Regular Strength Research Articles

Prediction of sloshing pressure using image-based deep learning

This study investigates an image-based deep learning method for predicting various types of sloshing pressure, including regular, weak impulse, random, and peak pressures. The time series of pressure at a specific location on a tank wall are estimated from independent features of the sequential images of free-surface waves in sloshing flow. A dataset of wave images labeled with pressure is inserted into the residual neural network (ResNet) to enable the training of neural networks without encountering the problem of vanishing gradients. The hyperparameters of the ResNet algorithm are tuned to determine the optimal learning rate, mini-batch size, and strength of L1 regularization. Compared with experimental measurements, the prediction performance regarding the time history of pressure is remarkably good in the non-resonance regime, and the predictions exhibit reasonable agreement in the resonance regime. Notably, the method for pressure normalization has a significant impact on the prediction performance for peak pressure in the resonance regime, affecting both the magnitude and profile of the pressure. Furthermore, to enable practical applications of the proposed model, training is conducted using a comprehensive dataset of wave images including all pressure types. The model successfully estimates the time history for the different pressure types that occur sequentially.

Module-based regularization improves Gaussian graphical models when observing noisy data

AbstractInferring relations from correlational data allows researchers across the sciences to uncover complex connections between variables for insights into the underlying mechanisms. The researchers often represent inferred relations using Gaussian graphical models, requiring regularization to sparsify the models. Acknowledging that the modular structure of these inferred networks is often studied, we suggest module-based regularization to balance under- and overfitting. Compared with the graphical lasso, a standard approach using the Gaussian log-likelihood for estimating the regularization strength, this approach better recovers and infers modular structure in noisy synthetic and real data. The module-based regularization technique improves the usefulness of Gaussian graphical models in the many applications where they are employed.

Demonstrating Bioequivalence for Two Dose Strengths of Niraparib andAbiraterone AcetateDual-Action Tablets Versus Single Agents: Utility of Clinical Study Data Supplemented with Modeling and Simulation.

The combination of niraparib and abiraterone acetate (AA) plus prednisone is under investigation for the treatment of patients with metastatic castration-resistant prostate cancer (mCRPC) and metastatic castration-sensitive prostate cancer (mCSPC). Regular-strength (RS) and lower-strength (LS) dual-action tablets (DATs), comprising niraparib 100 mg/AA 500 mg and niraparib 50 mg/AA 500 mg, respectively, were developed to reduce pill burden and improve patient experience. A bioequivalence (BE)/bioavailability (BA) study was conducted under modified fasting conditions in patients with mCRPC to support approval of the DATs. This open-label randomized BA/BE study (NCT04577833) was conducted at 14 sites in the USA and Europe. The study had a sequential design, including a 21-day screening phase, a pharmacokinetic (PK) assessment phase comprising three periods [namely (1) single-dose with up to 1-week run-in, (2) daily doseon days 1-11, and (3) daily dose ondays 12-22], an extension where both niraparib and AA as single-agent combination (SAC; reference) or AA alone was continued from day 23 until discontinuation, and a 30-day follow-up phase. Patients were randomly assigned in a parallel-group design (four-sequence randomization) to receive a single oral dose of niraparib 100 mg/AA 1000 mg as a LS-DAT or SAC in period 1, and patients continued as randomized into a two-way crossover design during periods 2 and 3 where they received niraparib 200 mg/AA 1000 mg once daily as a RS-DAT or SAC. The design was powered on the basis of crossover assessment of RS-DAT versus SAC. During repeated dosing (periods 2 and 3, and extension phase), all patients also received prednisone/prednisolone 5 mg twice daily. Plasma samples were collected for measurement of niraparib and abiraterone plasma concentrations. Statistical assessment of the RS-DAT and LS-DAT versus SAC was performed on log-transformed pharmacokinetic parameters data from periods 2 and 3 (crossover) and from period 1 (parallel), respectively. Additional paired analyses and model-based bioequivalence assessments were conducted to evaluate the similarity between the LS-DAT and SAC. For the RS-DAT versus SAC, the 90% confidence intervals (CI) of geometric mean ratios (GMR) for maximum concentration at a steady state (Cmax,ss) and area under the plasma concentration-time curve from 0-24 h at a steady state (AUC 0-24h,ss) were respectively 99.18-106.12% and 97.91-104.31% for niraparib and 87.59-106.69 and 86.91-100.23% for abiraterone. For the LS-DAT vs SAC, the 90% CI of GMR for AUC0-72h of niraparib was 80.31-101.12% in primary analysis, the 90% CI of GMR for Cmax,ss and AUC 0-24h,ss of abiraterone was 85.41-118.34% and 86.51-121.64% respectively, and 96.4% of simulated LS-DAT versus SAC BE trials met the BE criteria for both niraparib and abiraterone. The RS-DAT met BE criteria (range 80%-125%) versus SAC based on 90% CI of GMR for Cmax,ss and AUC 0-24h,ss. The LS-DAT was considered BE to SAC on the basis of the niraparib component meeting the BE criteria in the primary analysis for AUC 0-72h; abiraterone meeting the BE criteria in additional paired analyses based on Cmax,ss and AUC 0-24h,ss; and the percentage of simulated LS-DAT versus SAC BE trials meeting the BE criteria for both. NCT04577833.

Analyzing and Improving Existing Neural Network-Based Approaches to Identify AI Generated Images

Images produced by AI software, particularly human faces, have the potential to spread misinformation, thus detection tools are essential to tell facts from fiction. However, current neural network-based tools misclassify images with extreme photography conditions, facial obstructions, or post-processing. This project aimed to address the limitations of existing detection models, especially regarding poorly-photographed or edited images. A MobileNet-v2-based neural network to identify synthetic images was analyzed for possible shortcomings. After obtaining its baseline accuracy, it was stress tested on validation data with brightness, contrast, hue, and saturation edited by set increments. The model performed poorly on processed images and faces with glasses or shadows. Possible limitations were insufficient data augmentation, the Global Average Pooling Layer, too few epochs trained, and inadequate regularization strength or dropout rate. The model was then modified to address these shortcomings by altering the brightness, contrast, hue, and saturation of training data, adding Gaussian noise to training images, implementing a Global Max Pooling Layer, augmenting the number of epochs trained, and changing the regularization strength and dropout rate. The modified model was evaluated on augmented and non-augmented images. Overall, the model improved when Gaussian noise was added with standard deviation 0.1, it was trained for 15 epochs, and there were no dropout layers. This was because the Gaussian noise layer approximated degraded image quality, and the model was too simple to learn image features when trained for fewer epochs or when neurons were disabled. This research could be expanded by testing multiple modifications at once.

Constrained Plug-and-Play Priors for Image Restoration.

The Plug-and-Play framework has demonstrated that a denoiser can implicitly serve as the image prior for model-based methods for solving various inverse problems such as image restoration tasks. This characteristic enables the integration of the flexibility of model-based methods with the effectiveness of learning-based denoisers. However, the regularization strength induced by denoisers in the traditional Plug-and-Play framework lacks a physical interpretation, necessitating demanding parameter tuning. This paper addresses this issue by introducing the Constrained Plug-and-Play (CPnP) method, which reformulates the traditional PnP as a constrained optimization problem. In this formulation, the regularization parameter directly corresponds to the amount of noise in the measurements. The solution to the constrained problem is obtained through the design of an efficient method based on the Alternating Direction Method of Multipliers (ADMM). Our experiments demonstrate that CPnP outperforms competing methods in terms of stability and robustness while also achieving competitive performance for image quality.

Disentangling Magnetic Fields in NGC 6946 with Wide-band Polarimetry

We present λ13 cm polarization observations of the nearby spiral galaxy NGC 6946 with the Westerbork Synthesis Radio Telescope (WSRT) to examine the nearside halo magnetic fields. Despite λ13 cm exhibiting similar two-dimensional morphology as observed at longer (λ18–22 cm) or shorter (λ3 and λ6 cm) wavelengths, more complete frequency coverage will be required to explain the gap in polarization in the southwest quadrant of the galaxy. We fit models of the turbulent and coherent line-of-sight magnetic fields to the fractional degree of linearly polarized emission at λ3, λ6, λ13, λ18, and λ22 cm from observations taken with the WSRT, Karl G. Jansky Very Large Array, and Effelsberg telescopes. The results favor a multilayer turbulent magneto-ionized medium consistent with current observations of edge-on galaxies. We constrain the physical properties of the synchrotron-emitting thin and thick disks (scale heights of 300 pc and 1.4 kpc, respectively) along with the thermal thick disk and halo (scale heights of 1 and 5 kpc, respectively). Our preferred model indicates a clumpy and highly turbulent medium within 1 kpc of the midplane, and a diffuse extraplanar layer with a substantially lower degree of Faraday depolarization. In the halo, we estimate a regular magnetic field strength of 0.4–2.2 μG and that turbulence and a total magnetic field strength of ∼6 μG result in a Faraday dispersion of σ RM = 4–48 rad m−2. This work is an example of how the advanced capabilities of modern radio telescopes are opening a new frontier for the study of cosmic magnetism.

Proximal MCMC for Bayesian Inference of Constrained and Regularized Estimation

This article advocates proximal Markov chain Monte Carlo (ProxMCMC) as a flexible and general Bayesian inference framework for constrained or regularized estimation. Originally introduced in the Bayesian imaging literature, ProxMCMC employs the Moreau-Yosida envelope for a smooth approximation of the total-variation regularization term, fixes variance and regularization strength parameters as constants, and uses the Langevin algorithm for the posterior sampling. We extend ProxMCMC to be fully Bayesian by providing data-adaptive estimation of all parameters including the regularization strength parameter. More powerful sampling algorithms such as Hamiltonian Monte Carlo are employed to scale ProxMCMC to high-dimensional problems. Analogous to the proximal algorithms in optimization, ProxMCMC offers a versatile and modularized procedure for conducting statistical inference on constrained and regularized problems. The power of ProxMCMC is illustrated on various statistical estimation and machine learning tasks, the inference of which is traditionally considered difficult from both frequentist and Bayesian perspectives.

High-Fidelity Synthetic Face Generation for Rosacea Skin Condition from Limited Data

Similarly to the majority of deep learning applications, diagnosing skin diseases using computer vision and deep learning often requires a large volume of data. However, obtaining sufficient data for particular types of facial skin conditions can be difficult, due to privacy concerns. As a result, conditions like rosacea are often understudied in computer-aided diagnosis. The limited availability of data for facial skin conditions has led to the investigation of alternative methods of computer-aided diagnosis. In recent years, generative adversarial networks (GANs), mainly variants of StyleGANs, have demonstrated promising results in generating synthetic facial images. In this study, for the first time, a small dataset of rosacea with 300 full-face images was utilized to further investigate the possibility of generating synthetic data. Our experimentation demonstrated that the strength of R1 regularization is crucial for generating high-fidelity rosacea images using a few hundred images. This was complemented by various experimental settings to ensure model convergence. We successfully generated 300 high-quality synthetic images, significantly contributing to the limited pool of rosacea images for computer-aided diagnosis. Additionally, our qualitative evaluations by 3 expert dermatologists and 23 non-specialists highlighted the realistic portrayal of rosacea features in the synthetic images. We also provide a critical analysis of the quantitative evaluations and discuss the limitations of solely relying on validation metrics in the field of computer-aided clinical image diagnosis.

Korean Adolescents before and after COVID-19: Changes in Physical Activity, Mental Health, and Hygiene Management.

Adolescence is when health behaviors, namely, physical activity (PA), mental health, and hygiene management, are established. As the young represent our future societal welfare, their health requires attention, particularly in the face of a pandemic. We explored changes in Korean adolescent PA, mental health, and hygiene management before and after the COVID-19 pandemic to identify recommendations for improving adolescent health. We collected 2019-2021 data from the Korea Youth Risk Behavior Surveys from the Korea Disease Control and Prevention Agency for 166,590 middle and high school students (2019: 57,069; 2020: 54,809; 2021: 54,712). We analyzed seven variables (regular PA, high-intensity PA, and strength exercise participation, stress recognition, sleep-time satisfaction, brushing practice, and handwashing practice) to examine changes in adolescent behavior during these years. The findings indicated statistically significant differences in all seven variables over the three years (P<0.05). Specifically, regular PA was higher in 2019 and 2021 than in 2020; high-intensity PAs were highest in 2019, followed by 2021 and 2020; strength exercises were the highest in 2020, followed by 2021 and 2019; stress recognition was higher in 2019 and 2021 than in 2020; sleep-time satisfaction was the highest in 2020, followed by 2021 and 2019; brushing practice was the highest in 2019, followed by 2020 and 2021; and handwashing practice was the highest in 2020, followed by 2021 and 2019. The results recommend the need to encourage structured moderate-to-vigorous adolescent PA and systematic health education to improve Korean adolescent health behavior after COVID-19.

Optimizer's dilemma: optimization strongly influences model selection in transcriptomic prediction.

Most models can be fit to data using various optimization approaches. While model choice is frequently reported in machine-learning-based research, optimizers are not often noted. We applied two different implementations of LASSO logistic regression implemented in Python's scikit-learn package, using two different optimization approaches (coordinate descent, implemented in the liblinear library, and stochastic gradient descent, or SGD), to predict mutation status and gene essentiality from gene expression across a variety of pan-cancer driver genes. For varying levels of regularization, we compared performance and model sparsity between optimizers. After model selection and tuning, we found that liblinear and SGD tended to perform comparably. liblinear models required more extensive tuning of regularization strength, performing best for high model sparsities (more nonzero coefficients), but did not require selection of a learning rate parameter. SGD models required tuning of the learning rate to perform well, but generally performed more robustly across different model sparsities as regularization strength decreased. Given these tradeoffs, we believe that the choice of optimizers should be clearly reported as a part of the model selection and validation process, to allow readers and reviewers to better understand the context in which results have been generated. The code used to carry out the analyses in this study is available at https://github.com/greenelab/pancancer-evaluation/tree/master/01_stratified_classification. Performance/regularization strength curves for all genes in the Vogelstein et al.(2013) dataset are available at https://doi.org/10.6084/m9.figshare.22728644.

Regularized Multi-Decoder Ensemble for an Error-Aware Scene Representation Network.

Feature grid Scene Representation Networks (SRNs) have been applied to scientific data as compact functional surrogates for analysis and visualization. As SRNs are black-box lossy data representations, assessing the prediction quality is critical for scientific visualization applications to ensure that scientists can trust the information being visualized. Currently, existing architectures do not support inference time reconstruction quality assessment, as coordinate-level errors cannot be evaluated in the absence of ground truth data. By employing the uncertain neural network architecture in feature grid SRNs, we obtain prediction variances during inference time to facilitate confidence-aware data reconstruction. Specifically, we propose a parameter-efficient multi-decoder SRN (MDSRN) architecture consisting of a shared feature grid with multiple lightweight multi-layer perceptron decoders. MDSRN can generate a set of plausible predictions for a given input coordinate to compute the mean as the prediction of the multi-decoder ensemble and the variance as a confidence score. The coordinate-level variance can be rendered along with the data to inform the reconstruction quality, or be integrated into uncertainty-aware volume visualization algorithms. To prevent the misalignment between the quantified variance and the prediction quality, we propose a novel variance regularization loss for ensemble learning that promotes the Regularized multi-decoder SRN (RMDSRN) to obtain a more reliable variance that correlates closely to the true model error. We comprehensively evaluate the quality of variance quantification and data reconstruction of Monte Carlo Dropout (MCD), Mean Field Variational Inference (MFVI), Deep Ensemble (DE), and Predicting Variance (PV) in comparison with our proposed MDSRN and RMDSRN applied to state-of-the-art feature grid SRNs across diverse scalar field datasets. We demonstrate that RMDSRN attains the most accurate data reconstruction and competitive variance-error correlation among uncertain SRNs under the same neural network parameter budgets. Furthermore, we present an adaptation of uncertainty-aware volume rendering and shed light on the potential of incorporating uncertain predictions in improving the quality of volume rendering for uncertain SRNs. Through ablation studies on the regularization strength and decoder count, we show that MDSRN and RMDSRN are expected to perform sufficiently well with a default configuration without requiring customized hyperparameter settings for different datasets.

Optimizing lung cancer classification through hyperparameter tuning.

Artificial intelligence is steadily permeating various sectors, including healthcare. This research specifically addresses lung cancer, the world's deadliest disease with the highest mortality rate. Two primary factors contribute to its onset: genetic predisposition and environmental factors, such as smoking and exposure to pollutants. Recognizing the need for more effective diagnosis techniques, our study embarked on devising a machine learning strategy tailored to boost precision in lung cancer detection. Our aim was to devise a diagnostic method that is both less invasive and cost-effective. To this end, we proposed four methods, benchmarking them against prevalent techniques using a universally recognized dataset from Kaggle. Among our methods, one emerged as particularly promising, outperforming the competition in accuracy, precision and sensitivity. This method utilized hyperparameter tuning, focusing on the Gamma and C parameters, which were set at a value of 10. These parameters influence kernel width and regularization strength, respectively. As a result, we achieved an accuracy of 99.16%, a precision of 98% and a sensitivity rate of 100%. In conclusion, our enhanced prediction mechanism has proven to surpass traditional and contemporary strategies in lung cancer detection.

Kernel-Based Independence Tests for Causal Structure Learning on Functional Data.

Measurements of systems taken along a continuous functional dimension, such as time or space, are ubiquitous in many fields, from the physical and biological sciences to economics and engineering. Such measurements can be viewed as realisations of an underlying smooth process sampled over the continuum. However, traditional methods for independence testing and causal learning are not directly applicable to such data, as they do not take into account the dependence along the functional dimension. By using specifically designed kernels, we introduce statistical tests for bivariate, joint, and conditional independence for functional variables. Our method not only extends the applicability to functional data of the Hilbert-Schmidt independence criterion (hsic) and its d-variate version (d-hsic), but also allows us to introduce a test for conditional independence by defining a novel statistic for the conditional permutation test (cpt) based on the Hilbert-Schmidt conditional independence criterion (hscic), with optimised regularisation strength estimated through an evaluation rejection rate. Our empirical results of the size and power of these tests on synthetic functional data show good performance, and we then exemplify their application to several constraint- and regression-based causal structure learning problems, including both synthetic examples and real socioeconomic data.

Full 3-D modulation transfer function estimation of tomosynthesis system using modified Richardson-Lucy deconvolution.

Tomosynthesis systems are three-dimensional (3-D) medical imaging devices that operate over limited acquisition angles using low radiation dosages. To measure the spatial resolution performance of a tomosynthesis system, the modulation transfer function (MTF) is widely used as a quantitative evaluationmetric. We previously introduced a method to estimate the full 3-D MTF of a cone-beam computed tomography system using two-dimensional (2-D) Richardson-Lucy (RL) deconvolution with Tikhonov-Miller regularization. However, this method can not be applied directly to estimate the 3-D MTF of a tomosynthesis system, since the unique artifacts (i.e., shadow artifacts, spreading tails, directional blurring, and high-level noise) of the system produce several errors that lower the estimation performance. Varying positions of the negative pixels due to shadow artifacts and spreading tails cause inconsistent deconvolution performances at each of the directional projections, and the severe noise in the reconstructed images cause noise amplification during estimation. This work proposes several modifications to the previous method to resolve the inconsistent performance and noise amplification errors to increase the full 3-D MTF estimationaccuracy. Three modifications were introduced to the 2-D RL deconvolution to prevent estimation errors and improve MTF estimation performance: non-negativity relaxation function, cost function to terminate the iterative process of RL deconvolution, and regularization strength for noise control. To validate the effectiveness of the proposed modifications, we reconstructed sphere phantoms from simulation and experimental tomosynthesis studies in the iso-center and offset-center positions as well as estimated the full 3-D MTFs using the previous and proposed methods. We compared the 3-D render images, central plane images, and center profiles of the estimated 3-D MTFs and calculated the full widths at half and tenth maximum for quantitativeevaluation. The previous method cannot estimate the full 3-D MTF of a tomosynthesis system; its inaccurate negative pixel relaxation produces circular-shaped errors, and the mean squared error based simple cost function for termination causes inconsistent estimation at each directional projection to diminish the clear edges of the low-frequency drop and missing sample regions. Noise amplification from lack of noise regularization is also seen in the previous method results. Compared to the previous method, the proposed method shows superior estimation performance at reducing errors in both the simulation and experimental studies regardless of object position. The proposed method preserves the low-frequency drop, missing sample regions from the limited acquisition angles, and missing cone region from the offset-center position; the estimated MTFs also show FWHM and FWTM values close to those of the ideal MTFs than with the previousmethod. This work presents a method to estimate the full 3-D MTF of a tomosynthesis system. The proposed modifications prevent circular-shaped errors and noise amplification due to the geometry for limited acquisition angles and high noise levels. Compared to our previous method, the proposed scheme show better performance for estimating the 3-D MTF of the tomosynthesissystem.

Sentiment Analysis of Arabic Course Reviews of a Saudi University Using Support Vector Machine

This study presents the development of a sentimental analysis system for high education students using Arabic text. There is a gap in the literature concerning understanding the perceptions and opinions of students in Saudi Arabia Universities regarding their education beyond COVID-19. The proposed SVM Sentimental Analysis for Arabic Students’ Course Reviews (SVM-SAA-SCR) algorithm is a general framework that involves collecting student reviews, preprocessing them, and using a machine learning model to classify them as positive, negative, or neutral. The suggested technique for preprocessing and classifying reviews includes steps such as collecting data, removing irrelevant information, tokenizing, removing stop words, stemming or lemmatization, and using pre-trained sentiment analysis models. The classifier is trained using the SVM algorithm and performance is evaluated using metrics such as accuracy, precision, and recall. Fine-tuning is done by adjusting parameters such as kernel type and regularization strength to optimize performance. A real dataset provided by the deanship of quality at Prince Sattam bin Abdulaziz University (PSAU) is used and contains students’ opinions on various aspects of their education. We also compared our algorithm with CAMeLBERT, a state-of-the-art Dialectal Arabic model. Our findings show that while the CAMeLBERT model classified 70.48% of the reviews as positive, our algorithm classified 69.62% as positive which proves the efficiency of the suggested SVM-SAA-SCR. The results of the proposed model provide valuable insights into the challenges and obstacles faced by Arab Universities post-COVID-19 and can help to improve their educational experience.

Optimizing the regularization in size-consistent second-order Brillouin-Wigner perturbation theory.

Despite its simplicity and relatively low computational cost, second-order Møller-Plesset perturbation theory (MP2) is well-known to overbind noncovalent interactions between polarizable monomers and some organometallic bonds. In such situations, the pairwise-additive correlation energy expression in MP2 is inadequate. Although energy-gap dependent amplitude regularization can substantially improve the accuracy of conventional MP2 in these regimes, the same regularization parameter worsens the accuracy for small molecule thermochemistry and density-dependent properties. Recently, we proposed a repartitioning of Brillouin-Wigner perturbation theory that is size-consistent to second order (BW-s2), and a free parameter (α) was set to recover the exact dissociation limit of H2 in a minimal basis set. Alternatively α can be viewed as a regularization parameter, where each value of α represents a valid variant of BW-s2, which we denote as BW-s2(α). In this work, we semi-empirically optimize α for noncovalent interactions, thermochemistry, alkane conformational energies, electronic response properties, and transition metal datasets, leading to improvements in accuracy relative to the abinitio parameterization of BW-s2 and MP2. We demonstrate that the optimal α parameter (α = 4) is more transferable across chemical problems than energy-gap-dependent regularization parameters. This is attributable to the fact that the BW-s2(α) regularization strength depends on all of the information encoded in the t amplitudes rather than just orbital energy differences. While the computational scaling of BW-s2(α) is iterative O(N5), this effective and transferable approach to amplitude regularization is a promising route to incorporate higher-order correlation effects at second-order cost.

Investigation of Image Enhancement Techniques for Advancing Colon Cancer Diagnosis

Colorectal cancer continues to pose a substantial worldwide health challenge, necessitating the development of advanced imaging techniques for early and accurate diagnosis. In this study, we propose a novel hybrid image enhancement approach that combines Total Variation (TV) regularization and shift-invariant filtering to improve the visibility and diagnostic quality of colon cancer images.&#x0D; The Total Variation regularization technique is employed to effectively reduce noise and enhance the edges in the input images, thereby preserving important structural details. Simultaneously, shift-invariant filtering is utilized to address spatial variations and artifacts that often arise in medical images, ensuring consistent and reliable enhancements across the entire image. Our hybrid approach synergistically integrates the strengths of both TV regularization and shift-invariant filtering, resulting in enhanced colon cancer images that offer improved contrast, reduced noise, and enhanced fine structures. This improved image quality aids medical professionals in better identifying and characterizing cancerous lesions, ultimately leading to more accurate and timely diagnoses.&#x0D; To evaluate the effectiveness of the proposed approach, we conducted extensive experimentations on a diverse dataset of colon cancer images. Quantitative and qualitative assessments demonstrate that our hybrid approach outperforms existing enhancement methods, leading to superior image quality and diagnostic accuracy. In conclusion, the hybrid image enhancement approach presented in this study offers a promising solution for enhancing colon cancer images, contributing to the early detection and effective management of this life-threatening disease. These advancements hold significant potential for improving patient outcomes and reducing the burden of colon cancer on healthcare systems worldwide.

A framework for time-dependent ice sheet uncertainty quantification, applied to three West Antarctic ice streams

Abstract. Ice sheet models are the main tool to generate forecasts of ice sheet mass loss, a significant contributor to sea level rise; thus, knowing the likelihood of such projections is of critical societal importance. However, to capture the complete range of possible projections of mass loss, ice sheet models need efficient methods to quantify the forecast uncertainty. Uncertainties originate from the model structure, from the climate and ocean forcing used to run the model, and from model calibration. Here we quantify the latter, applying an error propagation framework to a realistic setting in West Antarctica. As in many other ice sheet modelling studies we use a control method to calibrate grid-scale flow parameters (parameters describing the basal drag and ice stiffness) with remotely sensed observations. Yet our framework augments the control method with a Hessian-based Bayesian approach that estimates the posterior covariance of the inverted parameters. This enables us to quantify the impact of the calibration uncertainty on forecasts of sea level rise contribution or volume above flotation (VAF) due to the choice of different regularization strengths (prior strengths), sliding laws, and velocity inputs. We find that by choosing different satellite ice velocity products our model leads to different estimates of VAF after 40 years. We use this difference in model output to quantify the variance that projections of VAF are expected to have after 40 years and identify prior strengths that can reproduce that variability. We demonstrate that if we use prior strengths suggested by L-curve analysis, as is typically done in ice sheet calibration studies, our uncertainty quantification is not able to reproduce that same variability. The regularization suggested by the L curves is too strong, and thus propagating the observational error through to VAF uncertainties under this choice of prior leads to errors that are smaller than those suggested by our two-member “sample” of observed velocity fields.

The reciprocal elastic net

A new penalized likelihood method (reciprocal elastic net) is put forward for regularization and variable selection. Our proposal is based on a new class of reciprocal penalty functions, combining the strengths of the reciprocal LASSO regularization and ridge regression. We formulate the reciprocal elastic net problem as an equivalent reciprocal LASSO problem on augmented data, facilitating a direct utilization of the reciprocal LASSO algorithm to generate the entire reciprocal elastic net solution path. We further present the reciprocal adaptive elastic net, fusing the merits of ridge regression with the adaptively weighted reciprocal LASSO regularization. These methods, illustrated through simulated examples and real data analysis, demonstrate satisfactory performance in various diversified scenarios compared to published methods. Finally, we propose Bayesian methods to solve the reciprocal elastic net and reciprocal adaptive elastic net models using Gibbs samplers.

Effect of isokinetic muscle strength training on knee muscle strength, proprioception, and balance ability in athletes with anterior cruciate ligament reconstruction: a randomised control trial.

Objective: This study aimed to investigate the effects of regular isokinetic muscle strength training on knee muscle strength, proprioception, and balance ability in athletes after anterior cruciate ligament (ACL) reconstruction. Methods: Forty-one athletes who underwent ACL reconstruction were randomly divided into the experimental (n = 21) and control (n = 20) groups. The experimental group used an isokinetic muscle strength tester for 4weeks (five times/ week) of knee flexion and extension isokinetic muscle strength training. The control group used the knee joint trainer (pneumatic resistance) for the same exercise regimen as the experimental group. Results: 1) Four weeks when compared with the baseline. Experimental group: the knee flexion and extension PT (60°/s and 240°/s) increased by 31.7%, 40.3%, 23.4%, and 42.9% (p < 0.01), and the flexion muscular endurance increased by 21.4% and 19.7% (p < 0.01). The flexion and extension kinaesthesia and the 30° and 60° position sense decreased by 36.2%, 32.3%, 40.0%, and 18.9% (p < 0.05). The anterior-posterior and medial-lateral displacement and speed decreased by 30.2%, 44.2%, 38.4%, and 24.0% (p < 0.05). Control group: the knee peak torque (60°/s) increased by 18.8% (p < 0.01). The anterior-posterior and medial-lateral displacement and speed decreased by 14.9%, 40.0%, 26.8%, and 19.5% (p < 0.01). 2) After 4weeks, compared with the control group, the knee flexion and extension peak torque (60°/s), extension, peak torque (240°/s), and extension muscular endurance of the treatment group increased to varying degrees (p < 0.05). However, the kinaesthesia, 30° position sense, and anterior-posterior displacement decreased to varying degrees (p < 0.05). Conclusion: Adding regular isokinetic muscle strength training to rehabilitation training further improved the knee flexion and extensor strength and extensor endurance of athletes with ACL reconstruction, as well as enhanced the kinaesthesia and 30° position sense and the balance between the anterior and posterior directions. However, the treatment had limited effects on knee flexion kinaesthesia and muscle endurance.