Accurate Image Research Articles

Isotropic Brain MRI Reconstruction from Orthogonal Scans Using 3D Convolutional Neural Network.

As an alternative to true isotropic 3D imaging, image super-resolution (SR) has been applied to reconstruct an isotropic 3D volume from multiple anisotropic scans. However, traditional SR methods struggle with inadequate performance, prolonged processing times, and the necessity for manual feature extraction. Motivated by the exceptional representational ability and automatic feature extraction of convolutional neural networks (CNNs), in this work, we present an end-to-end isotropic MRI reconstruction strategy based on deep learning. The proposed method is based on 3D convolutional neural networks (3D CNNs), which can effectively capture the 3D structural features of MRI volumes and accurately predict potential structure. In addition, the proposed method takes multiple orthogonal scans as input and thus enables the model to use more complementary information from different dimensions for precise inference. Experimental results show that the proposed algorithm achieves promising performance in terms of both quantitative and qualitative assessments. In addition, it can process a 3D volume with a size of 256 × 256 × 256 in less than 1 min with the support of an NVIDIA GeForce GTX 1080Ti GPU, which suggests that it is not only a quantitatively superior method but also a practical one.

3D-Aligner: advanced computational tool for correcting image distortion in expansion microscopy

Expansion Microscopy (ExM) is an innovative and cost-effective super-resolution microscopy technique that has become popular in cell biology research. It achieves super-resolution by physically expanding specimens. Since its introduction, ExM has undergone continuous methodological developments to enhance its resolution and labeling capabilities. However, ExM imaging often encounters sample drift during image acquisition due to the physical movement of the expanded hydrogel, posing a significant challenge for accurate image reconstruction. Despite many proposed experimental solutions to mitigate sample drift, a universal solution has yet to be established. In response to this challenge, we developed 3D-Aligner, an advanced and user-friendly image analysis tool designed to computationally correct drift in ExM images for precise three-dimensional image reconstruction and downstream quantification. We demonstrate that 3D-Aligner effectively determines and corrects drift in ExM images with different expansion rates and various fluorescently labeled biological targets, showcasing its capabilities and robustness in drift correction. Additionally, we validate the precision of 3D-Aligner by comparing drift values across different labeled targets and highlight the importance of drift correction in quantification of biological structures.

Kilohertz volumetric imaging of in-vivo dynamics using squeezed light field microscopy.

Volumetric functional imaging of transient cellular signaling and motion dynamics poses a significant challenge to current microscopy techniques, primarily due to limitations in hardware bandwidth and the restricted photon budget within short exposure times. In response to this challenge, we present squeezed light field microscopy (SLIM), a computational imaging method that enables rapid detection of high-resolution three-dimensional (3D) light signals using only a single, low-format camera sensor area. SLIM pushes the boundaries of 3D optical microscopy, achieving over one thousand volumes per second across a large field of view of 550 μm in diameter and 300 μm in depth with a spatial resolution of 3.6 μm laterally and 6 μm axially. Using SLIM, we demonstrated blood cell velocimetry across the embryonic zebrafish brain and in a free-moving tail exhibiting high-frequency swinging motion. The millisecond temporal resolution also enables accurate voltage imaging of neural membrane potentials in the leech ganglion, and in the hippocampus of behaving mice. These results collectively establish SLIM as a versatile and robust imaging tool for high-speed microscopy applications.

Iliac vein obstruction: accuracy of Direct Multidetector Computed Tomographic Venography and duplex ultrasound.

To evaluate the accuracy of direct computed tomographic venography (DCTV) and duplex ultrasound (DUS) in the identification of iliac vein obstruction in highly symptomatic patients with severe chronic venous disease (CVD) compared with intravascular ultrasound (IVUS). this study involved patients who had advanced CVD (CEAP C3-6). All patients underwent DCTV, venous duplex scanning, and IVUS. The presence of iliac vein obstruction was detected, and degree of obstruction was classified into 3 grades (grade I, 0% to 49%; grade II, 50% to 79%; and grade III, 80% or greater). The sensitivity, specificity, PPV, NPV, and accuracy were calculated for each modality compared with IVUS. Inter-observer agreement was assessed using the κ coefficient. of 94 patients with CVD, IVUS identified iliac vein obstruction in 55 (58.5%) patients (25.5% was grade 1, 27.3% was grade 2, 47.3% was grade 3). The sensitivity, specificity of DCTV in diagnosing obstruction was (96%, 95% in grade 1; 100%, 100% grade 2; 100%, 100% in grade 3, respectively). The sensitivity, specificity of DUS was (63.9%, 65% in grade 1; 68%, 82% in grade 2, and 70%, 85% in grade 3, respectively). The overall agreement of DUS was 0.73 (95% CI, 0.70-0.79), and DCTV was 0.96 (95% CI, 0.91- 0.97). DUS is a reasonable initial imaging modality for the identification of significant iliac vein obstruction. DCTV provides an accurate reproducible imaging for accurate estimation needed for treatment planning. Direct CT Venography provides accurate reproducible radiological information required for treatment planning of patients with Iliac vein obstruction including accurate assessment of site of obstruction, its morphology, and degree of obstruction and it can replace the use of indirect CT venography using smaller amount of contrast media with accurate diagnosis.

Gradient transformation-weighted centroid: enhancing peak localization precision in the line chromatic confocal sensing system

Line chromatic confocal imaging (LCI) is an advanced system used for quick and accurate three-dimensional surface imaging without vertical mechanical scanning. It is extensively employed for fast industrial inspection. For high-speed and accurate measurement characteristics of the LCI technique, a rapid and accurate peak localization method is important. In this paper, we propose a gradient transform weight centroid method (GTWC) to improve the accuracy of peak positioning, without compromising high-speed characteristics. This method helps reconstruct the 3D profile at a higher axial resolution by reducing the full width at half maximum (FWHM) without altering the original structure of the LCI system. Both simulations and experiments show the feasibility and good performance of this method.

High-precision geometric positioning for optical remote sensing satellite in dynamic imaging

ABSTRACT Dynamic imaging of optical remote sensing satellites refers to the active acquisition of images while the satellite is maneuvering at a high angular velocity, which significantly enhances the efficiency and application value of remote sensing imagery. However, due to the influence of rapid maneuvering, the random error of star sensors significantly increases, resulting in a decrease in the geometric positioning accuracy of dynamic imaging remote sensing images. This paper proposes a dynamic fusion method for multisource attitude measurement data based on the noise adaptive estimation and bidirectional filter to achieve high-precision attitude determination and geometric positioning in dynamic imaging. Based on the measurement error model of star sensors, the weights of the star sensor and gyroscope are adaptively adjusted in multisource data fusion to reduce the impact of star sensor measurement errors on the gyroscope. Moreover, a bidirectional fusion filter that includes a low-velocity maneuvering stage is proposed to realize the optimal estimation of the satellite attitude parameter. The simulation data and onboard data of the Luojia3-01 (LJ3–01) satellite were tested to verify the effectiveness of the proposed method. The geometric positioning accuracy of the staring images of LJ3–01 improved from 10.048 m to 7.538 m. The registration accuracy of the sequential images improved from 3.568 pixels to 1.179 pixels. The proposed method can significantly improve the attitude determination accuracy and the geometric positioning accuracy of LJ3–01 satellite staring images. Moreover, for simulation data with various angular velocities, the attitude determination accuracies of the proposed method are better than 0.93”. The experimental results show that the proposed method can achieve high-precision attitude determination in dynamic imaging, reaching the accuracy in the traditional passive imaging.

Light-Triggered Plasmonic DNAzyme Walker Enables Precise Subcellular Molecular Imaging with Reduced Off-Mitochondria Signal Leakage.

The development of highly sensitive and precise imaging techniques capable of visualizing crucial molecules at the subcellular level is essential for elucidating mitochondrial functions and uncovering novel mechanisms in biological processes. However, traditional molecular imaging strategies are still limited by off-mitochondria signal leakage because of the "always-active" sensing mode. To address this limitation, we have developed a light-triggered activation sequence activated plasmonic DNAzyme walker (PDW) for accurate subcellular molecular imaging by the combination of an organelle localized strategy, upconversion nanotechnology, and a plasmon enhanced fluorescence (PEF) technique. Exploiting the advantage of light activation enables precise control over when and where to activate the probe's sensing function, effectively reducing off-mitochondria signal leakage as validated by the dynamic monitoring of changes in off-mitochondria signals during the mitochondrial entry process. Furthermore, by leveraging the PEF capability of triangular gold nanoprisms (Au NPRs), the fluorescence intensity can be enhanced by approximately 11.9 times, ensuring highly sensitive and accurate subcellular molecular imaging.

Gradient Magnitude Based Image Classification and Edge Detection for Pattern Recognition in Grayscale Images

Image classification is a crucial technique in digital image processing, used in various applications such as object recognition, surveillance systems, and medical image analysis. This research explores the use of gradient magnitude-based edge detection and Robert's Cross methods in improving the classification accuracy of grayscale images. Edge detection is used to identify object boundaries, while gradient magnitude amplifies intensity differences, thus clarifying existing patterns. Through experiments conducted on grayscale images, the results show that this method is able to detect edges with significant accuracy. The gradient values obtained from the combination of Rx and Ry matrices give a strong indication of the presence of edges, which plays an important role in image classification. With an accuracy of 75%, the method proved to be effective, although there are still challenges in dealing with images with high noise or low contrast. The conclusion of this study shows that the combination of edge detection and gradient magnitude is a promising approach for image classification, providing results that can be applied in various domains, including medical and surveillance. Further research is recommended to optimize this approach and extend its application to more complex datasets.

Selective and multi-scale fusion Mamba for medical image segmentation

Given the high variability in the morphology and size of lesion areas in medical images, accurate medical image segmentation requires both precise positioning of global contours and careful processing of local boundaries. This emphasizes the importance of fusing multi-scale local and global features. However, existing CNN and Transformer-based models are often limited by high parameter counts and complex calculations, making it difficult to efficiently integrate these features. To address this challenge, we proposed two innovative optimization architectures: Selective Fusion Mamba (SF-Mamba) and Multi-Scale Fusion Mamba (MF-Mamba). SF-Mamba can flexibly and dynamically adjust the fusion strategy of local and global features according to the characteristics of the lesions, effectively handling segmentation tasks with variable morphology. MF-Mamba enhances the model’s segmentation ability for lesions of different sizes by capturing global information across scales. Based on these two structures, we constructed a lightweight SMM-UNet model, which not only significantly reduces the computational burden (with only 0.038M parameters) but also demonstrates excellent generalization ability and can efficiently adapt to various types of medical images. Extensive tests on the ISIC2017, ISIC2018, and BUSI public datasets show that SMM-UNet achieves excellent segmentation performance with an extremely low parameter cost.

Single-Shot 3D Imaging Meta-Microscope.

Three-dimensional (3D) imaging enables high-precision and high-resolution axial positioning, which is crucial for biological imaging, semiconductor defect monitoring, and other applications. Conventional implementations rely on bulky optical elements or scanning mechanisms, resulting in low speed and complicated setups. Here, we generate the double-helix (DH) point spread function with an all-dielectric metasurface and thus innovate the 3D imaging microscope (hence dubbed meta-microscope), both in 4f and 2f imaging systems. The 4f-meta-microscope with a numerical aperture of 0.7 achieves an axial localization accuracy below 0.12 μm within a 15.47 μm detection range, while the 2f-DH meta-microscope with a numerical aperture of 0.3 shows a 1.12 μm accuracy within a 227.33 μm range. We also demonstrate single-shot and accurate 3D biological imaging of the mouse kidney tissue and peach anther, providing a comprehensive and efficient approach for 3D bioimaging and other applications through a single-shot 3D meta-microscope.

ADVANCEMENTS IN ALZHEIMER’S DIAGNOSIS THROUGH MRI USING BAYESIAN CONVOLUTIONAL NEURAL NETWORKS AND VARIATIONAL INFERENCE

Alzheimer’s disease is one of the brain disorders that can be deadly in older. The disease is less treated and less recognized, but Alzheimer’s disease is now a significant public health problem. Early detection of the disease can significantly reduce symptoms. However, the lack of medical personnel makes handling this disease complex. Therefore, an automatic diagnosis of Alzheimer’s disease is needed with a Magnetic Resonance Imaging (MRI) examination to get an accurate diagnosis of the disease. This study classified the type of Alzheimer’s disease with deep learning methods using the Bayesian Convolutional Neural Network (BCNN) and the Variational Inference (VI) technique. It aims to determine image classification and accuracy level at the level of Alzheimer’s disease by using 2,400 brain MRI images, divided into three classes (non-demented, very mild demented, and mild demented) based on severity. The data was acquired from the kaggle.com website. We use a dataset scenario of 80% for training and 20% for testing, 100x100 pixels, kernel size 3x3, and optimizer Adam with epoch 200. The accuracy of the image classification process is 80%. The non-demented label predicts that the uncertainty is 0.371, and the other uncertainty prediction is 0.002. The ability to anticipate uncertainty enables clinicians to make informed decisions regarding the reliability of the model’s output and the need for additional validation or confirmation.

Research on the fusion imaging method of sign coherence and time reversal for Lamb wave sparse array

Time-reversal imaging struggles to detect plate-like structures due to interference from Lamb wave mode conversion and the processing demands, leading to less effective outcomes. This paper proposes a sign coherence factor and time reversal fusion (SCF-TR) imaging method based on amplitude and phase estimation. This method removes the coherence of array signals during signal reversal and refocusing. It reintroduces the sign coherence component to reduce interference from non-target scattered waves and partially overcome the constraints imposed by the Rayleigh criterion. The method allows imaging at a resolution smaller than the wavelength of Lamb and enhances the quality of the resulting images. In addition, a sparse array design utilizing the White Shark Optimisation Algorithm (WSO) is proposed to streamline the SCF-TR calculation process. This design utilizes sparse full matrix data to improve imaging efficiency. The experimental results show that for single blind hole defects, the SCF-TR method improves the array performance metrics and signal-to-noise ratio by 22.46% and 42.50%, respectively, compared to the TR method. For multiple asymmetric blind hole defects, when the defect size exceeds the resolution threshold, SCF-TR accurately reflects the position and morphology of defects smaller than the wavelength. When the defect size is below the resolution threshold, SCF-TR achieves super-resolution imaging. The sparse array designed using the White Shark Optimization algorithm demonstrates good sidelobe characteristics, effectively reducing sidelobe noise without reducing the array aperture. Moreover, the SCF-TR imaging time is reduced by approximately half while maintaining imaging accuracy.

Deep learning-based detection of affected body parts in Parkinson’s disease and freezing of gait using time-series imaging

We proposed a deep learning method using a convolutional neural network on time-series (TS) images to detect and differentiate affected body parts in people with Parkinson’s disease (PD) and freezing of gait (FOG) during 360° turning tasks. The 360° turning task was performed by 90 participants (60 people with PD [30 freezers and 30 nonfreezers] and 30 age-matched older adults (controls) at their preferred speed. The position and acceleration underwent preprocessing. The analysis was expanded from temporal to visual data using TS imaging methods. According to the PD vs. controls classification, the right lower third of the lateral shank (RTIB) on the least affected side (LAS) and the right calcaneus (RHEE) on the LAS were the most relevant body segments in the position and acceleration TS images. The RHEE marker exhibited the highest accuracy in the acceleration TS images. The identified markers for the classification of freezers vs. nonfreezers vs. controls were the left lateral humeral epicondyle (LELB) on the more affected side and the left posterior superior iliac spine (LPSI). The LPSI marker in the acceleration TS images displayed the highest accuracy. This approach could be a useful supplementary tool for determining PD severity and FOG.

Application Exploration of the Photon Chip Technology in Real time Processing of Biomedical Images

In the biomedical field, photonic chip technology is gradually becoming the key to improving the speed and accuracy of medical imaging. The development of this technology has given us new hope in early disease diagnosis and cell monitoring. Recent research has not only sorted out the hotspots and future trends of biomedical photonics, but also paid special attention to the application of photon technology in these fields. Specifically, photonic chips have demonstrated their unique capabilities in multi-color two-photon excitation fluorescence microscopy imaging, two-photon fluorescence lifetime imaging, and two-photon fiber endoscopic imaging. These applications not only improve the clarity of imaging, but also accelerate processing speed, which is crucial for rapid diagnosis and treatment. At the end of the article, the author summarized the contributions of photonic chip technology in the field of biomedical imaging and provided prospects for future development directions.

Quantitative phase imaging verification in large field-of-view lensless holographic microscopy via two-photon 3D printing

Large field-of-view (FOV) microscopic imaging (over 100 mm2) with high lateral resolution (1–2 μm) plays a pivotal role in biomedicine and biophotonics, especially within the label-free regime. Lensless digital holographic microscopy (LDHM) is promising in this context but ensuring accurate quantitative phase imaging (QPI) in large FOV LDHM is challenging. While phantoms, 3D printed by two-photon polymerization (TPP), have facilitated testing small FOV lens-based QPI systems, an equivalent evaluation for lensless techniques remains elusive, compounded by issues such as twin-image and beam distortions, particularly towards the detector’s edges. Here, we propose an application of TPP over large area to examine phase consistency in LDHM. Our research involves fabricating widefield phase test targets with galvo and piezo scanning, scrutinizing them under single-shot twin-image corrupted conditions and multi-frame iterative twin-image minimization scenarios. By measuring the structures near the detector’s edges, we verified LDHM phase imaging errors across the entire FOV, with less than 12% phase value difference between areas. Our findings indicate that TPP, followed by LDHM and Linnik interferometry cross-verification, requires new design considerations for precise large-area photonic manufacturing. This research paves the way for quantitative benchmarking of large FOV lensless phase imaging, enhancing understanding and further development of LDHM technique.

Observation of dynamic position and morphological changes of temporomandibular joint discs under Angle's classification

Objective: To observe the dynamic changes of the temporomandibular joint (TMJ) disc in joint movement under different Angle's classification, providing reference for understanding joint functional movement and providing a basis for more accurate clinical imaging diagnosis. Methods: A total of 30 patients (13 males and 17 females) with temporomandibular disorders who were admitted to Beijing Friendship Hospital, Capital Medical University and General Hospital of the People's Liberation Army from January 2022 to April 2024 were enrolled. Thirty adults (13 males and 17 females) with different Angle's classification, with an average age of (34.4±8.5) years, were subjected to dynamic imaging of their TMJ from the closed position to the maximum opening position, and then to the closed position using MRI. The position and morphological changes of the articular discs were observed. Results: The results showed that volunteers with no displacement of the articular disc in class Ⅰ, Ⅱ, and Ⅲ relationships had different shapes of the articular disc during open and closed mouth movements. However, in the maximum opening position, the articular disc were all located directly below the maxillary nodules, and their shape is double concave. In terms of irreversible anterior displacement of the articular disc, in class Ⅰ Angle, the posterior zone of the disc contacts the anterior inclined plane of condyle from the maximum opening position back to the front of the closing position. In class Ⅱ, the posterior zone of the disc contacts the anterior inclined plane of condyle from the beginning of opening position to maximum opening position. In class Ⅲ, the posterior zone of the disc is always in contact with the anterior inclined plane of condyle throughout the entire movement process. And among them, the articular disc presents a forward displacement state at the closing position, its morphology undergoes folding phenomenon. When the openness is 2.5 cm, the articular disc moves up to a certain extent, and is closer to the anterior inclined plane of condyle, and its shape is also partially changed. When the openness is 4.3 cm, the shape of the articular disc, located between the anterior inclined plane of the joint node and the posterior inclined plane of the condyle, is typical double concave, which is sufficient to show that the articular disc is reversible when maximum opening position is reached. In terms of reversible anterior disc displacement, in class Ⅰ Angle, the posterior zone of the disc contact with the anterior inclined plane of condyle at the beginning of the opening position and the end of the closing position. In classⅡ Angle, the posterior zone of the articular disc is not in contact with the anterior inclined plane of condyle. In class Ⅲ Angle, the posterior zone of the articular disc contact with the anterior inclined plane of condyle at the end of the closing position. Conclusions: Multi level dynamic MR imaging data of the temporomandibular joint can dynamically observe the movement of the temporomandibular joint, intuitively and accurately display the position and shape of the articular disc during movement, and can serve as a useful supplement to static conventional MR imaging of the TMJ. The patient's TMJ needs to reach the maximum opening position in order to determine whether the joint disc displacement can be reversible or not.

In vivo Multi-perspective 3D + t Ultrasound Imaging and Motion Estimation of Abdominal Aortic Aneurysms.

Time-resolved three-dimensional ultrasound (3D + t US) is a promising imaging modality for monitoring abdominal aortic aneurysms (AAAs), providing their 3D geometry and motion. The lateral contrast of US is poor, a well-documented drawback which multi-perspective (MP) imaging could resolve. This study aims to show the feasibility of in vivo multi-perspective 3D + t ultrasound imaging of AAAs for improving the image contrast and displacement accuracy. To achieve this, single-perspective (SP) aortic ultrasound images from three different angles were spatiotemporally registered and fused, and the displacements were compounded. The fused MP had a significantly higher wall-lumen contrast than the SP images, for both patients and volunteers (P < .001). MP radial displacements patterns are smoother than SP patterns in 67% of volunteers and 92% of patients. The MP images from three angles have a decreased tracking error (P < .001 for all participants), and an improved SNRe compared to two out of three SP images (P < .05). This study has shown the added value of MP 3D + t US, improving both image contrast and displacement accuracy in AAA imaging. This is a step toward using multiple or large transducers in the clinic to capture the 3D geometry and strain more accurately, for patient-specific characterization of AAAs.

Comprehensive Review on Bubbles: Synthesis, Modification, Characterization and Biomedical Applications.

Accurate detection, treatment, and imaging of diseases are important for effective treatment outcomes in patients. In this regard, bubbles have gained much attention, due to their versatility. Bubbles usually 1 nm to 10 μm in size can be produced and loaded with a variety of lipids, polymers, proteins, and therapeutic and imaging agents. This review details the different production and loading methods for bubbles, for imaging and treatment of diseases/conditions such as cancer, tumor angiogenesis, thrombosis, and inflammation. Bubbles can also be used for perfusion measurements, important for diagnostic and therapeutic decision making in cardiac disease. The different factors important in the stability of bubbles and the different techniques for characterizing their physical and chemical properties are explained, for developing bubbles with advanced therapeutic and imaging features. Hence, the review provides important insights for researchers studying bubbles for biomedical applications.

Deep-Learning-Based Real-Time Passive Non-Line-of-Sight Imaging for Room-Scale Scenes.

Non-line-of-sight imaging is a technique for reconstructing scenes behind obstacles. We report a real-time passive non-line-of-sight (NLOS) imaging method for room-scale hidden scenes, which can be applied to smart home security monitoring sensing systems and indoor fast fuzzy navigation and positioning under the premise of protecting privacy. An unseen scene encoding enhancement network (USEEN) for hidden scene reconstruction is proposed, which is a convolutional neural network designed for NLOS imaging. The network is robust to ambient light interference conditions on diffuse reflective surfaces and maintains a fast reconstruction speed of 12.2 milliseconds per estimation. The consistency of the mean square error (MSE) is verified, and the peak signal-to-noise ratio (PSNR) values of 19.21 dB, 15.86 dB, and 13.62 dB are obtained for the training, validation, and test datasets, respectively. The average values of the structural similarity index (SSIM) are 0.83, 0.68, and 0.59, respectively, and are compared and discussed with the corresponding indicators of the other two models. The sensing system built using this method will show application potential in many fields that require accurate and real-time NLOS imaging, especially smart home security systems in room-scale scenes.

Novel Colorimetric and Near-Infrared Ratiometric Fluorescent Probe for Sensing Cysteine in Food Samples, Plants, and Living Cells.

Cysteine (Cys) is a crucial biothiol that acts a significant function in food samples and biological systems, including plant roots and living cells. Hence, we developed a novel colorimetric and near-infrared ratiometric fluorescent probe (CT), composed of coumarin and tetrahydroacridine-conjugated indole salt, for the detection of Cys. Upon reaction with Cys, the probe undergoes a specific N-substitution reaction, resulting in a notable colorimetric change and a significant ratiometric fluorescent response in both visible and near-infrared emission channels. These dual-channel ratiometric fluorescence changes are completely independent, enabling the probe to obtain great selectivity, sensitivity, and exceptional detection accuracy. Leveraging these attributes, the probe was employed to provide accurate quantitative analysis of Cys in food samples. Furthermore, confocal imaging demonstrated that the probe could monitor both exogenous and endogenous Cys levels in living cells and track Cys changes in plant roots under heavy metal stress. This work presents a dependable and accurate imaging solution for tracking and identifying Cys of real food, plants, and living cells.