Image Noise Research Articles

An automated AI-powered IoT algorithm with data processing and noise elimination for plant monitoring and actuating.

This article aims to develop a novel Artificial Intelligence-powered Internet of Things (AI-powered IoT) system that can automatically monitor the conditions of the plant (crop) and apply the necessary action without human interaction. The system can remotely send a report on the plant conditions to the farmers through IoT, enabling them for tracking the healthiness of plants. Chili plant has been selected to test the proposed AI-powered IoT monitoring and actuating system as it is so sensitive to the soil moisture, weather changes and can be attacked by several types of diseases. The structure of the proposed system is passed through five main stages, namely, AI-powered IoT system design, prototype fabrication, signal and image processing, noise elimination and proposed system testing. The prototype for monitoring is equipped with multiple sensors, namely, soil moisture, carbon dioxide (CO2) detector, temperature, and camera sensors, which are utilized to continuously monitor the conditions of the plant. Several signal and image processing operations have been applied on the acquired sensors data to prepare them for further post-processing stage. In the post processing step, a new AI based noise elimination algorithm has been introduced to eliminate the noise in the images and take the right actions which are performed using actuators such as pumps, fans to make the necessary actions. The experimental results show that the prototype is functioning well with the proposed AI-powered IoT algorithm, where the water pump, exhausted fan and pesticide pump are actuated when the sensors detect a low moisture level, high CO2 concentration level, and video processing-based pests' detection, respectively. The results also show that the algorithm is capable to detect the pests on the leaves with 75% successful rate.

Basketball player motion detection and motion mode analysis based on biomechanical sensors

Basketball player motion detection and analysis are crucial for optimizing performance and preventing injuries. Traditional methods often rely on visual observation and video analysis, lacking precision and real-time feedback. In this study, a unique novel Intelligent Bayesian tuned-augmented Support Vector Machine (IB-ASVM) was proposed for predicting basketball players’ motion modes and performance analysis using the biomechanical sensor data. Advancements in biomechanical sensors such as accelerometers, gyroscopes, and force sensors are deployed into ESP32 to build a player’s wearable gadget. This gadget provides dynamic players with real-time sensing data. Data are transmitted to the cloud via Wi-Fi 7.0 for motion analysis and this model is stimulated using Arduino IDE. The Kalman Filter reduces noise and smoothens sensor data such as acceleration, and angular velocity. Then, the filtered data is employed in the Discrete Wavelet Transform (DWT) to capture time-frequency characteristics of motion signals, making it ideal for extracting relevant features. The featured data are utilized in the ASVM model to classify and detect the motion modes of the basketball players via IB optimization. The Tensor Flow software is used to implement the IB-ASVM model. The result demonstrates that IB-ASVM most accurately predicts the jump shot, layup, dribbling, running, pivoting, passing, free throw, and motion states of the basketball players. The IB-ASVM model accurately classifies basketball motion states using biomechanical sensor data, enhancing performance optimization and injury prevention through precise motion detection.

Improving Image Quality and Visualization of Hepatocellular Carcinoma in Arterial Phase Imaging Using Contrast Enhancement-Boost Technique.

This study aimed to evaluate the image quality and visualization of hepatocellular carcinoma (HCC) on arterial phase computed tomography (CT) using the contrast enhancement (CE)-boost technique. This retrospective study included 527 consecutive patients who underwent dynamic liver CT between June 2021 and February 2022. Quantitative and qualitative image analyses were performed on 486 patients after excluding 41 patients. HCC conspicuity was evaluated in 40 of the 486 patients with at least one HCC in the liver. Iodinated images obtained by subtracting nonenhanced images from arterial phase images were combined to generate CE-boost images. For quantitative image analysis, image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured for the liver, pancreas, muscles, and aorta. For qualitative analysis, the overall image quality and noise were graded using a 3-point scale. Artifact, sharpness, and HCC lesion conspicuity were assessed using a 5-point scale. The paired-sample t test was used to compare quantitative measures, whereas the Wilcoxon signed-rank test was used to compare qualitative measures. The mean SNR and CNR of the aorta, liver, pancreas, and muscle were significantly higher, and the image noise was significantly lower in the CE-boost images than in the conventional images (P < 0.001). The mean CNR of HCC was also significantly higher in the CE-boost images than in the conventional images (P < 0.001). In the qualitative analysis, CE-boost images showed higher scores for HCC lesion conspicuity than conventional images (P < 0.001). The overall image quality and visibility of HCC were improved using the CE-boost technique.

Minimizing prostate diffusion weighted MRI examination time through deep learning reconstruction

PurposeTo study the diagnostic image quality of high b-value diffusion weighted images (DWI) derived from standard and variably reduced datasets reconstructed with a commercially available deep learning reconstruction (DLR) algorithm. Materials and methodsThis was a retrospective study of 52 patients undergoing conventional prostate MRI with raw image data reconstructed using both conventional 2D Cartesian and DLR algorithms. Simulated shortened DWI acquisition times were performed by reconstructing images using DLR datasets harboring a reduced number of excitations (NEX). Two radiologists independently evaluated the image quality using a 4-point Likert scale. Signal-to-noise ratio (SNR) analysis was performed for the entire cohort and a subset of patients identified as having clinically significant prostate cancer identified at MRI, and later confirmed by pathology. ResultsRadiologists perceived less image noise for both restricted and large field of view (FOV) standard NEX dataset DLR diffusion images compared to conventionally reconstructed images with good interreader agreement. Diagnostic image quality was maintained for all DLR images using variably reduced NEX compared to conventionally reconstructed images employing the standard NEX. Improved signal to noise was observed for the restricted FOV DLR images compared to conventional reconstruction using standard NEX. DLR diffusion images derived from reduced NEX datasets translated to time reductions of up to 68 % and 39 % for the restricted and large FOV series acquisitions, respectively. ConclusionDLR of diffusion weighted images can reduce image noise at standard NEX and potentially reduce prostate MRI exam time when utilizing reduced NEX datasets without sacrificing diagnostic image quality.

An automated pheochromocytoma and paraganglioma lesion segmentation AI-model at whole-body 68Ga- DOTATATE PET/CT.

Somatostatin receptor (SSR) targeting radiotracer 68Ga-DOTATATE is used for Positron Emission Tomography (PET)/Computed Tomography (CT) imaging to assess patients with Pheochromocytoma and paraganglioma (PPGL), rare types of Neuroendocrine tumor (NET) which can metastasize thereby becoming difficult to quantify. The goal of this study is to develop an artificial intelligence (AI) model for automated lesion segmentation on whole-body 3D DOTATATE-PET/CT and to automate the tumor burden calculation. 132 68Ga-DOTATATE PET/CT scans from 38 patients with metastatic and inoperable PPGL, were split into 70, and 62 scans, from 20, and 18 patients for training, and test sets, respectively. The training set was further divided into patient-stratified 5 folds for cross-validation. 3D-full resolution nnUNet configuration was trained with 5-fold cross-validation. The model's detection performance was evaluated at both scan and lesion levels for the PPGL test set and two other clinical cohorts with NET (n = 9) and olfactory neuroblastoma (ONB, n = 5). Additionally, quantitative statistical analysis of PET parameters including SUVmax, total lesion uptake (TLU), and total tumor volume (TTV), was conducted. The nnUNet AI model achieved an average 5-fold validation dice similarity coefficient of 0.84 at the scan level. The model achieved dice similarity coefficients (DSC) of 0.88, 0.6, and 0.67 at the scan level, the sensitivity of 86%, 61.13%, and 61.64%, and a positive predictive value of 89%, 74%, and 86.54% at the lesion level for the PPGL test, NET and ONB cohorts, respectively. For PPGL cohorts, smaller lesions with low uptake were missed by the AI model (p < 0.001). Anatomical region-based failure analysis showed most of the false negative and false positive lesions within the liver for all the cohorts, mainly due to the high physiologic liver background activity and image noise on 68Ga- DOTATATE PET scans. The developed deep learning-based AI model showed reliable performance for automated segmentation of metastatic PPGL lesions on whole-body 68Ga-DOTATATE-PET/CT images, which may be beneficial for tumor burden estimation for objective evaluation during therapy follow-up. https://www. gov/study/NCT03206060 , https://www. gov/study/NCT04086485 , https://www. gov/study/NCT05012098 .

CoaddNet: Enhancing signal-to-noise ratio in single-shot images using convolutional neural networks with coadded image effect

Noise in astronomical images significantly impacts observations and analyses. Traditional denoising methods, such as increasing exposure time and image stacking, are limited when dealing with single-shot images or studying rapidly changing astronomical objects. To address this, we developed a novel deep-learning denoising model, CoaddNet, designed to improve the image quality of single-shot images and enhance the detection of faint sources. To train and validate the model, we constructed a dataset containing high and low signal-to-noise ratio (SNR) images, comprising coadded and single-shot types. CoaddNet combines the efficiency of convolutional operations with the advantages of the Transformer architecture, enhancing spatial feature extraction through a multi-branch structure and reparameterization techniques. Performance evaluation shows that CoaddNet surpasses the baseline model, NAFNet, by increasing the Peak Signal-to-Noise Ratio (PSNR) by 0.03 dB and the Structural Similarity Index (SSIM) by 0.005 while also improving throughput by 35.18%. The model significantly improves the SNR of single-shot images, with an average increase of 22.8, surpassing the noise reduction achieved by stacking 70-90 images. By boosting the SNR, CoaddNet significantly enhances the detection of faint sources, enabling SExtractor to detect an additional 22.88% of faint sources. Meanwhile, CoaddNet reduced the Mean Absolute Percentage Error (MAPE) of flux measurements for detected sources by at least 27.74%.

Processing of Idiopathic Pulmonary Fibrosis images based on spatial interpolation using DFT, FFT, DCT and LPF

The work focuses on the study of existing issues with some medical images obtained from patients with Idiopathic Pulmonary Fibrosis (IPF) who have survived this COVID-19 pandemic. This study analyzes potential causes of incorrect medical diagnoses with this disease. In this regard, we employ numerical algorithms such as DFT (Discrete Fourier Transform), FFT (Fast Fourier Transform), DCT (Discrete Cosine Transform), and LPF (Low-pass filter). The main objective of this work is to demonstrate that with these numerical algorithms based on Continuous and Discrete Fourier Theory, it is possible to filter existing noise in IPF radiography images. Furthermore, it is possible to compress, enhance, and improve their resolution so that better decisions can be made in a medical protocol. In this way, it contributes to the understanding of the true images that would be used for an optimal medical diagnosis.

Analysis of sagittal plane cine magnetic resonance imaging for measurement of pancreatic tumor residual motion during breath hold and evaluation of gating margins used in radiotherapy treatment.

In pancreatic radiotherapy, residual tumor motion during treatment increases the risk of toxicity. Cine imaging acquired during magnetic resonance guided radiotherapy (MRgRT) enables real-time treatment gating in response to anatomical motion, which can reduce this risk; however, treatment gating can negatively impact the efficiency of treatment. This study aimed to quantify the extent of residual tumor motion during breath hold and evaluate the appropriateness of the treatment gating margins used in current clinical practice. Cine imaging acquired during pancreatic MRgRT of 11 patients on the ViewRay MRIdian was analyzed. The total duration of treatment analyzed was 12 h 13 min. Improved methods for processing and analyzing cine imaging were developed: breath holds were systematically separated with frequency analysis, residual motion was measured with consideration of both the tracking structure contour and centroid, and residual motion measurements were supported by phantom measurements of image scaling, resolution, and noise. Residual motion was measured at angles 0°, 45°, 90°, and 135° to the superior-inferior (SI) direction. Total residual motion was measured by combining directional measurements. The minimum tracking structure displacement resolvable through cine imaging was found to be 1.5mm; therefore, residual motion analysis was limited to 1.5mm spatial resolution. Total residual motion was contained within margins ±1.5, ±3, and ±4.5mm with mean percentage frequencies of 97.0%, 91.1%, and 67.8%. Most residual motion was observed in the SI direction, and significantly more residual motion was measured for the tracking structure contour than the centroid. The results demonstrate that patients are largely able to maintain breath hold positions to within a 3mm margin, thus provide evidence that supports the use of a 3mm gating margin in clinical practice. Residual motion frequently exceeded 1.5mm so a reduction in gating margin would have an undesirable impact on treatment efficiency.

Figure-ground segmentation based medical image denoising using deep convolutional neural networks

Medical image noise can have a substantial impact on both diagnosis accuracy and image quality. Noise can cause problems with tasks related to diagnosis, like defining object boundaries, which are essential for precise diagnosis. By applying denoising techniques, medical imaging professionals can significantly improve image quality, reduce errors, and enhance the accuracy of diagnoses and treatments. This article presents a method for reducing noises like Gaussian noise, salt and pepper noise, speckle noise, and ring artefacts in medical images, such as magnetic resonance imaging (MRI), computed tomography (CT), and chest X-ray images. This study explores the integration of adaptive CNNs with guided image filtering for enhanced image quality and deep learning-driven Figure-Ground segmentation. The proposed method's performance is extensively evaluated on chest X-ray and MRI/CT images under diverse noise scenarios, with comparisons to existing techniques using established statistical metrics. The results validate that the proposed approach attains superior performance, yielding the highest values across these metrics.

Cardiac functional assessment using prospectively electrocardiography-triggered computed tomography in children with congenital heart disease: Comparison of radiation dose and image quality between heart rate-dependent single-extended and heart rate-independent dual-focused scans

PurposeTo evaluate radiation dose (RD) reduction potential of heart rate-independent dual-focused scan of prospectively electrocardiography (ECG)-triggered computed tomography (CT) for cardiac functional assessment in children with congenital heart disease (CHD), RD and image quality of the scan mode were compared to those of heart rate-dependent single-extended scan. MethodsThis study encompassed 1,252 prospectively ECG-triggered pediatric cardiothoracic CT examinations, including single-focused (a reference in matched comparisons), single-extended (younger patients), and dual-focused (older patients) scans. Propensity score matching was used to reduce the confounding effect of age and sex in two matched groups (MPs) (younger MP: single-focused vs. single-extended; older MP: single-focused vs. dual-focused). CT RD, image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) of the MPs were compared. ResultsThe effective dose of single-extended (1.4 ± 0.5 mSv) and dual-focused (1.1 ± 0.4 mSv) scans were approximately 2.0–3.2 times higher than (depending on heart rate) and approximately 1.8 times (irrespective of heart rate) that of the age- and sex-matched single-focused scans (0.6 ± 0.2 mSv), respectively. Image noise and SNR of single-extended and dual-focused scans were similar to those of the age- and sex-matched single-focused scans (p values > 0.05). The CNR was also comparable between single-focused and single-extended scans (younger MP) (p > 0.05), but a slightly lower CNR of the dual-focused scans compared to single-focused scans was observed in the older MP (p < 0.02). ConclusionFor cardiac functional assessment in children with CHD, heart rate-independent dual-focused prospectively ECG-triggered scan can reduce CT RD, especially at lower heart rates, with comparable image quality, compared to heart rate-dependent single-extended scan.

Precision without GPS: Multi-Sensor Fusion for Autonomous Drone Navigation in Complex Environments

The rapid evolution of drone technology has expanded its applications across variodus domains, including delivery services, environmental monitoring, and search and rescue operations. However, many of these applications face significant challenges in GPS-denied environments, such as dense urban areas and heavily forested regions, where traditional navigation methods falter. This paper presents a novel multi-sensor fusion algorithm designed to enhance the localization accuracy of autonomous drones without reliance on GPS. By integrating data from an Inertial Measurement Unit (IMU), LiDAR, and visual sensors, the proposed approach effectively compensates for the limitations of individual sensors, enabling robust navigation in complex environments. Experimental results demonstrate that the algorithm achieves an average localization accuracy of 1.2 meters in urban areas and 1.5 meters in forested settings, showcasing its resilience against sensor noise and environmental challenges. The implementation of loop closure techniques further improves long-term navigation accuracy, making it suitable for prolonged missions. This research contributes to the growing body of knowledge in autonomous drone navigation and offers significant implications for enhancing the operational capabilities of drones in real-world scenarios. Future work will focus on integrating additional sensors, exploring machine learning techniques for adaptive fusion, and conducting extensive field trials to validate the system's performance in dynamic environments..

Feasibility of Ultra-low Radiation and Contrast Medium Dosage in Aortic CTA Using Deep Learning Reconstruction at 60 kVp: An Image Quality Assessment

ObjectiveTo assess the viability of using ultra-low radiation and contrast medium (CM) dosage in aortic computed tomography angiography (CTA) through the application of low tube voltage (60kVp) and a novel deep learning image reconstruction algorithm (ClearInfinity, DLIR-CI). MethodsIodine attenuation curves obtained from a phantom study informed the administration of CM protocols. Non-obese participants undergoing aortic CTA were prospectively allocated into two groups and then obtained three reconstruction group. The conventional group (100kVp-CVgroup) underwent imaging at 100kVp and received 210 mg iodine/kg in combination with a hybrid iterative reconstruction algorithm (ClearView, HIR-CV). The experimental group was imaged at 60kVp with 105 mg iodine/kg, while images were reconstructed with HIR-CV (60 kVp-CV group) and with DLIR-CI (60 kVp-CI group). Student's t-test was used to compare differences in CM protocol and radiation dose. One-way ANOVA compared CT attenuation, image noise, SNR, and CNR among the three reconstruction groups, while the Kruskal-Wallis H test assessed subjective image quality scores. Post hoc analysis was performed with Bonferroni correction for multiple comparisons, and consistency analysis conducted in subjective image quality assessment was measured using Cohen's kappa. ResultsThe radiation dose (1.12±0.23 mSv vs. 2.03±0.82 mSv) and CM dosage (19.04±3.03 mL vs. 38.11±6.47 mL) provided the reduction of 45% and 50% in the experimental group compared to the conventional group. The CT attenuation, SNR and CNR of 60kVp-CI were superior to or equal to those of 100kVp-CV. Compared to the 60 kVp-CV group, images in 60kVp-CI showed higher SNR and CNR (all P < 0.001). There was no difference between the 60kVp-CI and 100kVp-CVgroup in terms of the subjective image quality of the aorta in various locations (all P > 0.05), with 60kVp-CI images were deemed diagnostically sufficient across all vascular segments. ConclusionsFor non-obese patients, the combined use of 60kVp and DLIR-CI algorithm can preserving image quality while enabling radiation dose and contrast medium savings for aortic CTA compared with 100kVp using HIR-CV.

Evaluation of Image Segmentation Methods for In Situ Quality Assessment in Additive Manufacturing

Additive manufacturing (AM), or 3D printing, has revolutionized the fabrication of complex parts, but assessing their quality remains a challenge. Quality assessment, especially for the interior part geometry, relies on post-print inspection techniques unsuitable for real-time in situ analysis. Vision-based approaches could be employed to capture images of any layer during fabrication, and then segmentation methods could be used to identify in-layer features in order to establish dimensional conformity and detect defects for in situ evaluation of the overall part quality. This research evaluated five image segmentation methods (simple thresholding, adaptive thresholding, Sobel edge detector, Canny edge detector, and watershed transform) on the same platform for their effectiveness in isolating and identifying features in 3D-printed layers under different contrast conditions for in situ quality assessment. The performance metrics used are accuracy, precision, recall, and the Jaccard index. The experimental set-up is based on an open-frame fused filament fabrication printer augmented with a vision system. The control system software for printing and imaging (acquisition and processing) was custom developed in Python running on a Raspberry Pi. Most of the segmentation methods reliably segmented the external geometry and high-contrast internal features. The simple thresholding, Canny edge detector, and watershed transform methods did not perform well with low-contrast parts and could not reliably segment internal features when the previous layer was visible. The adaptive thresholding and Sobel edge detector methods segmented high- and low-contrast features. However, the segmentation outputs were heavily affected by textural and image noise. The research identified factors affecting the performance and limitations of these segmentation methods and contributing to the broader effort of improving in situ quality assessment in AM, such as automatic dimensional analysis of internal and external features and the overall geometry.

Enhanced satellite image resolution with a residual network and correlation filter

This study addresses the predominant challenge of very low-resolution satellite images in remote sensing applications, a common issue in satellite image-based surveillance. Existing satellite image recognition algorithms struggle with such low-resolution images, and traditional Super-Resolution (SR) techniques fall short for very low-resolution cases. We propose the Progressive Satellite Image Super-Resolution (PSISR) model to bridge this gap. Unlike current learning-based SR methods, the PSISR model specifically targets very low-resolution satellite images. In satellite image super-resolution, problems with feature fusion that result in image noise, blind spots, poor perceptual quality, and checkboard artifacts are encountered during the reconstruction process. Current models try to improve perceptual quality, but they frequently show challenges in attaining acceptable outcomes because of losses during reconstruction. Using a combined loss function, correlation filters, and a loss-aware upscaling network layer, the PSISR model presents a revolutionary methodology. The model adopts a cascading structure with dense skip connections, sequentially upscaling images by factors of 2×, 4×, and 8× through three modules. To validate the model's superiority, a study is conducted, confirming its effectiveness compared to baseline models and also training the other models using the available dataset to prove the effectiveness of the model. The PSISR model effectively addresses the challenge of extracting more features with minimal losses, resulting in high magnification during reconstruction. Our method outperforms state-of-the-art techniques, including Swin2-MoSE, MambaFormer, SRFBN and RCAN, with a PSNR improvement of up to 0.4 dB and a 0.003 SSIM enhancement across various datasets. This demonstrates the effectiveness of our approach in producing high-quality outputs, achieving a 99.25 % correlation efficiency between the generated and original images.

A genetic programming Rician noise reduction and explainable deep learning model for Alzheimer's diseases severity prediction.

Degradation of magnetic resonance imaging (MRI) remains a challenging issue, with noise being a key damaging component introduced due to a variety of environmental and mechanical factors. The aim of this research work is to addresses the issue of noise reduction and to predict Alzheimer's disease detection efficiently. First, we present a genetic programming (GP) technique for reducing Rician noise in MRI images to pre-process the dataset. To effectively reduce Rician noise, this GP approach combines a Feature Extraction component, GP Optimal Expression, and an Optimum Removal Estimation component. In the second phase, we design and develop an explainable Deep Learning framework. This framework uses a local data-driven interpretation technique based on SHAP values to investigate the relationship between the neural network's estimated AD diagnosis and the input MRI images. In addition, we handle class distribution by combining an oversampling strategy with a minority approach. Several assessment metrics are used to analyze the performance of our proposed model. The proposed method is tested on a variety of medical samples, and the results are compared to those obtained using other comparable approaches. We also test and compare our model to three cutting-edge models: DenseNet169, VGGNet15, and Inceptionv3. The empirical results show that our proposed model outperforms others, particularly in handling basic structures with limited spectral features, lower computational complexity, and less overfitting. This research worked addressed Rician noise issue in MRI images and predict AD severity prediction using explainable deep learning framework.

Learned Tensor Neural Network Texture Prior for Photon-Counting CT Reconstruction.

Photon-counting computed tomography (PCCT) reconstructs multiple energy-channel images to describe the same object, where there exists a strong correlation among different channel images. In addition, reconstruction of each channel image suffers photon count starving problem. To make full use of the correlation among different channel images to suppress the data noise and enhance the texture details in reconstructing each channel image, this paper proposes a tensor neural network (TNN) architecture to learn a multi-channel texture prior for PCCT reconstruction. Specifically, we first learn a spatial texture prior in each individual channel image by modeling the relationship between the center pixels and its corresponding neighbor pixels using a neural network. Then, we merge the single channel spatial texture prior into multi-channel neural network to learn the spectral local correlation information among different channel images. Since our proposed TNN is trained on a series of unpaired small spatial-spectral cubes which are extracted from one single reference multi-channel image, the local correlation in the spatial-spectral cubes is considered by TNN. To boost the TNN performance, a low-rank representation is also employed to consider the global correlation among different channel images. Finally, we integrate the learned TNN and the low-rank representation as priors into Bayesian reconstruction framework. To evaluate the performance of the proposed method, four references are considered. One is simulated images from ultra-high-resolution CT. One is spectral images from dual-energy CT. The other two are animal tissue and preclinical mouse images from a custom-made PCCT systems. Our TNN prior Bayesian reconstruction demonstrated better performance than other state-of-the-art competing algorithms, in terms of not only preserving texture feature but also suppressing image noise in each channel image.

Performance and Robustness of Physiological Controllers With Integrated Pressure Sensors on Inflow Cannula.

The control of left ventricular assist devices (LVADs) requires sensors and/or estimators to account for the physiological state of the patient and apply advanced controllers. Sensor characteristics are a challenge when using implantable pressure sensors because they influence the quality of physiological control and the robustness of the controlled system. The objective of this work is to investigate the performance and robustness of LVAD controllers that operate based on LVAD integrated pressure sensors. Four pressure-based LVAD controllers are tested with a HeartMate 3 that has an integrated pressure sensor. Controller sensitivity as well as robustness to sensor drift and noise are evaluated based on controller response to large changes in preload and afterload. A fail-safe strategy for sensor failure used also to investigate the reliable operation of the LVAD in realistic conditions. All tested controllers are sensitive to drifting pressure signals, which can lead to unstable behavior. The results show that two controllers are robust to noise. The other two controllers show high deviation and oscillation in cardiac output even for small noise. Additionally, a noticeable difference of the controller's response between simulated and measured pressure input was observed, indicating the need for a robust controller design. Reliable operation even in the event of a sensor failure was achieved on the basis of a fail-safe control system.

Non-iterative model-based inversion for low channel-count optical ultrasound imaginga).

Ultrasound image reconstruction is typically performed using the computationally efficient delay-and-sum algorithm. However, this algorithm is suboptimal for systems of low channel counts, where it causes significant image artefacts. These artefacts can be suppressed through model-based inversion approaches; however, their computational costs typically prohibit real-time implementations. In this work, the emerging optical ultrasound (OpUS) modality is considered, where ultrasound waves are both generated and detected using light. With this modality, imaging probes comprise very low channel counts, resulting in significant image artefacts that limit the imaging dynamic range. However, this low channel counts offer an opportunity for non-iterative ("direct") model-based inversion (DMI) on modest computational resources available in a typical workstation. When applied to both synthetic and experimental OpUS data, the presented DMI method achieved substantial reduction in image artefacts and noise, improved recovery of image amplitudes, and-after one-off pre-computation of the system matrices-significantly reduced reconstruction time, even in imaging scenarios exhibiting mild spatial inhomogeneity. Whilst here applied to an OpUS imaging system, DMI can be applied to other low channel-count imaging systems, and is therefore expected to achieve better image quality, reduce system complexity, or both, in a wide range of settings.

Simultaneous Multislice Cardiac Multimapping based on Locally Low-Rank and Sparsity Constraints

Although quantitative myocardial T1 and T2 mappings are clinically used to evaluate myocardial diseases, their application needs a minimum of 6 breath-holds to cover 3 short-axis slices. The purpose of this work is to simultaneously quantify multi-slice myocardial T1 and T2 across 3 short-axis slices in one breath-hold by combining simultaneous multi-slice (SMS) with Multimapping. An SMS-Multimapping sequence with multi-band RF excitations and Cartesian FLASH readouts was developed for data acquisition. When 3 slices are simultaneously acquired, the acceleration rate is around 12-fold, causing a highly ill-conditioned reconstruction problem. To mitigate image artifacts and noise caused by the ill-conditioning, a reconstruction algorithm based on Locally Low-Rank and Sparsity (LLRS) was developed. Validation was performed in phantoms and in vivo imaging, with 20 healthy subjects and 4 patients, regarding regional mean, precision, and scan-rescan reproducibility. The phantom imaging shows that SMS-Multimapping with LLRS accurately reconstructed multi-slice T1 and T2 maps despite a 6-fold acceleration of scan time. Healthy subject imaging shows that the proposed LLRS algorithm substantially improved image quality relative to split slice-GRAPPA. Compared with MOLLI, SMS-Multimapping exhibited higher T1 mean (1118±43ms vs 1190±49ms, P<0.01), lower precision (67±17ms vs 90±17ms, P<0.01), and acceptable scan-rescan reproducibility measured by two scans 10-minute apart (bias=1.4ms for MOLLI and 9.0ms for SMS-Multimapping). Compared with bSSFP T2 mapping, SMS-Multimapping exhibited similar T2 mean (43.5±3.3ms vs 43.0±3.5ms, P=0.64), similar precision (4.9±2.1ms vs 5.1±1.0ms, P=0.93), and acceptable scan-rescan reproducibility (bias=0.13ms for bSSFP T2 mapping and 0.55ms for SMS-Multimapping). In patients, SMS-Multimapping clearly showed the abnormality in a similar fashion as the reference methods despite using only one breath-hold. SMS-Multimapping with the proposed LLRS reconstruction can measure multi-slice T1 and T2 maps in one breath-hold with good accuracy, reasonable precision, and acceptable reproducibility, achieving a 6-fold reduction of scan time and an improvement of patient comfort.

Heuristic dense reward shaping for learning-based map-free navigation of industrial automatic mobile robots

This paper presents a map-free navigation approach for industrial automatic mobile robots (AMRs), designed to ensure computational efficiency, cost-effectiveness, and adaptability. Utilizing deep reinforcement learning (DRL), the system enables real-time decision-making without fixed markers or frequent map updates. The central contribution is the Heuristic Dense Reward Shaping (HDRS), inspired by potential field methods, which integrates domain knowledge to improve learning efficiency and minimize suboptimal actions. To address the simulation-to-reality gap, data augmentation with controlled sensor noise is applied during training, ensuring robustness and generalization for real-world deployment without fine-tuning. Training results underscore HDRS’s superior convergence speed, training stability, and policy learning efficiency compared to baselines. Simulation and real-world evaluations establish HDRS-DRL as a competitive alternative, outperforming traditional approaches, and offering practical applicability in industrial settings.