Interactive virtual 3D image reconstruction to assist renal surgery in patients with fusion anomalies of the kidney

Objective: Renal fusion anomalies are rare and usually present as horseshoe kidneys or crossed fusion ectopia. The complex renal anatomy seen in patients with these anomalies can present a challenge. Pre-operative planning is therefore paramount in the surgical management of these cases. Herein we report the use of interactive virtual three-dimensional (3D) reconstruction to aid renal surgery in patients with fusion anomalies of the kidney. Materials and Methods: A total of seven cases were performed between May 2016 and October 2020. 3D reconstruction was rendered by Innersight Labs using pre-operative computed tomography (CT) scans. Results: Five patients had malignant disease and two patients had benign pathology. Robotic and open operations were performed in four and three patients, respectively. Conclusion: The use of 3D reconstruction in the cases reported in this series allowed for the identification of variations in renal vasculature, and this informed the choice of operative approach. Oxford Centre for Evidence-Based Medicine Evidence Level: 4

Three-Dimensional Image Reconstruction for Virtual Talent Training Scene

In real training, the training conditions are often undesirable, and the use of equipment is severely limited. These problems can be solved by virtual practical training, which breaks the limit of space, lowers the training cost, while ensuring the training quality. However, the existing methods work poorly in image reconstruction, because they fail to consider the fact that the environmental perception of actual scene is strongly regular by nature. Therefore, this paper investigates the three-dimensional (3D) image reconstruction for virtual talent training scene. Specifically, a fusion network model was deigned, and the deep-seated correlation between target detection and semantic segmentation was discussed for images shot in two-dimensional (2D) scenes, in order to enhance the extraction effect of image features. Next, the vertical and horizontal parallaxes of the scene were solved, and the depth-based virtual talent training scene was reconstructed three dimensionally, based on the continuity of scene depth. Finally, the proposed algorithm was proved effective through experiments.

Assessing the Feasibility of Using Kinect 3D Images to Predict Light Lamb Carcasses Composition from Leg Volume

This study aimed to evaluate the accuracy of the leg volume obtained by the Microsoft Kinect sensor to predict the composition of light lamb carcasses. The trial was performed on carcasses of twenty-two male lambs (17.6 ± 1.8 kg, body weight). The carcasses were split into eight cuts, divided into three groups according to their commercial value: high-value, medium value, and low-value group. Linear, area, and volume of leg measurements were obtained to predict carcass and cuts composition. The leg volume was acquired by two different methodologies: 3D image reconstruction using a Microsoft Kinect sensor and Archimedes principle. The correlation between these two leg measurements was significant (r = 0.815, p < 0.01). The models to predict cuts and carcass traits that include leg Kinect 3D sensor volume are very good in predicting the weight of the medium value and leg cuts (R2 of 0.763 and 0.829, respectively). Furthermore, the model, which includes the Kinect leg volume, explained 85% of its variation for the carcass muscle. The results of this study confirm the good ability to estimate cuts and carcass traits of light lamb carcasses with leg volume obtained with the Kinect 3D sensor.

Research on the Application of Visual Sensing Technology in Art Education

It is essential to have a new understanding of the development of visual sensing technology in digital image art at this stage, in order to make traditional art education have new professional ability teaching. Based on the current research results in related fields, a three-dimensional (3D) image visual communication system based on digital image automatic reconstruction is proposed with two schemes as the premise. In scheme 1, the hardware part is divided into two modules. The hardware used by the analysis of the 3D image layer module is the HUJ-23 3D image processor. The acquisition of a 3D image layer module uses the hardware of a realistic infrared camera. The software of the system consists of two parts: a 3D image computer expression module and a 3D image reconstruction module. A simulation platform is established. The test data of 3D image reconstruction accuracy and visual communication integrity of the designed system show that both of them show a good trend. In scheme 2, regarding digital image processing, the 3D image visual perception reconstruction is affected by the modeling conditions, and some images are incomplete and damaged. The depth camera and image processor that can be used in the visual communication technology are selected, and their internal parameters are modified to borrow them in the original system hardware. Gaussian filtering model combined with scale-invariant feature transform (SIFT) feature point extraction algorithm is adopted to select image feature points. Previous system reconstruction technology is used to upgrade the 3D digital image, and the feature point detection equation is adopted to detect the accuracy of the upgraded results. Based on the above hardware and software research, the 3D digital image system based on visual communication is successfully upgraded. The test platform is established, and the test samples are selected. Unlike the previous systems, the 3D image reconstruction accuracy of the designed visual communication system can be as high as 98%; the upgraded system has better image integrity and stronger performance than the previous systems and achieves higher visual sensing technology. In art education, it can provide a new content perspective for digital image art teaching.

Development of open-source software for free-hand 3D vascular ultrasound: Dialysis fistula application

Background: The arteriovenous fistula (AVF) is the preferred vascular access for End Stage Renal Disease, having superior patency and lower infection risks than prosthetic graft and catheter access. When AVF dysfunction or delayed maturation does occur, the gold standard for diagnosis is the fistula angiogram (a.k.a. fistulogram). 3D ultrasound is available for obstetrical and other specialized uses, but it is cost prohibitive and has a field of view that is too small to cover the region of interest for the dialysis fistula application. We sought to develop a point of care 3D solution using freehand 2D ultrasound data acquisition. Methods: We developed open-source software for 3D image reconstruction and projection of an angiogram-like image of the vascular access using a 2D freehand ultrasound scanner. We evaluated this software by comparing the ultrasound “sono-angiogram” images to fistulogram images in five subjects, using visual inspection and by applying the Percent of Exact Match (PEM) as a statistic test. Results: The sono-angiograms showed identifiable characteristics that matched the fistulogram results in all five subjects. The PEM ranged between 42.8% and 77.0%, with Doppler and grayscale ultrasound data, showing complementary advantages and disadvantages when used for sono-angiogram image construction. Motion from freehand ultrasound acquisition was a significant source of mismatch. 3D image generation is a potential advantage with ultrasound data. Conclusions: While further work is needed to improve the accuracy with free hand scanning, fistulogram-like “sono-angiograms” can be generated using point of care 2D ultrasound. Methods such as these may be able to assist in point-of-care diagnosis in the future. The software is open-source, and importantly, the ultrasound data used are non-proprietary and available from any standard ultrasound machine. The simplicity and accessibility of this approach warrant further study.

Study on 3D Image Reconstruction Model of Sparring Action Based on Graph Neural Network (GNN)

With the advent of the information age, human demand for information is increasing day by day. The emergence of the concept of big data has triggered a new round of technological revolution, and visual information plays an important role in information. In order to obtain a better 3D model, this paper studies the reconstruction model of training motion 3D images based on a graphical neural network algorithm. This paper studies the problem of Sanda from the following two aspects. First, we try to apply two deep learning algorithms, graphical neural network and recurrent neural network, to the boxing movement recognition task and compare the effects with quadratic discriminant analysis and support vector machine. By comparing and analyzing the influence of different network structures on the deep learning algorithm, it is concluded that recurrent neural network has more practical application advantages than graph neural network in network structure parameter tuning.

Three-stage segmentation of lung region from CT images using deep neural networks

Background Lung region segmentation is an important stage of automated image-based approaches for the diagnosis of respiratory diseases. Manual methods executed by experts are considered the gold standard, but it is time consuming and the accuracy is dependent on radiologists’ experience. Automated methods are relatively fast and reproducible with potential to facilitate physician interpretation of images. However, these benefits are possible only after overcoming several challenges. The traditional methods that are formulated as a three-stage segmentation demonstrate promising results on normal CT data but perform poorly in the presence of pathological features and variations in image quality attributes. The implementation of deep learning methods that can demonstrate superior performance over traditional methods is dependent on the quantity, quality, cost and the time it takes to generate training data. Thus, efficient and clinically relevant automated segmentation method is desired for the diagnosis of respiratory diseases. Methods We implement each of the three stages of traditional methods using deep learning methods trained on five different configurations of training data with ground truths obtained from the 3D Image Reconstruction for Comparison of Algorithm Database (3DIRCAD) and the Interstitial Lung Diseases (ILD) database. The data was augmented with the Lung Image Database Consortium (LIDC-IDRI) image collection and a realistic phantom. A convolutional neural network (CNN) at the preprocessing stage classifies the input into lung and none lung regions. The processing stage was implemented using a CNN-based U-net while the postprocessing stage utilize another U-net and CNN for contour refinement and filtering out false positives, respectively. Results The performance of the proposed method was evaluated on 1230 and 1100 CT slices from the 3DIRCAD and ILD databases. We investigate the performance of the proposed method on five configurations of training data and three configurations of the segmentation system; three-stage segmentation and three-stage segmentation without a CNN classifier and contrast enhancement, respectively. The Dice-score recorded by the proposed method range from 0.76 to 0.95. Conclusion The clinical relevance and segmentation accuracy of deep learning models can improve though deep learning-based three-stage segmentation, image quality evaluation and enhancement as well as augmenting the training data with large volume of cheap and quality training data. We propose a new and novel deep learning-based method of contour refinement.