The use of pencil ionization chamber with temporal readout capabilities to measure CT beam full width half maximum

BACKGROUND

Measurement of Computed Tomography (CT) beam width is required by accrediting and regulating bodies for routine physics evaluations due to its direct correlation to patient dose. Current methods for performing CT beam width measurement require special hardware, software, and/or consumable films. Today, most 100-mm pencil chambers with a digital interface used to evaluate Computed Tomography Dose Index (CTDIvol) have a sufficiently high sampling rate to reconstruct a high-resolution dose profile for any acquisition mode.

PURPOSE

The goal of this study is to measure the CT beam width from the sampled dose profile under a single helical acquisition with the 100-mm pencil chamber used for CTDIvol measurements.

METHODS

The dose profiles for different scanners were measured for helical scans with varying collimation settings using a 100-mm pencil chamber placed at the isocenter and co-moving with the patient table. The measured dose profiles from the 100-mm pencil chamber were corrected for table attenuation by extracting a periodic correction function (PCF) to eliminate table interference. The corrected dose profiles were then deconvolved with the response function of the chamber to compute the beam profile. The beam width was defined by the full width half maximum (FWHM) of the resulting beam profile. Reference dose profiles were also measured using Gafchromic film for comparison.

RESULTS

The beam widths, estimated using the innovative deconvolution method from the 100-mm pencil chamber, exhibit an average percentage difference of 1.6 ± 1.8 when compared with measurements obtained through Gafchromic film for beam width assessment.

CONCLUSION

The proposed approach to deconvolve the pencil chamber response demonstrates the potential of obtaining the CT beam width at high accuracy without the need of special hardware, software, or consumable films. This technique can improve workflow for routine performance evaluation of CT systems.

ArreSTick motif controls β-arrestin-binding stability and extends phosphorylation-dependent β-arrestin interactions to non-receptor proteins

The binding and function of β-arrestins are regulated by specific phosphorylation motifs present in G protein-coupled receptors (GPCRs). However, the exact arrangement of phosphorylated amino acids responsible for establishing a stable interaction remains unclear. We employ a 1D sequence convolution model trained on GPCRs with established β-arrestin-binding properties. With this approach, amino acid motifs characteristic of GPCRs that form stable interactions with β-arrestins can be identified, a pattern that we name "arreSTick." Intriguingly, the arreSTick pattern is also present in numerous non-receptor proteins. Using proximity biotinylation assay and mass spectrometry analysis, we demonstrate that the arreSTick motif controls the interaction between many non-receptor proteins and β-arrestin2. The HIV-1 Tat-specific factor 1 (HTSF1 or HTATSF1), a nuclear transcription factor, contains the arreSTick pattern, and its subcellular localization is influenced by β-arrestin2. Our findings unveil a broader role for β-arrestins in phosphorylation-dependent interactions, extending beyond GPCRs to encompass non-receptor proteins as well.

CLINet: A novel deep learning network for ECG signal classification

Machine learning is poised to revolutionize medicine with algorithms that spot cardiac arrhythmia. An automated diagnostic approach can boost the efficacy of diagnosing life-threatening arrhythmia disorders in routine medical procedures. In this paper, we propose a deep learning network CLINet for ECG signal classification. Our network uses convolution, LSTM and involution layers to bring their unique advantages together. For both convolution and involution layers, we use multiple, large size kernels for multi-scale representation learning. CLINet does not require complicated pre-processing and can handle electrocardiograms of any length. Our network achieves 99.90% accuracy on the ICCAD dataset and 99.94% accuracy on the MIT-BIH dataset. With only 297 K parameters, our model can be easily embedded in smart wearable devices. The source code of CLINet is available at https://github.com/CandleLabAI/CLINet-ECG-Classification-2024.

A convolution-based in vitro-in vivo correlation model for methylphenidate hydrochloride delayed-release and extended-release capsule

Delayed-release and extended-release methylphenidate hydrochloride (JORNAY PM®) is a novel capsule formulation of methylphenidate hydrochloride, used to treat attention deficit hyperactivity disorder in patients 6 years and older. In this paper, we develop a Level A in vitro-in vivo correlation (IVIVC) model for extended-release methylphenidate hydrochloride to support post-approval manufacturing changes by evaluating a point-to-point correlation between the fraction of drug dissolved in vitro and the fraction of drug absorbed in vivo. Dissolution data from an in vitro study of three different release formulations: fast, medium, and slow, and pharmacokinetic data from two in vivo studies were used to develop an IVIVC model using a convolution-based approach. The time-course of the drug concentration resulting from an arbitrary dose was considered as a function of the in vivo drug absorption and the disposition and elimination processes defined by the unit impulse response function using the convolution integral. An IVIVC was incorporated in the model due to the temporal difference seen in the scatterplots of the estimated fraction of drug absorbed in vivo and the fraction of drug dissolved in vitro and Levy plots. Finally, the IVIVC model was subjected to evaluation of internal predictability. This IVIVC model can be used to predict in vivo profiles for different in vitro profiles of extended-release methylphenidate hydrochloride.

TEPCAM: Prediction of T-cell receptor-epitope binding specificity via interpretable deep learning

The recognition of T-cell receptor (TCR) on the surface of T cell to specific epitope presented by the major histocompatibility complex is the key to trigger the immune response. Identifying the binding rules of TCR-epitope pair is crucial for developing immunotherapies, including neoantigen vaccine and drugs. Accurate prediction of TCR-epitope binding specificity via deep learning remains challenging, especially in test cases which are unseen in the training set. Here, we propose TEPCAM (TCR-EPitope identification based on Cross-Attention and Multi-channel convolution), a deep learning model that incorporates self-attention, cross-attention mechanism, and multi-channel convolution to improve the generalizability and enhance the model interpretability. Experimental results demonstrate that our model outperformed several state-of-the-art models on two challenging tasks including a strictly split dataset and an external dataset. Furthermore, the model can learn some interaction patterns between TCR and epitope by extracting the interpretable matrix from cross-attention layer and mapping them to the three-dimensional structures. The source code and data are freely available at https://github.com/Chenjw99/TEPCAM.

Vibration-Based Wear Condition Estimation of Journal Bearings Using Convolutional Autoencoders

Predictive maintenance is considered a proactive approach that capitalizes on advanced sensing technologies and data analytics to anticipate potential equipment malfunctions, enabling cost savings and improved operational efficiency. For journal bearings, predictive maintenance assumes critical significance due to the inherent complexity and vital role of these components in mechanical systems. The primary objective of this study is to develop a data-driven methodology for indirectly determining the wear condition by leveraging experimentally collected vibration data. To accomplish this goal, a novel experimental procedure was devised to expedite wear formation on journal bearings. Seventeen bearings were tested and the collected sensor data were employed to evaluate the predictive capabilities of various sensors and mounting configurations. The effects of different downsampling methods and sampling rates on the sensor data were also explored within the framework of feature engineering. The downsampled sensor data were further processed using convolutional autoencoders (CAEs) to extract a latent state vector, which was found to exhibit a strong correlation with the wear state of the bearing. Remarkably, the CAE, trained on unlabeled measurements, demonstrated an impressive performance in wear estimation, achieving an average Pearson coefficient of 91% in four different experimental configurations. In essence, the proposed methodology facilitated an accurate estimation of the wear of the journal bearings, even when working with a limited amount of labeled data.

Generalized Quantum Convolution for Multidimensional Data

The convolution operation plays a vital role in a wide range of critical algorithms across various domains, such as digital image processing, convolutional neural networks, and quantum machine learning. In existing implementations, particularly in quantum neural networks, convolution operations are usually approximated by the application of filters with data strides that are equal to the filter window sizes. One challenge with these implementations is preserving the spatial and temporal localities of the input features, specifically for data with higher dimensions. In addition, the deep circuits required to perform quantum convolution with a unity stride, especially for multidimensional data, increase the risk of violating decoherence constraints. In this work, we propose depth-optimized circuits for performing generalized multidimensional quantum convolution operations with unity stride targeting applications that process data with high dimensions, such as hyperspectral imagery and remote sensing. We experimentally evaluate and demonstrate the applicability of the proposed techniques by using real-world, high-resolution, multidimensional image data on a state-of-the-art quantum simulator from IBM Quantum.

Deep Multi-Branch CNN Architecture for Early Alzheimer's Detection from Brain MRIs

Alzheimer's disease (AD) is a neurodegenerative disease that can cause dementia and result in a severe reduction in brain function, inhibiting simple tasks, especially if no preventative care is taken. Over 1 in 9 Americans suffer from AD-induced dementia, and unpaid care for people with AD-related dementia is valued at USD 271.6 billion. Hence, various approaches have been developed for early AD diagnosis to prevent its further progression. In this paper, we first review other approaches that could be used for the early detection of AD. We then give an overview of our dataset and propose a deep convolutional neural network (CNN) architecture consisting of 7,866,819 parameters. This model comprises three different convolutional branches, each having a different length. Each branch is comprised of different kernel sizes. This model can predict whether a patient is non-demented, mild-demented, or moderately demented with a 99.05% three-class accuracy. In summary, the deep CNN model demonstrated exceptional accuracy in the early diagnosis of AD, offering a significant advancement in the field and the potential to improve patient care.

A convolution-superposition fluence model for the Siemens HD120 multi leaf collimator with application to a 3D VMAT dose engine

Purpose. To construct a fast-calculating fluence modelfor the Siemens HD120 multi leaf collimator (MLC) using convolution-superposition techniques, and to develop a 3D VMAT dose engine using this fluence model.  This work offers analternative to time-consuming open-source Monte Carlo simulations for thosedeveloping in-house dose-calculating software for research or clinical needs.Methods. EPID-acquired images of sweeping-window and sweeping-checker field profiles were used to commission transmission, 2 Dinterleaf leakage, and tongue-and-groove maps specific to the HD120 MLC.  These maps, along with a 2D head-scattermodel were incorporated into a convolution-superposition algorithm to provide a fluence model for the HD120 MLC. This fluence model was used to develop a 3D VMAT dose engine, where 3D pre-computed 6MV dose kernels (EGSnrc) and a 3D fluence curvature-correction map were incorporatedto calculate 3D VMAT doses in a 22 cm diameter cylindrical phantom. Four VMAT patient plans witha large range of PTV sizes (36 cc to 604 cc) were chosen to test the fluence model and dose engine.Results. Excellent agreement was observed between the simulated commissioning fields and measured EPID-responses. 2D 2%/2 mm gamma analysis yielded a 98.9% pass rate for 1 cm, 2 cm, and 4 cm sweeping-window fields. 2D 2%/2mm gamma analysis for outer/inner MLC leaves yielded 89.1%/77.0% and 95.2%/91.1% pass rates from 1 cm and 2 cm sweeping-checker fields. Mean 3%/3 mm gamma analysis showed excellent agreement between our dose engine and Eclipse (Acuros) regardless of PTV size: 98.7% pass rate, with 95.1% pass rate in the high-dose volume.  Fluence calculation times were13.6 seconds per dynamic MLC field and 1.4 minutes/arc for 3D VMAT dose on a standard PC. Conclusions. A fast-calculating convolution-superposition fluence model has been commissioned for the Siemens HD120 MLC and incorporatedinto a 3D VMAT dose engine.  This work can be used to facilitate the development of fast in-house dose-calculating software.

A Deep Learning Model Based on Capsule Networks for COVID Diagnostics through X-ray Images

X-ray diagnostics are widely used to detect various diseases, such as bone fracture, pneumonia, or intracranial hemorrhage. This method is simple and accessible in most hospitals, but requires an expert who is sometimes unavailable. Today, some diagnoses are made with the help of deep learning algorithms based on Convolutional Neural Networks (CNN), but these algorithms show limitations. Recently, Capsule Networks (CapsNet) have been proposed to overcome these problems. In our work, CapsNet is used to detect whether a chest X-ray image has disease (COVID or pneumonia) or is healthy. An improved model called DRCaps is proposed, which combines the advantage of CapsNet and the dilation rate (dr) parameter to manage images with 226 × 226 resolution. We performed experiments with 16,669 chest images, in which our model achieved an accuracy of 90%. Furthermore, the model size is 11M with a reconstruction stage, which helps to avoid overfitting. Experiments show how the reconstruction stage works and how we can avoid the max-pooling operation for networks with a stride and dilation rate to downsampling the convolution layers. In this paper, DRCaps is superior to other comparable models in terms of accuracy, parameters, and image size handling. The main idea is to keep the model as simple as possible without using data augmentation or a complex preprocessing stage.

Multi-Task Segmentation and Classification Network for Artery/Vein Classification in Retina Fundus

Automatic classification of arteries and veins (A/V) in fundus images has gained considerable attention from researchers due to its potential to detect vascular abnormalities and facilitate the diagnosis of some systemic diseases. However, the variability in vessel structures and the marginal distinction between arteries and veins poses challenges to accurate A/V classification. This paper proposes a novel Multi-task Segmentation and Classification Network (MSC-Net) that utilizes the vessel features extracted by a specific module to improve A/V classification and alleviate the aforementioned limitations. The proposed method introduces three modules to enhance the performance of A/V classification: a Multi-scale Vessel Extraction (MVE) module, which distinguishes between vessel pixels and background using semantics of vessels, a Multi-structure A/V Extraction (MAE) module that classifies arteries and veins by combining the original image with the vessel features produced by the MVE module, and a Multi-source Feature Integration (MFI) module that merges the outputs from the former two modules to obtain the final A/V classification results. Extensive empirical experiments verify the high performance of the proposed MSC-Net for retinal A/V classification over state-of-the-art methods on several public datasets.

DeepAlgPro: an interpretable deep neural network model for predicting allergenic proteins

Allergies have become an emerging public health problem worldwide. The most effective way to prevent allergies is to find the causative allergen at the source and avoid re-exposure. However, most of the current computational methods used to identify allergens were based on homology or conventional machine learning methods, which were inefficient and still had room to be improved for the detection of allergens with low homology. In addition, few methods based on deep learning were reported, although deep learning has been successfully applied to several tasks in protein sequence analysis. In the present work, a deep neural network-based model, called DeepAlgPro, was proposed to identify allergens. We showed its great accuracy and applicability to large-scale forecasts by comparing it to other available tools. Additionally, we used ablation experiments to demonstrate the critical importance of the convolutional module in our model. Moreover, further analyses showed that epitope features contributed to model decision-making, thus improving the model's interpretability. Finally, we found that DeepAlgPro was capable of detecting potential new allergens. Overall, DeepAlgPro can serve as powerful software for identifying allergens.

Investigating Beta-Variational Convolutional Autoencoders for the Unsupervised Classification of Chest Pneumonia

The world's population is increasing and so is the challenge on existing healthcare infrastructure to cope with the growing demand in medical diagnosis and evaluation. Although human experts are primarily tasked with the diagnosis of different medical conditions, artificial intelligence (AI)-assisted diagnoses have become considerably useful in recent times. One of the critical lung infections, which requires early diagnosis and subsequent treatment to reduce the mortality rate, is pneumonia. There are different methods for obtaining a pneumonia diagnosis; however, the adoption of chest X-rays is popular since it is non-invasive. The AI systems for a pneumonia diagnosis using chest X-rays are often built on supervised machine-learning (ML) models, which require labeled datasets for development. However, collecting labeled datasets is sometimes infeasible due to constraints such as human resources, cost, and time. As such, the problem that we address in this paper is the unsupervised classification of pneumonia using unsupervised ML models including the beta-variational convolutional autoencoder (β-VCAE) and other variants, such as convolutional autoencoders (CAE), denoising convolutional autoencoders (DCAE), and sparse convolutional autoencoders (SCAE). Namely, the pneumonia classification problem is cast into an anomaly detection to develop the aforementioned ML models. The experimental results show that pneumonia can be diagnosed with high recall, precision, f₁-score, and f₂-score using the proposed unsupervised models. In addition, we observe that the proposed models are competitive with the state-of-the-art models, which are trained on a labeled dataset.

CNN Hardware Accelerator for Real-Time Bearing Fault Diagnosis

This paper introduces a one-dimensional convolutional neural network (CNN) hardware accelerator. It is crafted to conduct real-time assessments of bearing conditions using economical hardware components, implemented on a field-programmable gate array evaluation platform, negating the necessity to transfer data to a cloud-based server. The adoption of the down-sampling technique augments the visible time span of the signal in an image, thereby enhancing the accuracy of the bearing condition diagnosis. Furthermore, the proposed method of quaternary quantization enhances precision and shrinks the memory demand for the neural network model by an impressive 89%. Provided that the current signal data sampling rate stands at 64 K samples/s, the proposed design can accomplish real-time fault diagnosis at a clock frequency of 100 MHz. Impressively, the response duration of the proposed CNN hardware system is a mere 0.28 s, with the fault diagnosis precision reaching a remarkable 96.37%.

Quantum entropy and central limit theorem

We introduce a framework to study discrete-variable (DV) quantum systems based on qudits. It relies on notions of a mean state (MS), a minimal stabilizer-projection state (MSPS), and a new convolution. Some interesting consequences are: The MS is the closest MSPS to a given state with respect to the relative entropy; the MS is extremal with respect to the von Neumann entropy, demonstrating a "maximal entropy principle in DV systems." We obtain a series of inequalities for quantum entropies and for Fisher information based on convolution, giving a "second law of thermodynamics for quantum convolutions." We show that the convolution of two stabilizer states is a stabilizer state. We establish a central limit theorem, based on iterating the convolution of a zero-mean quantum state, and show this converges to its MS. The rate of convergence is characterized by the "magic gap," which we define in terms of the support of the characteristic function of the state. We elaborate on two examples: the DV beam splitter and the DV amplifier.

Neuromorphic-P2M: processing-in-pixel-in-memory paradigm for neuromorphic image sensors

Edge devices equipped with computer vision must deal with vast amounts of sensory data with limited computing resources. Hence, researchers have been exploring different energy-efficient solutions such as near-sensor, in-sensor, and in-pixel processing, bringing the computation closer to the sensor. In particular, in-pixel processing embeds the computation capabilities inside the pixel array and achieves high energy efficiency by generating low-level features instead of the raw data stream from CMOS image sensors. Many different in-pixel processing techniques and approaches have been demonstrated on conventional frame-based CMOS imagers; however, the processing-in-pixel approach for neuromorphic vision sensors has not been explored so far. In this work, for the first time, we propose an asynchronous non-von-Neumann analog processing-in-pixel paradigm to perform convolution operations by integrating in-situ multi-bit multi-channel convolution inside the pixel array performing analog multiply and accumulate (MAC) operations that consume significantly less energy than their digital MAC alternative. To make this approach viable, we incorporate the circuit's non-ideality, leakage, and process variations into a novel hardware-algorithm co-design framework that leverages extensive HSpice simulations of our proposed circuit using the GF22nm FD-SOI technology node. We verified our framework on state-of-the-art neuromorphic vision sensor datasets and show that our solution consumes ~2× lower backend-processor energy while maintaining almost similar front-end (sensor) energy on the IBM DVS128-Gesture dataset than the state-of-the-art while maintaining a high test accuracy of 88.36%.

Butterfly Transforms for Efficient Representation of Spatially Variant Point Spread Functions in Bayesian Imaging

Bayesian imaging algorithms are becoming increasingly important in, e.g., astronomy, medicine and biology. Given that many of these algorithms compute iterative solutions to high-dimensional inverse problems, the efficiency and accuracy of the instrument response representation are of high importance for the imaging process. For efficiency reasons, point spread functions, which make up a large fraction of the response functions of telescopes and microscopes, are usually assumed to be spatially invariant in a given field of view and can thus be represented by a convolution. For many instruments, this assumption does not hold and degrades the accuracy of the instrument representation. Here, we discuss the application of butterfly transforms, which are linear neural network structures whose sizes scale sub-quadratically with the number of data points. Butterfly transforms are efficient by design, since they are inspired by the structure of the Cooley-Tukey fast Fourier transform. In this work, we combine them in several ways into butterfly networks, compare the different architectures with respect to their performance and identify a representation that is suitable for the efficient representation of a synthetic spatially variant point spread function up to a 1% error. Furthermore, we show its application in a short synthetic example.

PVDF Membrane-Based Dual-Channel Acoustic Sensor Integrating the Fabry-Pérot and Piezoelectric Effects

A resonant acoustic wave detector combined with Fabry-Pérot interference (FPI) and piezoelectric (PE) effects based on a polyvinylidene fluoride (PVDF) piezoelectric film was proposed to enhance the ability of the sensor to detect acoustic signals in a specific frequency band. The deformation of circular thin films was indicated by the interference and piezoelectric effects simultaneously, and the noise level was decreased by the real-time convolution of the two-way parallel signal. This study reveals that, at the film's resonance frequency, the minimum detection limits for the FPI and piezoelectric impacts on acoustic waves are 3.39 μPa/Hz^1/2 and 20.8 μPa/Hz^1/2, respectively. The convolution result shows that the background noise was reduced by 98.81% concerning the piezoelectric signal, and by 85.21% concerning the FPI signal. The convolution's signal-to-noise ratio (SNR) was several times greater than the other two signals at 10 mPa. Therefore, this resonance sensor, which the FPI and the piezoelectric effect synergistically enhance, can be applied to scenarios of acoustic wave detection in a specific frequency band and with ultrahigh sensitivity requirements.

Convolution, Correlation and Generalized Shift Operations Based on the Fresnel Transform

The Fresnel transform (FrT) is commonly used to describe the free-space propagation of optical waves. In this work, we present new definitions for the convolution, correlation and generalized shift operations based on the FrT. The generalized shift operation is defined by using simultaneous space and phase shifts. The generalized shift operation is useful for centred optical systems in the Fresnel domain (FrD) when the data distributions at the input plane of the optical system are shifted. The new convolution and correlation operations defined in terms of the FrT, the wavelength and the propagation distance, can be considered as a generalization of the usual convolution and correlation operations. The sampling theorem for distributions, whose resulting FrT has finite support, is formulated by using the new convolution operation introduced in this work and a new definition of the Dirac comb function. These new definitions and results could be applied to describe, design and implement optical processing systems related to the FrT. Finally, we present a centred optical systems used in holography and optical security systems that can be described or modelled by the new definitions of the operations proposed in this paper.

Smoothed Quantile Regression with Large-Scale Inference

Quantile regression is a powerful tool for learning the relationship between a response variable and a multivariate predictor while exploring heterogeneous effects. This paper focuses on statistical inference for quantile regression in the "increasing dimension" regime. We provide a comprehensive analysis of a convolution smoothed approach that achieves adequate approximation to computation and inference for quantile regression. This method, which we refer to as conquer, turns the non-differentiable check function into a twice-differentiable, convex and locally strongly convex surrogate, which admits fast and scalable gradient-based algorithms to perform optimization, and multiplier bootstrap for statistical inference. Theoretically, we establish explicit non-asymptotic bounds on estimation and Bahadur-Kiefer linearization errors, from which we show that the asymptotic normality of the conquer estimator holds under a weaker requirement on dimensionality than needed for conventional quantile regression. The validity of multiplier bootstrap is also provided. Numerical studies confirm conquer as a practical and reliable approach to large-scale inference for quantile regression. Software implementing the methodology is available in the R package conquer.

Deep Learning-Based Multiclass Brain Tissue Segmentation in Fetal MRIs

Fetal brain tissue segmentation is essential for quantifying the presence of congenital disorders in the developing fetus. Manual segmentation of fetal brain tissue is cumbersome and time-consuming, so using an automatic segmentation method can greatly simplify the process. In addition, the fetal brain undergoes a variety of changes throughout pregnancy, such as increased brain volume, neuronal migration, and synaptogenesis. In this case, the contrast between tissues, especially between gray matter and white matter, constantly changes throughout pregnancy, increasing the complexity and difficulty of our segmentation. To reduce the burden of manual refinement of segmentation, we proposed a new deep learning-based segmentation method. Our approach utilized a novel attentional structural block, the contextual transformer block (CoT-Block), which was applied in the backbone network model of the encoder-decoder to guide the learning of dynamic attentional matrices and enhance image feature extraction. Additionally, in the last layer of the decoder, we introduced a hybrid dilated convolution module, which can expand the receptive field and retain detailed spatial information, effectively extracting the global contextual information in fetal brain MRI. We quantitatively evaluated our method according to several performance measures: dice, precision, sensitivity, and specificity. In 80 fetal brain MRI scans with gestational ages ranging from 20 to 35 weeks, we obtained an average Dice similarity coefficient (DSC) of 83.79%, an average Volume Similarity (VS) of 84.84%, and an average Hausdorff95 Distance (HD95) of 35.66 mm. We also used several advanced deep learning segmentation models for comparison under equivalent conditions, and the results showed that our method was superior to other methods and exhibited an excellent segmentation performance.

Landslide Susceptibility Mapping by Fusing Convolutional Neural Networks and Vision Transformer

Landslide susceptibility mapping (LSM) is an important decision basis for regional landslide hazard risk management, territorial spatial planning and landslide decision making. The current convolutional neural network (CNN)-based landslide susceptibility mapping models do not adequately take into account the spatial nature of texture features, and vision transformer (ViT)-based LSM models have high requirements for the amount of training data. In this study, we overcome the shortcomings of CNN and ViT by fusing these two deep learning models (bottleneck transformer network (BoTNet) and convolutional vision transformer network (ConViT)), and the fused model was used to predict the probability of landslide occurrence. First, we integrated historical landslide data and landslide evaluation factors and analysed whether there was covariance in the landslide evaluation factors. Then, the testing accuracy and generalisation ability of the CNN, ViT, BoTNet and ConViT models were compared and analysed. Finally, four landslide susceptibility mapping models were used to predict the probability of landslide occurrence in Pingwu County, Sichuan Province, China. Among them, BoTNet and ConViT had the highest accuracy, both at 87.78%, an improvement of 1.11% compared to a single model, while ConViT had the highest F1-socre at 87.64%, an improvement of 1.28% compared to a single model. The results indicate that the fusion model of CNN and ViT has better LSM performance than the single model. Meanwhile, the evaluation results of this study can be used as one of the basic tools for landslide hazard risk quantification and disaster prevention in Pingwu County.

RBL-2H3 Mast Cell Receptor Dynamics in the Immunological Synapse

The RBL-2H3 mast cell immunological synapse dynamics is often simulated with reaction-diffusion and Fokker-Planck equations. The equations focus on how the cell synapse captures receptors following an immune response, where the receptor capture at the immunological site appears to be a delayed process. This article investigates the physical nature and mathematics behind such time-dependent delays. Using signal processing methods, convolution and cross-correlation-type delay capture simulations give a χ -squared range of 22 to 60, in good agreement with experimental results. The cell polarization event is offered as a possible explanation for these capture delays, where polarizing rates measure how fast the cell polarization event occurs. In the case of RBL-2H3 mast cells, polarization appears to be associated with cytoskeletal rearrangement; thus, both cytoskeletal and diffusional components are considered. From these simulations, a maximum polarizing rate ranging from 0.0057 s-2 to 0.031 s-2 is obtained. These results indicate that RBL-2H3 mast cells possess both temporal and spatial memory, and cell polarization is possibly linked to a Turing-type pattern formation.

YOLOv4-Tiny-Based Coal Gangue Image Recognition and FPGA Implementation

Nowadays, most of the deep learning coal gangue identification methods need to be performed on high-performance CPU or GPU hardware devices, which are inconvenient to use in complex underground coal mine environments due to their high power consumption, huge size, and significant heat generation. Aiming to resolve these problems, this paper proposes a coal gangue identification method based on YOLOv4-tiny and deploys it on the low-power hardware platform FPGA. First, the YOLOv4-tiny model is well trained on the computer platform, and the computation of the model is reduced through the 16-bit fixed-point quantization and the integration of a BN layer and convolution layer. Second, convolution and pooling IP kernels are designed on the FPGA platform to accelerate the computation of convolution and pooling, in which three optimization methods, including input and output channel parallelism, pipeline, and ping-pong operation, are used. Finally, the FPGA hardware system design of the whole algorithm is completed. The experimental results of the self-made coal gangue data set indicate that the precision of the algorithm proposed in this paper for coal gangue recognition on the FPGA platform are slightly lower than those of CPU and GPU, and the mAP value is 96.56%; the recognition speed of each image is 0.376 s, which is between those of CPU and GPU; the hardware power consumption of the FPGA platform is only 2.86 W; and the energy efficiency ratio is 10.42 and 3.47 times that of CPU and GPU, respectively.

Modeling the Dynamics of Mixture Toxicity and Effects of Organic Micropollutants in a Small River under Unsteady Flow Conditions

The presence of anthropogenic organic micropollutants in rivers poses a long-term threat to surface water quality. To describe and quantify the in-stream fate of single micropollutants, the advection-dispersion-reaction (ADR) equation has been used previously. Understanding the dynamics of the mixture effects and cytotoxicity that are cumulatively caused by micropollutant mixtures along their flow path in rivers requires a new concept. Thus, we extended the ADR equation from single micropollutants to defined mixtures and then to the measured mixture effects of micropollutants extracted from the same river water samples. Effects (single and mixture) are expressed as effect units and toxic units, the inverse of effect concentrations and inhibitory concentrations, respectively, quantified using a panel of in vitro bioassays. We performed a Lagrangian sampling campaign under unsteady flow, collecting river water that was impacted by a wastewater treatment plant (WWTP) effluent. To reduce the computational time, the solution of the ADR equation was expressed by a convolution-based reactive transport approach, which was used to simulate the dynamics of the effects. The dissipation dynamics of the individual micropollutants were reproduced by the deterministic model following first-order kinetics. The dynamics of experimental mixture effects without known compositions were captured by the model ensemble obtained through Bayesian calibration. The highly fluctuating WWTP effluent discharge dominated the temporal patterns of the effect fluxes in the river. Minor inputs likely from surface runoff and pesticide diffusion might contribute to the general effect and cytotoxicity pattern but could not be confirmed by the model-based analysis of the available effect and chemical data.

Data-driven emergence of convolutional structure in neural networks

Exploiting data invariances is crucial for efficient learning in both artificial and biological neural circuits. Understanding how neural networks can discover appropriate representations capable of harnessing the underlying symmetries of their inputs is thus crucial in machine learning and neuroscience. Convolutional neural networks, for example, were designed to exploit translation symmetry, and their capabilities triggered the first wave of deep learning successes. However, learning convolutions directly from translation-invariant data with a fully connected network has so far proven elusive. Here we show how initially fully connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs, resulting in localized, space-tiling receptive fields. These receptive fields match the filters of a convolutional network trained on the same task. By carefully designing data models for the visual scene, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs, which has long been recognized as the hallmark of natural images. We provide an analytical and numerical characterization of the pattern formation mechanism responsible for this phenomenon in a simple model and find an unexpected link between receptive field formation and tensor decomposition of higher-order input correlations. These results provide a perspective on the development of low-level feature detectors in various sensory modalities and pave the way for studying the impact of higher-order statistics on learning in neural networks.

Discrete Quadratic-Phase Fourier Transform: Theory and Convolution Structures

The discrete Fourier transform is considered as one of the most powerful tools in digital signal processing, which enable us to find the spectrum of finite-duration signals. In this article, we introduce the notion of discrete quadratic-phase Fourier transform, which encompasses a wider class of discrete Fourier transforms, including classical discrete Fourier transform, discrete fractional Fourier transform, discrete linear canonical transform, discrete Fresnal transform, and so on. To begin with, we examine the fundamental aspects of the discrete quadratic-phase Fourier transform, including the formulation of Parseval's and reconstruction formulae. To extend the scope of the present study, we establish weighted and non-weighted convolution and correlation structures associated with the discrete quadratic-phase Fourier transform.

TransConver: transformer and convolution parallel network for developing automatic brain tumor segmentation in MRI images

BACKGROUND

Medical image segmentation plays a vital role in computer-aided diagnosis (CAD) systems. Both convolutional neural networks (CNNs) with strong local information extraction capacities and transformers with excellent global representation capacities have achieved remarkable performance in medical image segmentation. However, because of the semantic differences between local and global features, how to combine convolution and transformers effectively is an important challenge in medical image segmentation.

METHODS

In this paper, we proposed TransConver, a U-shaped segmentation network based on convolution and transformer for automatic and accurate brain tumor segmentation in MRI images. Unlike the recently proposed transformer and convolution based models, we proposed a parallel module named transformer-convolution inception (TC-inception), which extracts local and global information via convolution blocks and transformer blocks, respectively, and integrates them by a cross-attention fusion with global and local feature (CAFGL) mechanism. Meanwhile, the improved skip connection structure named skip connection with cross-attention fusion (SCCAF) mechanism can alleviate the semantic differences between encoder features and decoder features for better feature fusion. In addition, we designed 2D-TransConver and 3D-TransConver for 2D and 3D brain tumor segmentation tasks, respectively, and verified the performance and advantage of our model through brain tumor datasets.

RESULTS

We trained our model on 335 cases from the training dataset of MICCAI BraTS2019 and evaluated the model's performance based on 66 cases from MICCAI BraTS2018 and 125 cases from MICCAI BraTS2019. Our TransConver achieved the best average Dice score of 83.72% and 86.32% on BraTS2019 and BraTS2018, respectively.

CONCLUSIONS

We proposed a transformer and convolution parallel network named TransConver for brain tumor segmentation. The TC-Inception module effectively extracts global information while retaining local details. The experimental results demonstrated that good segmentation requires the model to extract local fine-grained details and global semantic information simultaneously, and our TransConver effectively improves the accuracy of brain tumor segmentation.

On the shape of convolution kernels in MRI reconstruction: Rectangles versus ellipsoids

PURPOSE

Many MRI reconstruction methods (including GRAPPA, SPIRiT, ESPIRiT, LORAKS, and convolutional neural network [CNN] methods) involve shift-invariant convolution models. Rectangular convolution kernel shapes are often chosen by default, although ellipsoidal kernel shapes have potentially appealing theoretical characteristics. In this work, we systematically investigate the differences between different kernel shape choices in several contexts.

THEORY

It is well-understood that a rectangular region of k-space is associated with anisotropic spatial resolution, while ellipsoidal regions can be associated with more isotropic resolution. Further, for a fixed spatial resolution, ellipsoidal kernels are associated with substantially fewer parameters than rectangular kernels. These characteristics suggest that ellipsoidal kernels may have certain advantages over rectangular kernels.

METHODS

We used real retrospectively undersampled k-space data to empirically study the characteristics of rectangular and ellipsoidal kernels in the context of seven methods (GRAPPA, SPIRiT, ESPIRiT, SAKE, LORAKS, AC-LORAKS, and CNN-based reconstructions).

RESULTS

Empirical results suggest that both kernel shapes can produce reconstructed images with similar error metrics, although the ellipsoidal shape can often achieve this with reduced computation time and memory usage and/or fewer model parameters.

CONCLUSION

Ellipsoidal kernel shapes may offer advantages over rectangular kernel shapes in various MRI applications.

Two-stage ECG signal denoising based on deep convolutional network

Background.An electrocardiogram (ECG) is an effective and non-invasive indicator for the detection and prevention of arrhythmia. ECG signals are susceptible to noise contamination, which can lead to errors in ECG interpretation. Therefore, ECG pretreatment is important for accurate analysis.Methods.The ECG data used are from CPSC2018, and the noise signal is from MIT-BIH Noise Stress Test Database. In the experiment, the signal-to-noise ratio (SNR), the root mean square error (RMSE), and the correlation coefficientPare used to evaluate the performance of the network. The method proposed is divided into two stages. In the first stage, a U_de-net model is designed for ECG signal denoising to eliminate noise. The DR-net model in the second stage is used to reconstruct the ECG signal and to correct the waveform distortion caused by noise removal in the first stage. In this paper, the U_de-net and the DR-net are constructed by the convolution method to achieve end-to-end mapping from noisy ECG signals to clean ECG signals.Result.In SNR, RMSE andPindicators, U_de-net + DR-net proposed in this paper can achieve the best performance compared with the other five schemes (FCN, U-net etc). In the three data sets, SNR can be increased by 11.61 dB, 13.71 dB and 14.40 dB and RMSE can be reduced by 10.46 × 10^-2, 21.55 × 10^-2and 15.98 × 10^-2.Conclusions.Despite the contradictory results, the proposed two-stages method can achieve both the elimination of noise and the preservation of effective details to a large extent of the signals. The proposed method has good application prospects in clinical practice.

WEAK SINDY FOR PARTIAL DIFFERENTIAL EQUATIONS

Sparse Identification of Nonlinear Dynamics (SINDy) is a method of system discovery that has been shown to successfully recover governing dynamical systems from data [6, 39]. Recently, several groups have independently discovered that the weak formulation provides orders of magnitude better robustness to noise. Here we extend our Weak SINDy (WSINDy) framework introduced in [28] to the setting of partial differential equations (PDEs). The elimination of pointwise derivative approximations via the weak form enables effective machine-precision recovery of model coefficients from noise-free data (i.e. below the tolerance of the simulation scheme) as well as robust identification of PDEs in the large noise regime (with signal-to-noise ratio approaching one in many well-known cases). This is accomplished by discretizing a convolutional weak form of the PDE and exploiting separability of test functions for efficient model identification using the Fast Fourier Transform. The resulting WSINDy algorithm for PDEs has a worst-case computational complexity of O ( N D + 1 log ( N ) ) for datasets with N points in each of D + 1 dimensions. Furthermore, our Fourier-based implementation reveals a connection between robustness to noise and the spectra of test functions, which we utilize in an a priori selection algorithm for test functions. Finally, we introduce a learning algorithm for the threshold in sequential-thresholding least-squares (STLS) that enables model identification from large libraries, and we utilize scale invariance at the continuum level to identify PDEs from poorly-scaled datasets. We demonstrate WSINDy's robustness, speed and accuracy on several challenging PDEs. Code is publicly available on GitHub at https://github.com/MathBioCU/WSINDy_PDE.

Deep Convolutional Neural Network Regularization for Alcoholism Detection Using EEG Signals

Alcoholism is attributed to regular or excessive drinking of alcohol and leads to the disturbance of the neuronal system in the human brain. This results in certain malfunctioning of neurons that can be detected by an electroencephalogram (EEG) using several electrodes on a human skull at appropriate positions. It is of great interest to be able to classify an EEG activity as that of a normal person or an alcoholic person using data from the minimum possible electrodes (or channels). Due to the complex nature of EEG signals, accurate classification of alcoholism using only a small dataset is a challenging task. Artificial neural networks, specifically convolutional neural networks (CNNs), provide efficient and accurate results in various pattern-based classification problems. In this work, we apply CNN on raw EEG data and demonstrate how we achieved 98% average accuracy by optimizing a baseline CNN model and outperforming its results in a range of performance evaluation metrics on the University of California at Irvine Machine Learning (UCI-ML) EEG dataset. This article explains the stepwise improvement of the baseline model using the dropout, batch normalization, and kernel regularization techniques and provides a comparison of the two models that can be beneficial for aspiring practitioners who aim to develop similar classification models in CNN. A performance comparison is also provided with other approaches using the same dataset.

Model-based signal processing enables bidirectional inferring between local field potential and spikes evoked by noxious stimulation

BACKGROUND

Recording spontaneous and evoked activities by means of unitary extracellular recordings and local field potential (LFP) are key understanding the mechanisms of neural coding. The LFP is one of the most popular and easy methods to measure the activity of a population of neurons. LFP is also a composite signal known to be difficult to interpret and model. There is a growing need to highlight the relationship between spiking activity and LFP. Here, we hypothesized that LFP could be inferred from spikes under evoked noxious conditions.

METHOD

Recording was performed from the medullary dorsal horn (MDH) in deeply anesthetized rats. We detail a process to highlight the C-fiber (nociceptive) evoked activity, by removing the A-fiber evoked activity using a model-based approach. Then, we applied the convolution kernel theory and optimization algorithms to infer the C-fiber LFP from the single cell spikes. Finally, we used a probability density function and an optimization algorithm to infer the spikes distribution from the LFP.

RESULTS

We successfully extracted C-fiber LFP in all data recordings. We observed that C-fibers spikes preceded the C-fiber LFP and were rather correlated to the LFP derivative. Finally, we inferred LFP from spikes with excellent correlation coefficient (r = 0.9) and reverse generated the spikes distribution from LFP with good correlation coefficients (r = 0.7) on spikes number.

CONCLUSION

We introduced the kernel convolution theory to successfully infer the LFP from spikes, and we demonstrated that we could generate the spikes distribution from the LFP.

Deep Neural Network for Slip Detection on Ice Surface

Slip-induced falls are among the most common causes of major occupational injuries and economic loss in Canada. Identifying the risk factors associated with slip events is key to developing preventive solutions to reduce falls. One factor is the slip-resistance quality of footwear, which is fundamental to reducing the number of falls. Measuring footwear slip resistance with the recently developed Maximum Achievable Angle (MAA) test requires a trained researcher to identify slip events in a simulated winter environment. The human capacity for information processing is limited and human error is natural, especially in a cold environment. Therefore, to remove conflicts associated with human errors, in this paper a deep three-dimensional convolutional neural network is proposed to detect the slips in real-time. The model has been trained by a new dataset that includes data from 18 different participants with various clothing, footwear, walking directions, inclined angles, and surface types. The model was evaluated on three types of slips: Maxi-slip, midi-slip, and mini-slip. This classification is based on the slip perception and recovery of the participants. The model was evaluated based on both 5-fold and Leave-One-Subject-Out (LOSO) cross validation. The best accuracy of 97% was achieved when identifying the maxi-slips. The minimum accuracy of 77% was achieved when classifying the no-slip and mini-slip trials. The overall slip detection accuracy was 86% with sensitivity and specificity of 81% and 91%, respectively. The overall accuracy dropped by about 2% in LOSO cross validation. The proposed slip detection algorithm is not only beneficial for footwear manufactures to improve their footwear slip resistance quality, but it also has other potential applications, such as improving the slip resistance properties of flooring in healthcare facilities, commercial kitchens, and oil drilling platforms.

HFNet: A CNN Architecture Co-designed for Neuromorphic Hardware With a Crossbar Array of Synapses

The hardware-software co-optimization of neural network architectures is a field of research that emerged with the advent of commercial neuromorphic chips, such as the IBM TrueNorth and Intel Loihi. Development of simulation and automated mapping software tools in tandem with the design of neuromorphic hardware, whilst taking into consideration the hardware constraints, will play an increasingly significant role in deployment of system-level applications. This paper illustrates the importance and benefits of co-design of convolutional neural networks (CNN) that are to be mapped onto neuromorphic hardware with a crossbar array of synapses. Toward this end, we first study which convolution techniques are more hardware friendly and propose different mapping techniques for different convolutions. We show that, for a seven-layered CNN, our proposed mapping technique can reduce the number of cores used by 4.9-13.8 times for crossbar sizes ranging from 128 × 256 to 1,024 × 1,024, and this can be compared to the toeplitz method of mapping. We next develop an iterative co-design process for the systematic design of more hardware-friendly CNNs whilst considering hardware constraints, such as core sizes. A python wrapper, developed for the mapping process, is also useful for validating hardware design and studies on traffic volume and energy consumption. Finally, a new neural network dubbed HFNet is proposed using the above co-design process; it achieves a classification accuracy of 71.3% on the IMAGENET dataset (comparable to the VGG-16) but uses 11 times less cores for neuromorphic hardware with core size of 1,024 × 1,024. We also modified the HFNet to fit onto different core sizes and report on the corresponding classification accuracies. Various aspects of the paper are patent pending.

Perception Understanding Action: Adding Understanding to the Perception Action Cycle With Spiking Segmentation

Traditionally the Perception Action cycle is the first stage of building an autonomous robotic system and a practical way to implement a low latency reactive system within a low Size, Weight and Power (SWaP) package. However, within complex scenarios, this method can lack contextual understanding about the scene, such as object recognition-based tracking or system attention. Object detection, identification and tracking along with semantic segmentation and attention are all modern computer vision tasks in which Convolutional Neural Networks (CNN) have shown significant success, although such networks often have a large computational overhead and power requirements, which are not ideal in smaller robotics tasks. Furthermore, cloud computing and massively parallel processing like in Graphic Processing Units (GPUs) are outside the specification of many tasks due to their respective latency and SWaP constraints. In response to this, Spiking Convolutional Neural Networks (SCNNs) look to provide the feature extraction benefits of CNNs, while maintaining low latency and power overhead thanks to their asynchronous spiking event-based processing. A novel Neuromorphic Perception Understanding Action (PUA) system is presented, that aims to combine the feature extraction benefits of CNNs with low latency processing of SCNNs. The PUA utilizes a Neuromorphic Vision Sensor for Perception that facilitates asynchronous processing within a Spiking fully Convolutional Neural Network (SpikeCNN) to provide semantic segmentation and Understanding of the scene. The output is fed to a spiking control system providing Actions. With this approach, the aim is to bring features of deep learning into the lower levels of autonomous robotics, while maintaining a biologically plausible STDP rule throughout the learned encoding part of the network. The network will be shown to provide a more robust and predictable management of spiking activity with an improved thresholding response. The reported experiments show that this system can deliver robust results of over 96 and 81% for accuracy and Intersection over Union, ensuring such a system can be successfully used within object recognition, classification and tracking problem. This demonstrates that the attention of the system can be tracked accurately, while the asynchronous processing means the controller can give precise track updates with minimal latency.

NIMBLE for Bayesian Disease Mapping

This tutorial describes the basic implementation of Bayesian hierarchical models for spatial health data using the R package nimble. To quote the nimble R description: A system for writing hierarchical statistical models largely compatible with 'BUGS' and 'JAGS', writing nimbleFunctions to operate models and do basic R-style math, and compiling both models and nimbleFunctions via custom-generated C++. 'NIMBLE' includes default methods for MCMC, particle filtering, Monte Carlo Expectation Maximization, and some other tools. The nimbleFunction system makes it easy to do things like implement new MCMC samplers from R, customize the assignment of samplers to different parts of a model from R, and compile the new samplers automatically via C++ alongside the samplers 'NIMBLE' provides. Examples of the use of the package for a small range of Bayesian Disease Mapping (BDM) models is explored and focus on different approaches to model fitting and analysis are discussed. Examples of publicly available small area health data is used throughout.

PLGA Microspheres with Alginate-Coated Large Pores for the Formulation of an Injectable Depot of Donepezil Hydrochloride

As the main symptom of Alzheimer's disease-related dementia is memory loss, patient compliance for donepezil hydrochloride (donepezil), administered as once-daily oral formulations, is poor. Thus, we aimed to design poly(lactic-co-glycolic acid) (PLGA) microspheres (MS) with alginate-coated large pores as an injectable depot of donepezil exhibiting sustained release over 2-3 weeks. The PLGA MS with large pores could provide large space for loading drugs with high loading capacity, and thereby sufficient amounts of drugs were considered to be delivered with minimal use of PLGA MS being injected. However, initial burst release of donepezil from the porous PLGA MS was observed. To reduce this initial burst release, the surface pores were closed with calcium alginate coating using a spray-ionotropic gelation method. The final pore-closed PLGA MS showed in vitro sustained release for approximately 3 weeks, and the initial burst release was remarkably decreased by the calcium alginate coating. In the prediction of plasma drug concentration profiles using convolution method, the mean residence time of the pore-closed PLGA MS was 2.7-fold longer than that of the porous PLGA MS. Therefore, our results reveal that our pore-closed PLGA MS formulation is a promising candidate for the treatment of dementia with high patient compliance.

Universal Algorithm for Simulating and Evaluating Cyclic Voltammetry at Macroporous Electrodes by Considering Random Arrays of Microelectrodes

An algorithm for the simulation and evaluation of cyclic voltammetry (CV) at macroporous electrodes such as felts, foams, and layered structures is presented. By considering 1D, 2D, and 3D arrays of electrode sheets, cylindrical microelectrodes, hollow-cylindrical microelectrodes, and hollow-spherical microelectrodes the internal diffusion domains of the macroporous structures are approximated. A universal algorithm providing the time-dependent surface concentrations of the electrochemically active species, required for simulating cyclic voltammetry responses of the individual planar, cylindrical, and spherical microelectrodes, is presented as well. An essential ingredient of the algorithm, which is based on Laplace integral transformation techniques, is the use of a modified Talbot contour for the inverse Laplace transformation. It is demonstrated that first-order homogeneous chemical kinetics preceding and/or following the electrochemical reaction and electrochemically active species with non-equal diffusion coefficients can be included in all diffusion models as well. The proposed theory is supported by experimental data acquired for a reference reaction, the oxidation of [Fe(CN)6 ]4- at platinum electrodes as well as for a technically relevant reaction, the oxidation of VO2+ at carbon felt electrodes. Based on our calculation strategy, we provide a powerful open source tool for simulating and evaluating CV data implemented into a Python graphical user interface (GUI).

The hypoattenuating ocular lens on CT is not always due to cataract formation

Hypoattenuating ocular lenses on CT have been described with cataract formation in humans, however published studies are currently lacking regarding this finding in veterinary patients. The purpose of this retrospective and prospective study was to describe the varying CT appearances of the ocular lens in vivo, and investigate the causes for CT density variations in a population of cats and dogs. A total of 102 canine and feline patients with CT of the head acquired at the authors' hospital between May 2011 and March 2019 were included. A bilateral hypoattenuating halo surrounding an isoattenuating to mildly hypoattenuating core was described in the ocular lens center of every cat in which a Philips brand proprietary image construction filter was used. A similar but more varied hypoattenuating region was noted in the lenses of 45.8% of dogs where the same filter was applied, as well as 43.8% of dogs with a second, similar filter. Ophthalmic examination of three live cats and one dog with hypoattenuating lenses demonstrated normal lens translucency, excluding the presence of cataract. The effect of different proprietary filters on lens appearance was also described in three fresh cadavers with normal lenses identified on ophthalmic, macroscopic, and microscopic examination. Etiology of the hypoattenuating areas within the ocular lens was not conclusively determined. Recognition that such a variant may be seen in the absence of cataract is important, in order to prevent misdiagnosis.

Deconvolution Analysis by Non-linear Regression Using a Convolution-Based Model: Comparison of Nonparametric and Parametric Approaches

The convolution-based modeling approach has been shown to be flexible and easy to implement for performing a deconvolution analysis and for assessing in vitro/in vivo correlation using non-linear regression and a pre-specified model describing the in vivo drug absorption. A generalization of this method has been developed using a nonparametric description of the in vivo drug absorption process in replacement of a model-based definition. A comparison of the parametric and nonparametric deconvolution and convolution analyses was conducted on the pharmacokinetic (PK) data observed in four published studies after the administration of an extended-release formulation of methylphenidate at the dose of 18 mg. All the analyses were conducted using a conventional non-linear regression software (NONMEM). The results of the deconvolution analysis indicated that the parametric and nonparametric approaches performed similarly. The parametric approach described the input function using a double Weibull equation (6 parameters) while the nonparametric approach described the input function using a piecewise approximation (12-13 parameters). The validation of the results of the deconvolution analysis was conducted by comparing observed and predicted PK concentrations by the convolution analysis. The performance of the parametric and nonparametric approaches for assessing deconvolution was evaluated using the Akaike and the Bayesian information criteria. These criteria indicated that, despite the similar results obtained with the two approaches, the nonparametric approach provided better results. In conclusion, these results indicated that the nonparametric approach should be considered as the preferred approach for conducting a deconvolution analysis.

Parallel beamlet dose calculation via beamlet contexts in a distributed multi-GPU framework

PURPOSE

Dose calculation is one of the most computationally intensive, yet essential tasks in the treatment planning process. With the recent interest in automatic beam orientation and arc trajectory optimization techniques, there is a great need for more efficient model-based dose calculation algorithms that can accommodate hundreds to thousands of beam candidates at once. Foundational work has shown the translation of dose calculation algorithms to graphical processing units (GPUs), lending to remarkable gains in processing efficiency. But these methods provide parallelization of dose for only a single beamlet, serializing the calculation of multiple beamlets and under-utilizing the potential of modern GPUs. In this paper, the authors propose a framework enabling parallel computation of many beamlet doses using a novel beamlet context transformation and further embed this approach in a scalable network of multi-GPU computational nodes.

METHODS

The proposed context-based transformation separates beamlet-local density and TERMA into distinct beamlet contexts that independently provide sufficient data for beamlet dose calculation. Beamlet contexts are arranged in a composite context array with dosimetric isolation, and the context array is subjected to a GPU collapsed-cone convolution superposition procedure, producing the set of beamlet-specific dose distributions in a single pass. Dose from each context is converted to a sparse representation for efficient storage and retrieval during treatment plan optimization. The context radius is a new parameter permitting flexibility between the speed and fidelity of the dose calculation process. A distributed manager-worker architecture is constructed around the context-based GPU dose calculation approach supporting an arbitrary number of worker nodes and resident GPUs. Phantom experiments were executed to verify the accuracy of the context-based approach compared to Monte Carlo and a reference CPU-CCCS implementation for single beamlets and broad beams composed by addition of beamlets. Dose for representative 4π beam sets was calculated in lung and prostate cases to compare its efficiency with that of an existing beamlet-sequential GPU-CCCS implementation. Code profiling was also performed to evaluate the scalability of the framework across many networked GPUs.

RESULTS

The dosimetric accuracy of the context-based method displays <1.35% and 2.35% average error from the existing serialized CPU-CCCS algorithm and Monte Carlo simulation for beamlet-specific PDDs in water and slab phantoms, respectively. The context-based method demonstrates substantial speedup of up to two orders of magnitude over the beamlet-sequential GPU-CCCS method in the tested configurations. The context-based framework demonstrates near linear scaling in the number of distributed compute nodes and GPUs employed, indicating that it is flexible enough to meet the performance requirements of most users by simply increasing the hardware utilization.

CONCLUSIONS

The context-based approach demonstrates a new expectation of performance for beamlet-based dose calculation methods. This approach has been successful in accelerating the dose calculation process for very large-scale treatment planning problems - such as automatic 4π IMRT beam orientation and VMAT arc trajectory selection, with hundreds of thousands of beamlets - in clinically feasible timeframes. The flexibility of this framework makes it as a strong candidate for use in a variety of other very large-scale treatment planning tasks and clinical workflows.

Research of Method for Solving Relaxation Modulus Based on Three-Point Bending Creep Test

A method was developed for solving the relaxation modulus of high viscosity asphalt sand (HVAS) based on the three-point bending creep test, and was verified by comparison with experimental results. In this method, firstly, a transcendental equation was obtained by the convolution, and then equations were obtained by Taylor’s formula, which were solved by Mathmatica to obtain the relaxation modulus by Newton’s method. Subsequently, the laboratory investigations of the viscoelastic parameters of the Burgers model for the HVAS by three-point bending creep tests were carried out. In addition, the method was verified by comparing the relaxation moduli with the indoor relaxation experiments. Results showed that the numerical calculation and the test data were in good agreement, and the relaxation characteristics of the HVAS were reflected more accurately. The method can be used to study the relaxation characteristics of the asphalt mixtures effectively. In addition, this study provides a research basis for road crack prevention.

Mapping Tobacco Fields Using UAV RGB Images

Tobacco planting information is an important part of tobacco production management. Unmanned aerial vehicle (UAV) remote sensing systems have become a popular topic worldwide because they are mobile, rapid and economic. In this paper, an automatic identification method for tobacco fields based on UAV images is developed by combining supervised classifications with image morphological operations, and this method was used in the Yunnan Province, which is the top province for tobacco planting in China. The results show that the produce accuracy, user accuracy, and overall accuracy of tobacco field identification using the method proposed in this paper are 92.59%, 96.61% and 95.93%, respectively. The method proposed in this paper has the advantages of automation, flow process, high accuracy and easy operation, but the ground sampling distance (GSD) of the UAV image has an effect on the accuracy of the proposed method. When the image GSD was reduced to 1 m, the overall accuracy decreased by approximately 10%. To solve this problem, we further introduced the convolution method into the proposed method, which can ensure the recognition accuracy of tobacco field is above 90% when GSD is less than or equal to 1 m. Some other potential improvements of methods for mapping tobacco fields were also discussed in this paper.

Letter to the Editor

Hutson and Vexler (2018) demonstrate an example of aliasing with the beta and normal distribution. This letter presents another illustration of aliasing using the beta and normal distributions via an infinite mixture model, inspired by the problem of modeling placebo response.

FastSKAT: Sequence kernel association tests for very large sets of markers

The sequence kernel association test (SKAT) is widely used to test for associations between a phenotype and a set of genetic variants that are usually rare. Evaluating tail probabilities or quantiles of the null distribution for SKAT requires computing the eigenvalues of a matrix related to the genotype covariance between markers. Extracting the full set of eigenvalues of this matrix (an <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>n</mml:mi><mml:mo>×</mml:mo><mml:mi>n</mml:mi></mml:mrow></mml:math> matrix, for n subjects) has computational complexity proportional to n³ . As SKAT is often used when <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>n</mml:mi><mml:mo>></mml:mo><mml:msup><mml:mn>10</mml:mn><mml:mn>4</mml:mn></mml:msup></mml:mrow></mml:math> , this step becomes a major bottleneck in its use in practice. We therefore propose fastSKAT, a new computationally inexpensive but accurate approximations to the tail probabilities, in which the k largest eigenvalues of a weighted genotype covariance matrix or the largest singular values of a weighted genotype matrix are extracted, and a single term based on the Satterthwaite approximation is used for the remaining eigenvalues. While the method is not particularly sensitive to the choice of k, we also describe how to choose its value, and show how fastSKAT can automatically alert users to the rare cases where the choice may affect results. As well as providing faster implementation of SKAT, the new method also enables entirely new applications of SKAT that were not possible before; we give examples grouping variants by topologically associating domains, and comparing chromosome-wide association by class of histone marker.

Predicting events in clinical trials using two time-to-event outcomes

In clinical trials with time-to-event outcomes, it is of interest to predict when a prespecified number of events can be reached. Interim analysis is conducted to estimate the underlying survival function. When another correlated time-to-event endpoint is available, both outcome variables can be used to improve estimation efficiency. In this paper, we propose to use the convolution of two time-to-event variables to estimate the survival function of interest. Propositions and examples are provided based on exponential models that accommodate possible change points. We further propose a new estimation equation about the expected time that exploits the relationship of two endpoints. Simulations and the analysis of real data show that the proposed methods with bivariate information yield significant improvement in prediction over that of the univariate method.

An updated overview with simple and practical approach for developing in vitro-in vivo correlation

Preclinical Research & Development An in vitro-in vivo correlation (IVIVC) is as a predictive mathematical model that demonstrates a key role in the development, advancement, evaluation and optimization of extended release, modified release and immediate release pharmaceutical formulations. A validated IVIVC model can serve as a surrogate for bioequivalence studies and subsequently save time, effort and expenditure during pharmaceutical product development. This review discusses about different levels of correlations, general approaches to develop an IVIVC by mathematical modelling, validation, data analysis and various applications. In the current setting, the dearth of success associated with IVIVC is due to complexity of underlying scientific principles as well as the practice of fitting/matching in vivo plasma level-time data with in vitro dissolution profile. Hence, a simple, straightforward practical means to predict plasma drug levels by convolution technique and percentage drug absorbed computed from in vitro dissolution profile based on deconvolution method are illustrated. The bioavailability/bioequivalence assessment and evaluation are frequently validated by the pharmacokinetic parameters such as maximum concentration, time to reach maximum concentration, and area under the curve. The implementation of a quality by design manufacturing based on in vivo bioavailability and clinically relevant dissolution specification are recommended because corresponding design safe space will guarantee that all batches from relevant products are met with sufficient quality and bioperformance. Recently, United States Food and Drug Administration and European Medicines Agency have proposed that in silico/physiologically based pharmacokinetic modelling can be used in decision making during preclinical experiments as well as to recognize the dissolution profiles that can forecast and ensure the desired clinical performance.

Estimating lead-time bias in lung cancer diagnosis of patients with previous cancers

Surprisingly, survival from a diagnosis of lung cancer has been found to be longer for those who experienced a previous cancer than for those with no previous cancer. A possible explanation is lead-time bias, which, by advancing the time of diagnosis, apparently extends survival among those with a previous cancer even when they enjoy no real clinical advantage. We propose a discrete parametric model to jointly describe survival in a no-previous-cancer group (where, by definition, lead-time bias cannot exist) and in a previous-cancer group (where lead-time bias is possible). We model the lead time with a negative binomial distribution and the post-lead-time survival with a linear spline on the logit hazard scale, which allows for survival to differ between groups even in the absence of bias; we denote our model Logit-Spline/Negative Binomial. We fit Logit-Spline/Negative Binomial to a propensity-score matched subset of the Surveillance, Epidemiology, and End Results-Medicare linked data set, conducting sensitivity analyses to assess the effects of key assumptions. With lung cancer-specific death as the end point, the estimated mean lead time is roughly 11 months for stage I&II patients; with overall survival, it is roughly 3.4 months in stage I&II. For patients with higher-stage lung cancers, the mean lead time is 1 month or less for both outcomes. Accounting for lead-time bias reduces the survival advantage of the previous-cancer group when one exists, but it does not nullify it in all cases.

A comparison of the convolution and TMR10 treatment planning algorithms for Gamma Knife® radiosurgery

AIMS

To compare the accuracies of the convolution and TMR10 Gamma Knife treatment planning algorithms, and assess the impact upon clinical practice of implementing convolution-based treatment planning.

METHODS

Doses calculated by both algorithms were compared against ionisation chamber measurements in homogeneous and heterogeneous phantoms. Relative dose distributions calculated by both algorithms were compared against film-derived 2D isodose plots in a heterogeneous phantom, with distance-to-agreement (DTA) measured at the 80%, 50% and 20% isodose levels. A retrospective planning study compared 19 clinically acceptable metastasis convolution plans against TMR10 plans with matched shot times, allowing novel comparison of true dosimetric parameters rather than total beam-on-time. Gamma analysis and dose-difference analysis were performed on each pair of dose distributions.

RESULTS

Both algorithms matched point dose measurement within ±1.1% in homogeneous conditions. Convolution provided superior point-dose accuracy in the heterogeneous phantom (-1.1% v 4.0%), with no discernible differences in relative dose distribution accuracy. In our study convolution-calculated plans yielded D_99% 6.4% (95% CI:5.5%-7.3%,p<0.001) less than shot matched TMR10 plans. For gamma passing criteria 1%/1mm, 16% of targets had passing rates >95%. The range of dose differences in the targets was 0.2-4.6Gy.

CONCLUSIONS

Convolution provides superior accuracy versus TMR10 in heterogeneous conditions. Implementing convolution would result in increased target doses therefore its implementation may require a revaluation of prescription doses.