Low Resolution Research Articles

Spatial domain noise removal filtering for low-resolution digital images

<p>n this research work, six different filters are applied on a low resolution 8 b/pixel gray-scale images, which operate on small sub-images (windows of 3×3 to 11×11 pixels). The enhanced images are used to compare the efficiency of the different six filters using the peak signal to noise ratio (PSNR) image quality measure. Noise peak elimination filter (PSNR)=36.63) outperforms others, such as median filter (PSNR=36.61), while corruption estimation (PSNR=36.03) significantly cuts processing time by only processing the corrupted pixels while maintaining image details. Mean filter (PSNR=34.05) is sensitive to outliers, which cause the image's sharpness and fine features to be lost. By avoiding averaging across edges, bimodal-averaging filter (PSNR=35.30), which improves on the mean filter, chooses the mean of the biggest population. The median-mean filtering (PSNR=36.32), which combines median and mean filters and determines the output pixel by averaging the median and some nearby pixels, is another improvement above averaging.</p>

TQU-HG dataset and comparative study for hand gesture recognition of RGB-based images using deep learning

Hand gesture recognition has great applications in human-computer interaction (HCI), human-robot interaction (HRI), and supporting the deaf and mute. To build a hand gesture recognition model using deep learning (DL) with high results then needs to be trained on many data and in many different conditions and contexts. In this paper, we publish the TQU-HG dataset of large RGB images with low resolution (640×480) pixels, low light conditions, and fast speed (16 fps). TQU-HG dataset includes 60,000 images collected from 20 people (10 male, 10 female) with 15 gestures of both left and right hands. A comparative study with two branches: i) based on Mediapipe TML and ii) Based on convolutional neural networks (CNNs) (you only look once (YOLO); YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLO-Nas, single shot multiBox detector (SSD) VGG16, residual network (ResNet)18, ResNext50, ResNet152, ResNext50, MobileNet V3 small, and MobileNet V3 large), the architecture and operation of CNNs models are also introduced in detail. We especially fine-tune the model and evaluate it on TQU-HG and HaGRID datasets. The quantitative results of the training and testing are presented (F1-score of YOLOv8, YOLO-Nas, MobileNet V3 small, ResNet50 is 98.99%, 98.98%, 99.27%, 99.36%, respectively on the TQU-HG dataset and is 99.21%, 99.37%, 99.36%, 86.4%, 98.3%, respectively on the HaGRID dataset). The computation time of YOLOv8 is 6.19 fps on the CPU and 18.28 fps on the GPU.

Investigation of aluminium Whipple Shield response to hypervelocity impacts close to ballistic limit between 2.5 and 5 km/s

For impact velocities larger than a threshold velocity of 3 km/s, aluminium Whipple Shields present an enhanced protection capability with respect to monolithic protections with the same areal density. In particular, in the range between 3 and 7 km/s the projectile partially fragments and melts after impacting the bumper; modelling this transition might be complex due to the high number of parameters affecting the collision. In particular, limited data is available in literature with a systematic evaluation of such parameters.In this context, a campaign of 22 experiments was performed to assess the response of aluminium Whipple Shields to normal impacts of aluminium projectiles in the transition range up to 5 km/s. In the tests, the projectile diameter was fixed at 2.9 mm and the bumper thickness and impact velocity were systematically varied respectively at 1, 1.5, and 2 mm and between 2.6 and 5 km/s. Collected data included high-velocity videos of the debris cloud generated by the impacts at both low resolution (for all shots) and high resolution (for 19 out of 22 tests); in addition, for 10 experiments high-resolution images of both the bumper and the wall were acquired after the tests.In this paper, the experimental campaign is described and the main collected results are presented; in particular, the influence of the different impact parameters is discussed. Experimental results are finally compared with numerical simulations conducted with a smoothed particle hydrodynamics method available in IMPETUS Solver.

Geo-spatial epidemiology of gallbladder cancer in Bihar, India

Recently, a substantial increase in gallbladder cancer (GBC) cases has been reported in Bihar, India. The region's groundwater can naturally contain harmful concentrations of arsenic, which appears to be epidemiologically linked to the unusually high incidence. However, the root causes remain largely unexplored. Recent findings of uranium in the state's groundwater may also have associations. This study investigates the geo-spatial epidemiology of GBC in Bihar, India—with a focus on the correlation between environmental carcinogens, particularly arsenic and uranium in groundwater, and the incidence of GBC. Utilizing data from 8460 GBC patients' registration records over an 11-year period at a single health center, the research employs Semi-parametric Geographically Weighted Poisson Regression (S-GWPR) to account for non-stationarity associations and explores significant factors contributing to GBC prevalence at a subdistrict level. The S-GWPR model outperformed the standard Poisson regression model. The estimates suggest that arsenic and uranium concentrations in groundwater did not present significant associations; however, this could be due to the lower resolution of this data at the district level, necessitating higher resolution data for accurate estimates. Other socio-environmental factors included demonstrated significant regional heterogeneity in their association with GBC prevalence. Notably, each 1 % increase in the coverage of well- and canal-irrigated areas is associated with a maximum of 3.0 % and 5.2 % rise in the GBC incidence rate, respectively, likely attributable to carcinogen exposure from irrigation water. Moreover, distance to the health center and domestic electricity connections appear to influence the number of reported GBC cases. The latter suggests that access to electricity might have facilitated the use of groundwater pumps—increasing exposure to carcinogens. The results underscore the necessity for targeted health policies and interventions based on fine-resolution spatial analysis, as well as ongoing environmental monitoring and research to better understand the multifaceted risk factors contributing to GBC.

Inter-scanner super-resolution of 3D cine MRI using a transfer-learning network for MRgRT

Objective. Deep-learning networks for super-resolution (SR) reconstruction enhance the spatial-resolution of 3D magnetic resonance imaging (MRI) for MR-guided radiotherapy (MRgRT). However, variations between MRI scanners and patients impact the quality of SR for real-time 3D low-resolution (LR) cine MRI. In this study, we present a personalized super-resolution (psSR) network that incorporates transfer-learning to overcome the challenges in inter-scanner SR of 3D cine MRI. Approach: Development of the proposed psSR network comprises two-stages: (1) a cohort-specific SR (csSR) network using clinical patient datasets, and (2) a psSR network using transfer-learning to target datasets. The csSR network was developed by training on breath-hold and respiratory-gated high-resolution (HR) 3D MRIs and their k-space down-sampled LR MRIs from 53 thoracoabdominal patients scanned at 1.5 T. The psSR network was developed through transfer-learning to retrain the csSR network using a single breath-hold HR MRI and a corresponding 3D cine MRI from 5 healthy volunteers scanned at 0.55 T. Image quality was evaluated using the peak-signal-noise-ratio (PSNR) and the structure-similarity-index-measure (SSIM). The clinical feasibility was assessed by liver contouring on the psSR MRI using an auto-segmentation network and quantified using the dice-similarity-coefficient (DSC). Results. Mean PSNR and SSIM values of psSR MRIs were increased by 57.2% (13.8–21.7) and 94.7% (0.38–0.74) compared to cine MRIs, with the reference 0.55 T breath-hold HR MRI. In the contour evaluation, DSC was increased by 15% (0.79–0.91). Average time consumed for transfer-learning was 90 s, psSR was 4.51 ms per volume, and auto-segmentation was 210 ms, respectively. Significance. The proposed psSR reconstruction substantially increased image and segmentation quality of cine MRI in an average of 215 ms across the scanners and patients with less than 2 min of prerequisite transfer-learning. This approach would be effective in overcoming cohort- and scanner-dependency of deep-learning for MRgRT.

Fast image-free autofocus method for passive FSPI microscopy

Autofocus is crucial for capturing sharp images with imaging devices for information acquisition. Traditional autofocus strategies based on post-processing become less efficient for passive FSPI microscopy of yet low temporal resolution. In this Letter, a fast and image-free autofocus system is proposed for passive FSPI microscopy. Based on the complementary design of an optical path, the system can measure the focus degree at 5000 fps while maintaining a high light efficiency for imaging. The proposed system can be easily combined with existing trinocular microscopes, which provides a welcomed boost to the practicability of passive FSPI microscopy.

Mexican Coccoloba uvifera L. Leaf and Fruit Extracts: Identification of Pentacyclic Triterpenes and Volatile Profile by GC-MS.

Mexican Coccoloba uvifera fruit contains polyphenols, flavonoids, and anthocyanins, while in the leaves, lupeol, α- and β-amyrin have been previously identified by HPLC. However, the low resolution by HPLC of pentacyclic triterpenes (PTs) is a limitation. Moreover, the volatile profile of C. uvifera fruit is still unknown. Therefore, this study aimed to identify PTs in C. uvifera leaf and fruit extracts by CG-MS analysis and to determine the volatile profile of C. uvifera pulp by headspace solid-phase microextraction. The results showed trimethylsilylated compounds of standards lupeol, α- and β-amyrin, indicating that the silylation reaction was suitable. These trimethylsilylated compounds were identified in leaf and fruit extracts. The fruit volatile profile revealed the presence of 278 esters, 20 terpenes, 9 aldehydes, 5 alcohols, and 4 ketones. The fruit showed a high content of esters and terpenes. Due to their flavour properties, esters are essential for the food, cosmetics, and pharmaceutics industries. Moreover, terpenes in the fruit, such as menthone, β-elemene, junipene, and β-caryophyllene have the potential as anticancer and phytopathogen agents. The results indicated that GC-MS is an alternative to HPLC approaches for identifying PTs. Besides, identifying volatile compounds in the fruit will increase the value of this plant and expand its application. Identifying PTs and volatile compounds in Mexican C. uvifera leads to a better understanding of the potential benefits of this plant. This would increase the consumption of Mexican C. uvifera fresh or as functional ingredients in nutraceutical or pharmaceutical products.

Super resolution using sparse sampling at portable ultra-low field MR

Ultra-low field (ULF) magnetic resonance imaging (MRI) holds the potential to make MRI more accessible, given its cost-effectiveness, reduced power requirements, and portability. However, signal-to-noise ratio (SNR) drops with field strength, necessitating imaging with lower resolution and longer scan times. This study introduces a novel Fourier-based Super Resolution (FouSR) approach, designed to enhance the resolution of ULF MRI images with minimal increase in total scan time. FouSR combines spatial frequencies from two orthogonal ULF images of anisotropic resolution to create an isotropic T2-weighted fluid-attenuated inversion recovery (FLAIR) image. We hypothesized that FouSR could effectively recover information from under-sampled slice directions, thereby improving the delineation of multiple sclerosis (MS) lesions and other significant anatomical features. Importantly, the FouSR algorithm can be implemented on the scanner with changes to the k-space trajectory. Paired ULF (Hyperfine SWOOP, 0.064 tesla) and high field (Siemens, Skyra, 3 Tesla) FLAIR scans were collected on the same day from a phantom and a cohort of 10 participants with MS or suspected MS (6 female; mean ± SD age: 44.1 ± 4.1). ULF scans were acquired along both coronal and axial planes, featuring an in-plane resolution of 1.7 mm × 1.7 mm with a slice thickness of 5 mm. FouSR was evaluated against registered ULF coronal and axial scans, their average (ULF average) and a gold standard SR (ANTs SR). FouSR exhibited higher SNR (47.96 ± 12.6) compared to ULF coronal (36.7 ± 12.2) and higher lesion conspicuity (0.12 ± 0.06) compared to ULF axial (0.13 ± 0.07) but did not exhibit any significant differences contrast-to-noise-ratio (CNR) compared to other methods in patient scans. However, FouSR demonstrated superior image sharpness (0.025 ± 0.0040) compared to all other techniques (ULF coronal 0.021 ± 0.0037, q = 5.9, p-adj. = 0.011; ULF axial 0.018 ± 0.0026, q = 11.1, p-adj. = 0.0001; ULF average 0.019 ± 0.0034, q = 24.2, p-adj. < 0.0001) and higher lesion sharpness (−0.97 ± 0.31) when compared to the ULF average (−1.02 ± 0.37, t(543) = −10.174, p = <0.0001). Average blinded qualitative assessment by three experienced MS neurologists showed no significant difference in WML and sulci or gyri visualization between FouSR and other methods. FouSR can, in principle, be implemented on the scanner to produce clinically useful FLAIR images at higher resolution on the fly, providing a valuable tool for visualizing lesions and other anatomical structures in MS.

Super-resolution reconstruction of single image for latent features

AbstractSingle-image super-resolution (SISR) typically focuses on restoring various degraded low-resolution (LR) images to a single high-resolution (HR) image. However, during SISR tasks, it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture features. This challenge can lead to issues such as model collapse, lack of rich details and texture features in the reconstructed HR images, and excessive time consumption for model sampling. To address these problems, this paper proposes a Latent Feature-oriented Diffusion Probability Model (LDDPM). First, we designed a conditional encoder capable of effectively encoding LR images, reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed images. We then employed a normalized flow and multimodal adversarial training, learning from complex multimodal distributions, to model the denoising distribution. Doing so boosts the generative modeling capabilities within a minimal number of sampling steps. Experimental comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics, providing a fresh perspective for tackling SISR tasks.

Remote sensing applications for prescribed burn research

Prescribed burning is a key management strategy within fire-adapted systems, and improved monitoring approaches are needed to evaluate its effectiveness in achieving social-ecological outcomes. Remote sensing provides opportunities to analyse the impacts of prescribed burning, yet a comprehensive understanding of the applications of remote sensing for prescribed burn research is lacking. We conduct a literature review of 120 peer-reviewed publications to synthesise the research aims, methodologies, limitations and future directions of remote sensing for the analysis of prescribed fire. Studies evaluating management outcomes found prescribed burning effective for wildfire risk reduction, yet few analysed co-benefits or trade-offs with other management goals. Most studies use passive, spaceborne, low spatial resolution sensors, characterised in the literature as consistent and accessible data sources but limited in detecting small, low-severity and short-duration fires characteristic of prescribed burns. In contrast, active remote sensing approaches including LiDAR are less frequently employed, but show promise for highly accurate, spatially explicit 3D vegetation and fuel load mapping. Remote sensing advances toward higher spatial resolution, more frequent revisit, denser spectral sampling and more data across the electromagnetic spectrum are critical to advancing prescribed fire research, addressing current methodological gaps, and improving fuels and fire management capacity.

The correlation between left atrium voltage mapping and 3D late gadolinium enhancement cardiac magnetic resonance imaging findings in patients with known atrial fibrillation

Abstract Introduction The non-invasive cardiac magnetic resonance imaging-based late gadolinium enhancement (LGE-CMR) of the left atrium (LA) fibrosis distribution and degree can be utilized for preprocedural planning and real-time guidance of AF ablation. There is an ongoing discussion with somewhat conflicting results regarding the accuracy of LGE-CMR of the LA as compared with the invasive electroanatomical mapping (EAM). Earlier data mainly encompassed post-ablation patients with an EAM involving a relatively small number of acquired points. Aim To evaluate the accuracy of LGE-CMR-based LA fibrosis using the ADAS 3D LA software among ablation-naive AF patients as compared with EAM using the CARTO-3 navigation system. Methods Patients with non-permanent AF, naive to previous ablation/cardiac surgery, who underwent CARTO-based AF ablation with preceding LGE-CMR scan were selected for the study. The 3D substrate atrial maps were generated based on the LGE-CMR (CMR-fib) using the ADAS 3D LA software and the CARTO EAM (EAM-fib). A fibrosis threshold of 1.2-1.32 was established for the CMR-fib using the Image Intensity Ratio whereas a threshold of 0.2-0.5mV was applied for the EAM-fib. The two 3D-maps were assessed for fibrosis using anatomically synchronized quantitative point by point comparison. The EAM was set as the gold standard. Results The study included 26,777 points acquired from 17 patients (aged 65.8±7.92 yrs with 6 females and 10 paroxysmal AF cases). CMR-fib showed 22.0% of LGE, while the EAM-fib demonstrated 48.8% of low voltage areas. The CMR-fib vs. EAM-fib point by point comparison concluded with an agreement of 73.1% and interrater reliability of 0.45 (p<0.001). The sensitivity and specificity were 45.0% and 99.9%, respectively, while the PPV and NPV were 99.8% and 65.6%, respectively. The Area Under the Curve (AUC) was calculated as 72.4. Increase of the EAM-fib threshold for scar (to 0.5-1.5mV) led to a further decrease in the assessment diagnostic performance and the interrater reliability. The lowest agreement was tested in the LA roof segments and the higher in the inferoseptal and posterior walls. Conclusion The LGE-CMR underestimates the presence of scar as presented by the EAM, however, those detected are mostly accurate. The low sensitivity that was demonstrated between the two methods may be attributed to the lower spatial resolution of the CMR as well as the superficial subendocardial layer mapped by EAM as compared with the averaged transmural mapping of the CMR-fib. A follow-up larger-scale study is required to assess the accuracy of LGE-CMR-based assessment of fibrosis, considering various wall thicknesses and establishing the scar threshold on the ADAS software, particularly for ablation-naive patients.

D.5 Low density scalp electrical source imaging of the ictal onset zone network using source coherence maps

Background: This study investigates the utility of low-density scalp electrical source imaging (LD-ESI) of the Ictal Onset Zone (IOZ) and interictal spike ripple high frequency oscillation (ISRHFO) networks using Source Coherence Maps (SCM) in the surgical evaluation of children with medically refractory epilepsy. Invasive intracranial monitoring, the gold standard for determining epileptogenic zones, has limited spatial sampling. SCM presents a promising new non-invasive diagnostic technique. Methods: This was a retrospective review of 11 patients who underwent focal resections. SCMs were generated using Standardized Low Resolution Electromagnetic Tomography (sLORETA). SCM concordance to resection margins was assessed, noting outcomes at 3 years. Results: For 7/11 cases, ictal SCMs included the resection, and 5/7 achieved seizure freedom, indicating inclusion of the epileptogenic zone. For the 2/7 not seizure-free, the IOZ networks on the SCMs extended beyond resection margins, suggesting the epileptogenic zone also extended beyond the resection. Interictal spike ripple ESI and ISRHFO SCM were performed for 7/11, with 3/7 included in the resection and all 3 seizure-free. Conclusions: These findings may support LD-ESI of the IOZ and ISRHFO network using SCM as promising methods complementary to ictal and interictal ESI in pediatric epilepsy surgical workup, guiding electrode placement for intracranial monitoring to identify the epileptogenic zone.

Utilizing the Sentinel-6 Michael Freilich Equivalent Number of Looks for Sea State Applications

Sentinel-6 Michael Freilich (S6-MF) is the first altimeter operating in a continuous high-rate pulse mode, i.e., interleaved mode. This ensures the generation of low-resolution (LR) mode measurements with a pulse repetition frequency (PRF) of ∼9 kHz (variable along the orbit) for the Ku-band as well as the processing of high-resolution (HR) echoes on ground. This operating mode provides an elevated number of highly correlated single looks with respect to the fewer number, weakly correlated echoes of Jason-3 altimeter. A theoretical model is exploited to envisage the correlation properties of S6-MF pulse limited waveform echoes for different sea-state conditions; after that, the model is validated by comparison with the equivalent number of looks (ENL) empirically estimated from real data. The existence of a significant dependence of the statistical properties on the range is verified, and its impact on the precision and on the accuracy in the estimation of the geophysical parameters is assessed in case of the 9 kHz PRF of S6-MF. By applying pulse decimation before the multilook processing, an investigation on new processing techniques is performed, aimed at exploiting the higher ENL in S6-MF low-resolution mode waveforms. It is shown that a bias of less than 0.4 cm is found for SSH and about 1.5 cm for SWH at SWH = 2 m when the decimated waveforms processing is compared with full high-PRF processing.

#2411 Inflammation in hypervolemic hemodialysis patients is linked to increased caspase-4 activity

Abstract Background and Aims Hypervolemia is associated with inflammation in hemodialysis patients (HD). How hypervolemia triggers inflammation is not entirely known. We hypothesized that hypervolemia affects signaling via the NF-kB transcription factor system and thus analyzed the phosphorylated forms of RelA and RelB in immune cells of hypervolemic (H) and normovolemic (N) patients. Further on, we speculated that the higher inflammatory load in H in comparison to N patients could affect the sensor of intracellular LPS activity, namely caspase-4. Method Forty HD patients were categorized by bioimpedance measurement in normovolemic (N; 23) and hypervolemic (H; 17) groups. A caspase activity assay in combination with a specific Caspase-4/-5 inhibitor was used to detect Caspase-4/-5 activity in isolated PBMCs by flow-cytometry. The transcription factors RelA (pS529) and RelB (pS552) were analyzed by phopho-flow cytometry. Endotoxin as well as IL-6 and TNF-α were detected by ELISA technique. Results Hypervolemic patients were older and had more frequent diabetes. Although the endotoxin levels (EU/ml) in the serum of both groups were not different (N: 3.3 ± 1.6 vs. H 2.9 ± 1.4), Caspase-4/-5 activity (frequency of CD14+Caspase-4+), linked to intracellular endotoxin detection, was significantly elevated in H patients (N: 52 + 17% vs. H: 65 + 17%, p < 0.01). While the frequency and expression density of Rel A expressing immune cells were not different, the monocytic frequency of cells staining positive for RelB (pS552) was significantly elevated in N patients (N: 37.7 ± 27.1 vs. H: 20.4 ± 17.1, p=0.038). Among inflammatory cytokines (pg/ml) measured in the serum of patients, only IL-6 (N: 5.8 ± 10.5 vs H: 7.2 ± 6.8, p = 0.01) but not TNF-α (N: 7.7 ± 1.1 vs. H: 7.5 ± 0.7, p = 0.580) was increased in H patients. Conclusion The lower frequency of pRelB-positive monocytes in hypervolemic patients may be linked to a functional impairment to resolve inflammatory circles. The elevated caspase-4/-5 activity in H patients, however, can be interpreted as a higher intracellular LPS-load in H patients. Therefore, both higher inflammatory load and lower inflammatory resolution capacities are characteristics of H patients.

Multi-resolution Twinned Residual Auto-Encoders (MR-TRAE)—A Novel DL Model for Image Multi-resolution

AbstractIn this paper, we design and evaluate the performance of the Multi-resolution Twinned Residual Auto-Encoders (MR-TRAE) model, a deep learning (DL)-based architecture specifically designed for achieving multi-resolution super-resolved images from low-resolution (LR) inputs at various scaling factors. For this purpose, we expand on the recently introduced Twinned Residual Auto-Encoders (TRAE) paradigm for single-image super-resolution (SISR) to extend it to the multi-resolution (MR) domain. The main contributions of this work include (i) the architecture of the MR-TRAE model, which utilizes cascaded trainable up-sampling modules for progressively increasing the spatial resolution of low-resolution (LR) input images at multiple scaling factors; (ii) a novel loss function designed for the joint and semi-blind training of all MR-TRAE model components; and (iii) a comprehensive analysis of the MR-TRAE trade-off between model complexity and performance. Furthermore, we thoroughly explore the connections between the MR-TRAE architecture and broader cognitive paradigms, including knowledge distillation, the teacher-student learning model, and hierarchical cognition. Performance evaluations of the MR-TRAE benchmarked against state-of-the-art models (such as U-Net, generative adversarial network (GAN)-based, and single-resolution baselines) were conducted using publicly available datasets. These datasets consist of LR computer tomography (CT) scans from patients with COVID-19. Our tests, which explored multi-resolutions at scaling factors $$\times (2,4,8)$$ × ( 2 , 4 , 8 ) , showed a significant finding: the MR-TRAE model can reduce training times by up to $$60\%$$ 60 % compared to those of the baselines, without a noticeable impact on achieved performance.

Convolutional neural networks combined with conventional filtering to semantically segment plant roots in rapidly scanned X-ray computed tomography volumes with high noise levels

BackgroundX-ray computed tomography (CT) is a powerful tool for measuring plant root growth in soil. However, a rapid scan with larger pots, which is required for throughput-prioritized crop breeding, results in high noise levels, low resolution, and blurred root segments in the CT volumes. Moreover, while plant root segmentation is essential for root quantification, detailed conditional studies on segmenting noisy root segments are scarce. The present study aimed to investigate the effects of scanning time and deep learning-based restoration of image quality on semantic segmentation of blurry rice (Oryza sativa) root segments in CT volumes.ResultsVoxResNet, a convolutional neural network-based voxel-wise residual network, was used as the segmentation model. The training efficiency of the model was compared using CT volumes obtained at scan times of 33, 66, 150, 300, and 600 s. The learning efficiencies of the samples were similar, except for scan times of 33 and 66 s. In addition, The noise levels of predicted volumes differd among scanning conditions, indicating that the noise level of a scan time ≥ 150 s does not affect the model training efficiency. Conventional filtering methods, such as median filtering and edge detection, increased the training efficiency by approximately 10% under any conditions. However, the training efficiency of 33 and 66 s-scanned samples remained relatively low. We concluded that scan time must be at least 150 s to not affect segmentation. Finally, we constructed a semantic segmentation model for 150 s-scanned CT volumes, for which the Dice loss reached 0.093. This model could not predict the lateral roots, which were not included in the training data. This limitation will be addressed by preparing appropriate training data.ConclusionsA semantic segmentation model can be constructed even with rapidly scanned CT volumes with high noise levels. Given that scanning times ≥ 150 s did not affect the segmentation results, this technique holds promise for rapid and low-dose scanning. This study offers insights into images other than CT volumes with high noise levels that are challenging to determine when annotating.

A temporal enhanced semi-supervised training framework for needle segmentation in 3D ultrasound images

Objective. Automated biopsy needle segmentation in 3D ultrasound images can be used for biopsy navigation, but it is quite challenging due to the low ultrasound image resolution and interference similar to the needle appearance. For 3D medical image segmentation, such deep learning networks as convolutional neural network and transformer have been investigated. However, these segmentation methods require numerous labeled data for training, have difficulty in meeting the real-time segmentation requirement and involve high memory consumption. Approach. In this paper, we have proposed the temporal information-based semi-supervised training framework for fast and accurate needle segmentation. Firstly, a novel circle transformer module based on the static and dynamic features has been designed after the encoders for extracting and fusing the temporal information. Then, the consistency constraints of the outputs before and after combining temporal information are proposed to provide the semi-supervision for the unlabeled volume. Finally, the model is trained using the loss function which combines the cross-entropy and Dice similarity coefficient (DSC) based segmentation loss with mean square error based consistency loss. The trained model with the single ultrasound volume input is applied to realize the needle segmentation in ultrasound volume. Main results. Experimental results on three needle ultrasound datasets acquired during the beagle biopsy show that our approach is superior to the most competitive mainstream temporal segmentation model and semi-supervised method by providing higher DSC (77.1% versus 76.5%), smaller needle tip position (1.28 mm versus 1.87 mm) and length (1.78 mm versus 2.19 mm) errors on the kidney dataset as well as DSC (78.5% versus 76.9%), needle tip position (0.86 mm versus 1.12 mm) and length (1.01 mm versus 1.26 mm) errors on the prostate dataset. Significance. The proposed method can significantly enhance needle segmentation accuracy by training with sequential images at no additional cost. This enhancement may further improve the effectiveness of biopsy navigation systems.

The 2019 Raikoke eruption as a testbed used by the Volcano Response group for rapid assessment of volcanic atmospheric impacts

Abstract. ​​​​​​​The 21 June 2019 Raikoke eruption (48° N, 153° E) generated one of the largest amounts of sulfur emission to the stratosphere since the 1991 Mt. Pinatubo eruption. Satellite measurements indicate a consensus best estimate of 1.5 Tg for the sulfur dioxide (SO2) injected at an altitude of around 14–15 km. The peak Northern Hemisphere (NH) mean 525 nm stratospheric aerosol optical depth (SAOD) increased to 0.025, a factor of 3 higher than background levels. The Volcano Response (VolRes) initiative provided a platform for the community to share information about this eruption which significantly enhanced coordination efforts in the days after the eruption. A multi-platform satellite observation subgroup formed to prepare an initial report to present eruption parameters including SO2 emissions and their vertical distribution for the modeling community. It allowed us to make the first estimate of what would be the peak in SAOD 1 week after the eruption using a simple volcanic aerosol model. In this retrospective analysis, we show that revised volcanic SO2 injection profiles yield a higher peak injection of the SO2 mass. This highlights difficulties in accurately representing the vertical distribution for moderate SO2 explosive eruptions in the lowermost stratosphere due to limited vertical sensitivity of the current satellite sensors (±2 km accuracy) and low horizontal resolution of lidar observations. We also show that the SO2 lifetime initially assumed in the simple aerosol model was overestimated by 66 %, pointing to challenges for simple models to capture how the life cycle of volcanic gases and aerosols depends on the SO2 injection magnitude, latitude, and height. Using a revised injection profile, modeling results indicate a peak NH monthly mean SAOD at 525 nm of 0.024, in excellent agreement with observations, associated with a global monthly mean radiative forcing of −0.17 W m−2 resulting in an annual global mean surface temperature anomaly of −0.028 K. Given the relatively small magnitude of the forcing, it is unlikely that the surface response can be dissociated from surface temperature variability.

KBMeasure: small sample in-situ damage automatic measurement method based on the combination of keypoints detection and binocular 3D reconstruction for aero-engine blades

Size is one of the important bases for the level assessment of aero-engine blade damage and the disposal method selection for damaged blades. Therefore, research on in-situ damage measurement of aero-engine blades is conducted in this paper. We break the inherent pipeline of "3D reconstruction and manual annotation of keypoints" in traditional damage measurement methods, and propose an in-situ damage automatic measurement method (KBMeasure) based on the combination of damage keypoints intelligent detection and binocular 3D reconstruction. KBMeasure replaces the manual annotation of damage keypoints, improves the damage measurement efficiency, and reduces the dependence on professional inspectors. The proposed method also overcomes the problem of high computational cost and low efficiency caused by redundant 3D reconstruction of the entire damaged area. For the characteristics of large changes in damage scale, low image resolution, the requirement of high-precision keypoints positioning, limited annotated data, and lightweight deployment in aero-enginge blade damage measurement task, a novel blade damage keypoints detection model (DKeyDet) with top-down framework is designed by introducing coordinate classification, semi-supervised learning, and knowledge distillation. Then, intersecting optical axis binocular model is used to estimate the spatial coordinates of the detected keypoints and compute the size of damage. The keypoints detection average precision (AP) and average recall (AR) of our method are 87.6 and 91.3, and the damage measurement size error (SE) is 0.08, which is superior to existing methods. This research provides a new theoretical support for in-situ damage automatic measurement for aero-engine in service, and provides what we believe is a novel idea for damage measurement of industrial components in other fields.

Infrared image fault diagnosis based on dual-stream attention convolution network

Abstract The development of infrared image detection technology has improved the real-time performance and safety of fault diagnosis. To address the problem of insufficient features caused by the low resolution of infrared images, a fault diagnosis model for induction motor infrared images based on Dual-Stream Attention Convolutional (DSAC) is proposed. Firstly, this model employs a dual-stream convolutional neural network to extract spatial features in the X and Y directions of the infrared images separately. This dual-stream structure allows the network to simultaneously learn and utilize information from both X and Y directions, enabling a more comprehensive capture of temperature distribution and variation trends in the infrared images. Then, a convolutional attention mechanism is introduced to assign weights to the obtained features. The attention maps generated by the convolutional attention layer enhance key features while suppressing unimportant information, thereby enhancing the focusing performance of the DSAC model on critical features. Finally, the weighted dual-stream features are fused to achieve the goal of fault diagnosis for induction motor infrared images. The DSAC model is validated using the induction motor infrared image dataset from
Babol Noshirvani University of Technology, demonstrating excellent diagnostic accuracy and speed. Moreover, it alleviates the problem of sample imbalance to some extent in small-sample datasets, providing a feasible solution for fault diagnosis of induction motor infrared images.