum value and the least value do not affect the evaluation results in optimization process. For improving the evaluation efficiency, by de-selecting those points while considering the characteristic parameter was also studied based on the statistic method and experiment. Coefficient t should be less than 0.3 to ensure the evaluation accuracy. This research may be useful for developing the next generation measurement instrument for the global sizes and the way forward for the digital manufacturing.
Display omitted
•Mathematical explanation for the global diameter defined by ISO 14405-1 was given;•Three optimal evaluation models were established, respectively, as well as their operation procedure;•Calibration measuring concept and the instrument for the global diameter were described.•Measurement uncertainty evaluation method of the developed measuring instrument and evaluation method was present;•Comparison of convergence accuracy and efficiency of the proposed evaluation algorithms was made.
In ISO 14405-1, the global sizes, such as least-squares diameter, minimum circumscribed diameter and maximum inscribed diameter are defined. The diameters above can be measured by using cylindrical coordinate measuring method like the circular section measuring method of cylindricity error. The determination method of the least-squares diameter was firstly given based on the cylindrical measuring system, and the optimization models of the minimum circumscribed diameter and the maximum inscribed diameter were built, respectively. The corresponding objective functions were unified as “minimax” expressions. For the four axis parameters of the cylinder with the minimum circumscribed diameter or the maximum inscribed diameter, the searching ranges of cylinder’s axis parameters for their optimal solutions were defined numerically. Thereafter, the genetic, steepest decent and BFGS-0.618 algorithms were introduced, and the optimization evaluation algorithms of two kinds of diameters mentioned above were given. Based on many cylinders’ profiles obtained by the circular section measuring method on a measuring instrument of cylinder’s global sizes which was developed by Zhongyuan University of Technology, Zhengzhou, China. The accuracy, efficiency and suitability of three optimization algorithms were investigated through the evaluation of a lot of the minimum circumscribed diameters and the maximum inscribed diameters. The measurement uncertainty of the global sizes for the cylindrical specimen was analyzed, and the measurement uncertainties of the sizes in the radial and z directions are ±0.95μm and ±0.5μm, respectively. The total measurement uncertainties of the global sizes of the cylindrical specimens with the specifications of ϕ10×120mm and ϕ100×300mm are ±3.8μm and ±5.7μm, respectively. The investigation results showed that for the evaluation of the globe sizes, any one of three algorithms above is not absolutely prior to the other two algorithms while considering both evaluation accuracy and efficiency, and the difference of their evaluation results do not exceed 0.5μm. On the other hand, many points between the maximum value and the least value do not affect the evaluation results in optimization process. For improving the evaluation efficiency, by de-selecting those points while considering the characteristic parameter was also studied based on the statistic method and experiment. Coefficient t should be less than 0.3 to ensure the evaluation accuracy. This research may be useful for developing the next generation measurement instrument for the global sizes and the way forward for the digital manufacturing.
With the development of distribution networks, both the type of business and the amount of data transmitted in the network are gradually increasing. In order to guarantee the real-time and ...reliability transmission of the business data, it has drawn great concern about how to distribute the network resources reasonably in the distribution communication networks. In this paper, we propose a virtual resource scheduling model for distribution communication networks leveraged by the research on resource virtualization technology and the Software Defined Network (SDN). We also realize the virtualization of physical resources and the unified management and scheduling for the physical resources. This model has not only improved the utilization of communication resources, but also reduced the cost of the maintenance and expansion of the distribution network.
Chronic Kidney Disease has become a worldwide public health problem which demands careful assessments by pathologists. In this paper, we propose a novel architecture for fine-grained classification ...of glomerular lesions in renal pathology images sampling from patients with IgA nephropathy. The adversarial correlation loss is innovatively presented to guide a parallel convolutional neural network. In this well- designed loss function, bias between the prediction and the label was take into account while the relationship among different categories is well-aligned. Glomerular lesions in this study are divided into five subcategories, Neg (Negative samples such as tubule and artery), SS (sclerosis involving a portion of the glomerular tuft), GS (sclerosis involving 100% of the tuft), C (build-up of more than two layers of cells within Bowman's space, often with fibrin and collagen deposition) and NOA (none of above). Our model with 93.0% accuracy and 92.9% Fl-score for these five categories has proved superior to other models through experimental results.
There is a large body of literature linking anatomic and geometric characteristics of kidney tumors to perioperative and oncologic outcomes. Semantic segmentation of these tumors and their host ...kidneys is a promising tool for quantitatively characterizing these lesions, but its adoption is limited due to the manual effort required to produce high-quality 3D segmentations of these structures. Recently, methods based on deep learning have shown excellent results in automatic 3D segmentation, but they require large datasets for training, and there remains little consensus on which methods perform best. The 2019 Kidney and Kidney Tumor Segmentation challenge (KiTS19) was a competition held in conjunction with the 2019 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) which sought to address these issues and stimulate progress on this automatic segmentation problem. A training set of 210 cross sectional CT images with kidney tumors was publicly released with corresponding semantic segmentation masks. 106 teams from five continents used this data to develop automated systems to predict the true segmentation masks on a test set of 90 CT images for which the corresponding ground truth segmentations were kept private. These predictions were scored and ranked according to their average So rensen-Dice coefficient between the kidney and tumor across all 90 cases. The winning team achieved a Dice of 0.974 for kidney and 0.851 for tumor, approaching the inter-annotator performance on kidney (0.983) but falling short on tumor (0.923). This challenge has now entered an "open leaderboard" phase where it serves as a challenging benchmark in 3D semantic segmentation.
Ultrasound (US)‐mediated sonodynamic therapy (SDT) has the advantages of non‐invasiveness and deep tissue penetration. Nanosystems are prominently used in sonosensitization; however, most ...nano‐sonosensitizers have a low reactive oxygen species (ROS) yield, thus restraining the application of SDT. Sodium molybdenum bronze nanoparticles (SMB NPs) with rich oxygen vacancies are developed and interlayer gaps of molybdenum trioxide nanobelts are expanded. Owing to the increased oxygen vacancy density and wide interlayer gap‐induced narrower band gap of SMB NPs, the electrons (e–) and holes (h+) generated by US are separated more rapidly, and oxygen vacancies prevent electrons–holes recombination under US irradiation. SMB NPs exhibit a second near‐infrared (NIR‐II) photothermal effect to promote the generation of ROS by the sonosensitizer. The SMB NPs system is successfully realized to eliminate Staphylococcus aureus (S. aureus) and dissipate biofilm. Therefore, multimodal therapy using SMB NPs serves as an effective and promising regimen for deep‐seated bacterial infections. The newly developed Mo‐based sonosensitizer is presented for the first time to demonstrate excellent antimicrobial activity through hyperthermia‐promoting SDT therapeutics. This work proposes a novel strategy in the field of NIR‐II photo‐amplified SDT with Mo‐based materials for bacterial eradication and other important biomedical applications.
The newly developed Mo‐based sonosensitizer is presented for the first time to demonstrate excellent eradicate bacteria in deep‐seated infected tissues in vivo through photothermal‐amplified sonodynamic therapy (SDT) to accelerate abscess healing, which can open a new window in the field of NIR‐II photo‐amplified SDT with Mo‐based materials for bacterial eradication and other biomedical applications.