Search

Background. Continuous wavelet transform (CWT) based scalogram can be used for photoplethysmography (PPG) signal transformation to classify blood pressure (BP) with deep learning. We aimed to investigate the determinants that can improve the accuracy of BP classification based on PPG and deep learning and establish a better algorithm for the prediction. Methods. The dataset from PhysioNet was accessed to extract raw PPG signals for testing and its corresponding BPs as category labels. The BP category of normal or abnormal followed the criteria of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Hypertension Guidelines. The PPG signals were transformed into 224 ∗ 224 ∗ 3-pixel scalogram via different CWTs and segment units. All of them are fed into different convolutional neural networks (CNN) for training and validation. The receiver-operating characteristic and loss and accuracy curves were used to evaluate and compare the performance of different methods. Results. Both wavelet type and segment length could affect the accuracy, and Cgau1 wavelet and segment-300 revealed the best performance (accuracy 90%) without obvious overfitting. This method performed better than previously reported MATLAB Morse wavelet transformed scalogram on both of our proposed CNN and CNN-GoogLeNet. Conclusions. We have established a new algorithm with high accuracy to predict BP classification from PPG via matching of CWT type and segment length, which is a promising solution for rapid prediction of BP classification from real-time processing of PPG signal on a wearable device.

We proposed and experimentally demonstrated an optical funnel for living cells trap based on an annular-core optical fiber. The proposed optical funnel employed the annular dark field to trap the living cell to avoid the laser damages. To perform the dark field trap, we placed the living cells in the glycerol solution. The refractive index of glycerol liquid was higher than yeast cells, which helped to perform the dark field trap; the viscosity of glycerol was significant, which helped to perform the viscosity-assisted 3-D trapping. The experimental results showed that the donut shape intensity field introduced by a single fiber probe performed 3D trapping easily and efficiently. Such funnel-shaped hollow conical beam could find ample potentials in optical manipulation on biological living cells, avoiding optical damages. All-fiber and seamlessly integrated structure of the proposed scheme could find ample potentials in manipulating biological cells.

Respiratory liver motion is one of the major issues affecting the liver interventions, such as biopsy and ablation, in the clinical practice. Traditional 4D liver model methods using magnetic reson...

Purpose: In respiratory motion modeling for liver interventions, the respiratory signal is usually obtained by using special tracking devices to monitor external skin. However, due to intrinsic limits and cost consideration of these tracking devices, a purely ultrasound image-based approach to tracking the signal is a more feasible option. Methods: In this study, a novel image-based method is proposed to obtain the respiratory signal directly from 2D ultrasound images by automatically identifying and tracking the liver boundary. The boundary identification is a multistage process, which is the key to utilize a Hessian matrix-based 2D filter to enhance the line-like liver boundary and weaken other liver tissues. For tracking the identified boundary, a new dynamic template matching technique is first applied to estimate 2D displacements, and a boundary-specific selection mechanism is then introduced to extract the respiratory signal from the 2D displacements. Results: The experiments demonstrate that their method can obtain accurate breathing signals, which are in key phases comparable to the manually annotation and highly consistent to the electromagnetic-tracked ground-truth signals (average correlation coefficients 0.9209 and statistically significant p < 0.01). Meanwhile, the experiments also prove their method can achieve high real-time performance of about 80–160 Hz. Conclusions: This method provides a good alternative to traditional external-landmark-based tracking methods and may be widely applied for respiratory compensation in ultrasound-guided liver interventions.

Respiratory gating has been widely applied for respiratory correction or compensation in image acquisition and image-guided interventions. A novel image-based method is proposed to extract respiratory signal directly from 2D ultrasound liver images. The proposed method utilizes a typical manifold learning method, based on local tangent space alignment based technique, to detect principal respiratory motion from a sequence of ultrasound images. This technique assumes all the images lying on a low-dimensional manifold embedding into the high-dimensional image space, constructs an approximate tangent space of each point to represent its local geometry on the manifold, and then aligns the local tangent spaces to form the global coordinate system, where the respiratory signal is extracted. The experimental results show that the proposed method can detect relatively accurate respiratory signal with high correlation coefficient (0.9775) with respect to the ground-truth signal by tracking external markers, and achieve satisfactory computing performance (2.3s for an image sequence of 256 frames). The proposed method is also used to create breathing-corrected 3D ultrasound images to demonstrate its potential application values.

The analysis of vascular structure from volumetric datasets plays a crucial role in many medical applications. Many segmentation algorithms have been designed to extract the vessel features. However, to date, these algorithms have not been efficiently evaluated and a large-scale manual analysis is always impossible. In this paper, we propose a quantitative evaluation function based on connectivity, overlap volume ratio, skeleton coincidence and branches structure error to deal with this evaluation task. The function is applicable to 3D tree-like structure segmentations and does not depend on the segmentation algorithms used. The performances of this evaluation function are tested on real liver datasets. The results show that the values of evaluation function are very close to the human scoring value (standard value), and the average value of relative errors is only 7.3% over all the eight datasets, while other evaluation measurements are 20% or more. So, it provides the greatest correlation with human qua...

In model-based respiratory motion estimation for the liver or other abdominal organs, the surrogate respiratory signal is usually obtained by using special tracking devices from skin or diaphragm, and subsequently applied to parameterize a 4D motion model for prediction or compensation. However, due to the intrinsic limits and economical costs of these tracking devices, the identification of the respiratory signal directly from intra-operative ultrasound images is a more attractive alternative. We propose a fast and robust method to extract the respiratory motion of the liver from an intra-operative 2D ultrasound image sequence. Our method employs a preprocess to remove speckle-like noises in the ultrasound images and utilizes the normalized cross-correlation to measure the image similarity fast. More importantly, we present a novel adaptive search strategy, which makes full use of the inter-frame dependency of the image sequence. This search strategy narrows the search range of the optimal matching, thus greatly reduces the search time, and makes the matching process more robust and accurate. The experimental results on four volunteers demonstrate that our method is able to extract the respiratory signal from an image sequence of 256 image frames in 5 s. The quantitative evaluation using the correlation coefficient reveals that the respiratory motion, extracted near the liver boundaries and vessels, is highly consistent with the reference motion tracked by an EM device. Our method can use 2D ultrasound to track natural landmarks from the liver as surrogate respiratory signal and hence provide a feasible solution to replace special tracking devices.

Multidimensional adaptive sampling technique is crucial for generating high quality images with effects such as motion blur, depth-of-field and soft shadows, but it costs a lot of memory and computation time. We propose a novel kd-tree based parallel adaptive rendering approach. First, a two-level framework for adaptive sampling in parallel is introduced to reduce the computation time and control the memory cost: in the prepare stage, we coarsely sample the entire multidimensional space and use kd-tree structure to separate it into several multidimensional subspaces; in the main stage, each subspace is refined by a sub kd-tree and rendered in parallel. Second, novel kd-tree based strategies are introduced to measure space’s error value and generate anisotropic Poisson disk samples. The experimental results show that our algorithm produces better quality images than previous ones.

We propose and demonstrate a stable three-dimensional trap and manipulation of a micron-sized strongly absorbing particle in pure liquid glycerol by using a single tight focused TEM00 Gaussian beam. We employ a bottom-side bidirectional view observation system to observe the trapped particle. We use the light at 980 nm to trap the absorbing particle and the light at 532 nm to indicate the distribution of the temperature field around the trapped particle. The trapping position of the absorbing particle is related to the incident laser power; the lower the incident laser power, the longer the particle shift distance. Our approach provides full control over trapped absorbing particles and expands optical manipulation of strong absorbing particles into a liquid media.

Bokeh effect is an important characteristic for realistic image synthesis. However, existing bokeh rendering methods are incapable of simulating realistic bokeh effects due to not taking into account optical characteristics of real lenses, especially optical aberrations. In this paper, a novel realistic bokeh rendering method, based on an accurate camera lens model and distributed ray tracing, is presented. An optical analysis of the relationship between bokeh and optical aberrations, including spherical aberration, coma, astigmatism and field curvature, is firstly introduced. Based on this analysis, a physically-based camera lens model, which takes detailed lens prescription as input, is then introduced for accurately modeling the aberrations. The position and diameter of the entrance and exit pupils are calculated by tracing rays inside the lens for achieving efficient ray sampling, and a general sequential ray tracing algorithm is proposed to better combine with bidirectional ray tracing. Furthermore, correct integration between the lens model and bidirectional ray tracing is also analyzed. The rendering results demonstrate a variety of realistic bokeh effects caused by the aberrations.

In respiratory motion modeling for liver interventions, the respiratory signal is usually obtained by using special tracking devices to monitor external skin. However, due to intrinsic limits and cost consideration of these tracking devices, a purely ultrasound image-based approach to tracking the signal is a more feasible option.In this study, a novel image-based method is proposed to obtain the respiratory signal directly from 2D ultrasound images by automatically identifying and tracking the liver boundary. The boundary identification is a multistage process, which is the key to utilize a Hessian matrix-based 2D filter to enhance the line-like liver boundary and weaken other liver tissues. For tracking the identified boundary, a new dynamic template matching technique is first applied to estimate 2D displacements, and a boundary-specific selection mechanism is then introduced to extract the respiratory signal from the 2D displacements.The experiments demonstrate that their method can obtain accurate breathing signals, which are in key phases comparable to the manually annotation and highly consistent to the electromagnetic-tracked ground-truth signals (average correlation coefficients 0.9209 and statistically significant p0.01). Meanwhile, the experiments also prove their method can achieve high real-time performance of about 80-160 Hz.This method provides a good alternative to traditional external-landmark-based tracking methods and may be widely applied for respiratory compensation in ultrasound-guided liver interventions.

In respiratory motion modeling for the liver, the breathing pattern is usually obtained by using special tracking devices from skin or diaphragm, and subsequently applied as input to a 4D motion model for motion estimation. However, due to the intrinsic limits and economical costs of these tracking devices, the identification of the breathing pattern directly from intra-operative ultrasound images is a more attractive option. In this paper, a new method is proposed to automatically track the breathing pattern from 2D ultrasound image sequences of the liver. The proposed method firstly utilizes a Hessian matrix-based 2D line filter to identify the liver boundary, then uses an adaptive search strategy to in real-time match a template block centered inside the identified boundary, and consequently extract the translational motion of the boundary as the respiratory pattern. The experiments on four volunteers demonstrate that the respiratory pattern extracted by our method is highly consistent to those acquired by an EM tracking system with the correlation coefficient of at least 0.91.

Testing objects are important for the validation of developing biopsy systems. Unfortunately, they are very hard to obtain. Motivated by this issue, the purpose of this study is to develop a technique for the easy creation of a model to simulate tumors of different sizes inside porcine livers, which could be used for ultrasound image-guided liver biopsy amongst other applications, and evaluate its performance by comparing to the more widely-used approaches in-vivo and ex-vivo. In this study, a Vaseline-based tumor model, and a more widely-used agar-based tumor model to provide comparison with the proposed method were created and injected into porcine livers as biopsy targets. The clinician located simulated tumors using real-time 2D imaging under the guidance of a robotic arm to delivery the biopsy in ex-vivo and in-vivo experiments. The results show that the optimum tumor model was created from a mixture of Vaseline, glycerol, and barium sulfate which can be easily produced and injected. All Vaseline-based simulated tumors were of solid, palpable mass on gross examination, and ultrasound imaging revealed clearly visible lesions. The clinician successfully performed ultrasound image guided liver biopsy in all the trials (10/10) in the ex-vivo experiment, and 2 out of 3 trials (2/3) in the in-vivo experiment on this optimum tumor model. We described a novel technique of creating solid liver tumor models that can be used for ultrasound image-guided liver biopsy.

In this paper, an accurate and practical camera lens model is proposed to be applied in realistic rendering of lens-related effects. The optical modeling of this new model is firstly presented from two aspects: lens surface modeling and formulation of ray tracing equations. Then, a number of tunable models for controlling lens properties are introduced and combined together to determine overall imaging performance. An implementation framework for this lens model is presented from two important aspects: its internal working framework and a new rendering pipeline for integrating it into a general ray tracer. In contrast to existing lens models, this new one is characterized by its ability to accurately model the image formation process and its friendly tunable models to control its lens properties. As a consequence, it is capable of simulating complex lens-related effects without too much expertise on lens optics. Finally, multiple rendering experiments are performed to demonstrate the ability and usage of this novel model to simulate a variety of complex lens-related effects.

Lens dispersion is a kind of light dispersion phenomenon occurring inside physical lens systems, which caused by the dispersion property of lens materials. Lens dispersion creates a variety of colorful optical effects on the image, such as colored fringes and defocused highlights. Simulating these dispersion-related effects can improve realism of computer-generated images and also serve as special effects for artistic purpose. A few papers have been presented to model and simulate common light dispersion phenomenon, but none has referred to special lens dispersion. In our work, a novel method for simulating lens dispersion is proposed by combining a new realistic lens model and an efficient spectral rendering method. The new realistic lens model is extended from Wu's lens model [Wu et al. 2010] to model the light dispersion inside the lens, and the efficient spectral rendering method is based on the stratified wavelength cluster(SWC) technique, proposed by Evans and McCool [1999].

In the process of road extraction, the road region often mixes with some redundancies caused by the phenomenon of "different objects with the same spectrum," which exists widely in remote sensing images. To eliminate these redundancies and preserve neat and accurate road regions, a method based on skeleton analysis is proposed. At first, the shortest path of a specified road segment is constructed with mixed region skeletons. The road width at the skeleton point is then selected as the smoothing scale on this point, on which basis an accumulative and bidirectional smoothing is taken. With the smoothed path and its average width, a buffer algorithm reconstructs the road region. The results show that redundancies of the road can be eliminated correctly, and accurate road center lines and neat road regions can be obtained.