Accurate Image Research Articles

Enhancing Cardiac Imaging with Deep Learning: New Frontiers in Diagnosis

The incorporation of Deep Learning (DL) in imaging represents a significant leap forward in medical diagnostics transforming the approach to identifying, diagnosing, and treating cardiovascular diseases. By utilizing algorithms and extensive datasets DL greatly enhances the precision, speed, and predictive capabilities of diagnostic methods like echocardiography, magnetic resonance imaging (MRI) and computed tomography (CT). This piece explores the influence of DL on cardiac imaging by illustrating how these technologies not only enhance image clarity and accuracy but also streamline and improve the analysis of intricate imaging data. The integration of DL enables detection of cardiac conditions with greater accuracy enabling tailored treatment strategies for individual patients and potentially saving lives. Furthermore, automating procedures reduces the risk of human errors and speeds up decision making processes ultimately enhancing patient outcomes and operational efficacy in healthcare environments. As we delve into the state and future prospects of DL in cardiac imaging, we uncover broader implications for predictive medicine and healthcare analytics that suggest a promising future, for technology driven healthcare solutions that are both groundbreaking and beneficial. Keywords: Deep Learning, Cardiac Imaging, Echocardiography, Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Diagnostic Automation, Predictive Medicine, Healthcare Analytics

Rational-Function-Model-Based Rigorous Bundle Adjustment for Improving the Relative Geometric Positioning Accuracy of Multiple Korea Multi-Purpose Satellite-3A Images

Rational-Function-Model-Based Rigorous Bundle Adjustment for Improving the Relative Geometric Positioning Accuracy of Multiple Korea Multi-Purpose Satellite-3A Images

Geometric calibration of large-scale SAR images using wind turbines as ground control points

ABSTRACT The geometric positioning accuracy of synthetic aperture radar images is a key factor impacting their application, necessitating geometric calibration for improved accuracy. Traditional geometric calibration methods rely on ground calibration fields and supportive equipment such as corner reflectors, which often fall short in meeting both domestic and international geometric calibration demands. To enable convenient, swift, and large-scale or even global SAR calibration, we proposed a method that employs wind turbines as ground control points for SAR geometric calibration, and have constructed a wind turbine database for China. This database allows for precise positioning of wind turbine targets in SAR images. In this study, we applied the wind turbine geometric calibration model for the first time to four imaging modes of Gaofen-3 images. The results indicate that the positioning accuracy post-calibration ranges from 2.70 m to 8.72 m. The mean square error of positioning for the multi-scene calibration of 12 images spanning six provinces is less than 9 m, validating the method's applicability for large-scale calibration. Our proposed method presents a cost-effective solution for large-scale geometric calibration and can notably enhance the positioning accuracy of Gaofen-3 images, thereby increasing their application potential in remote sensing mapping, environmental monitoring, and resource management.

Stability Detection of Canopy RGB Images for Different Underlying Surfaces Based on SVM

This study aims to investigate the impact of different environmental conditions on the stability of RGB images in ecological sites and the anti-interference properties of vegetation images on different underlying surfaces. Three vegetation types including sugarcane, forest, and karst (mainly shrub and grass) are used to segment green vegetation using machine learning, and the RGB vegetation indices are calculated using color channel data. Then, The effect of weather, season, and time period on different types of vegetation indices are studied, which provide technical references for quantitative application of RGB image data. The results indicate the following: ① For the vegetation with high canopy density, the SVM machine learning segmentation algorithm used in this study is more applicable, as the RGB image segmentation accuracy of sugarcane and forest is significantly higher than karst. For different weather conditions, the segmentation accuracy of sugarcane and forest RGB images on sunny or cloudy days is higher than that on rainy or foggy days, but the effect on sparse vegetation in karst is not obvious. Additionally, the segmentation accuracy of different vegetation types has a small increase with NLM filter processing. ② Change in weather, season, and time can affect the stability of the image index. For different weather conditions, the vegetation indices of sugarcane and forest images on sunny or cloudy days are the more stable (ExGR, outlier proportion between 3.25% and 5.63%), while on rainy and foggy days they are less stable (ExR, outlier proportion between 17.60% and 21.59%). For different seasons, the stability of the sugarcane and karst image index obtained in spring is better (ExG, outlier proportion between 3.32% and 6.88%), while the stability of the forest image index obtained in summer is better (ExR, outlier proportion = 10.32%). For different times within a day, the sugarcane image index obtained in the morning is more stable (ExGR, outlier proportion = 4.62%), while the stability of the forest image index obtained in the afternoon is better (ExGR, outlier proportion = 9.24%). ③ The stability of the sugarcane image index is more affected by weather and season. For forest, the influence of the weather is more obvious than the season. But, for karst, the effect of season on the vegetation index is greater than that of weather.

Osteochondritis Dissecans of the Knee: A Case Report

Osteochondritis dissecans (OCD) is a focal, idiopathic alteration of subchondral bone with the potential for instability and disruption of adjacent articular cartilage causing pain or discomfort in the knee and, if left untreated, may cause increasing disability and progressive arthritic change of the joint. This article presents a case study of a 16-year-old female who presented with intermittent right lateral knee pain for several years with no acute traumatic injury. Diagnostic imaging demonstrated subchondral changes of the lateral femoral condyle consistent with osteochondritis dissecans. The patient was diagnosed with an unstable OCD of the right knee and underwent surgical treatment. Following treatment, the patient attended formal physical therapy two times per week and began a gradual return to sport protocol at 4 months post-operatively. This case study will demonstrate the complexities surrounding the clinical presentation, diagnostic workup, and treatment options for this challenging diagnosis. The importance of accurate imaging and aggressive treatment of unstable OCD lesions will be highlighted.

Prediction method of long-term creep deformation of mobile phone resin lens module in a harsh temp environment

In recent years, the replacement cycle of digital products such as mobile phones in the market has significantly increased, and the accurate prediction of attenuation under long-term usage conditions of the lens module, storage, and other components has become important. The vast majority of mobile phone lens modules are processed from optical resins, and the degradation of their imaging performance is mainly caused by assembly deformation and long-term creep deformation after a high-temperature reliability test. Therefore, there is an urgent need for a reliable model to describe the long-term creep deformation of the mobile phone resin lens module in a harsh temperature environment. This study proposed a modified derivation method for the parameters of the Maxwell constitutive model to predict the long-term creep deformation of the mobile phone resin lens module in a high-temperature environment. The fitting accuracy of the modified generalized Maxwell model and the time-hardening model were compared through uniaxial compression creep experiments. Long-term creep simulations and experiments were conducted on the mobile phone resin lens module; the deformation law of each component of the mobile phone resin lens module was identified; and the creep deformation mechanism was elucidated. The result shows that the modified generalized Maxwell model has much higher accuracy than the time-hardening model in predicting the long-term creep deformation of different resin materials. When considering the influence of the creep phenomenon, the simulation result matches well with the experimental result, and the high-temperature creep mechanism of the mobile phone resin lens module was clarified. In addition, it was found that the maximum deformation of the lens occurred at the lens with the largest size, which was 2.1 µm, and the position with the largest surface deviation was at the edge of the lens. The prediction method proposed in this study can provide guidance and correction suggestions for the structural design of a mobile phone resin lens module and the rationality of optical design; thus, the imaging accuracy of lens modules will be guaranteed to the greatest extent possible.

Automated Detection of Plant Diseases and Pests Using Advanced Image Processing and Neural Networks

Pests and viral plant diseases have proliferated in agriculture. Rats spread plant diseases. Hard to detect viral infections or insect infestations that impair plant growth and yield. It decreases crop size, quality, and marketability, incurring major economic losses. Strong, disease-resistant crops are needed for agriculture. Bacterial, fungal, and insect/pest diseases destroy 70% and 15% of crops. Above 85% of disease severity destroys crops. Early detection systems track pests and diseases. High-definition equipment is costly for rural farmers. Correct diagnosis permits quick infection severity and type assessment and control. In precision agriculture, neural networks and image processing eliminate human error. Sustainable agriculture and organic agricultural cultivation must start early. The proposed system classifies diseases and pests for plant safety. A healthy, fast-growing plantation is our goal. Early detection of viral diseases and insect infestations boosts crop productivity. Plant disease diagnostics focuses on leaves and buds. Stem and insect infections are covered. The suggested method classifies plant health and disease automatically. Peanut leaves develop Bacterial Blight from contagious illnesses like Cercospora Leaf Peanut. Boll Weevils, European Corn Borers, and Fall Army Worms attack cotton. The conditions are categorised. Image accuracy depends on lighting, resolution, location, and backdrop complexity. Various low-resolution pictures are enriched to better classification. Complex leaf intensity variations cause the system to misdiagnose illness or insect infestation. To avoid misclassification, the system segments the sample image using various methods to find lesions. Multilevel Segmentation (MS) improves Deep neural network feature extraction using approximate outlines. This method protects important pixels, eliminates unneeded pixels, and amplifies the lesion to control segmentation and reduce insufficient and excessive segmentation.

Effect of single-photon emission computed tomography acquisition method and sampling angles on image quality and quantitative accuracy in xSPECT-reconstructed images.

The aim of this study was to evaluate the effects of the single-photon emission computed tomography (SPECT) acquisition method and sampling angles on the qualitative and quantitative interpretations of xSPECT-reconstructed images. The spatial resolution was evaluated using a JSP phantom, and the uniformity and quantitative accuracy were verified with a NEMA IEC Body Phantom using an SIEMENS Symbia Intevo SPECT/computed tomography system. SPECT was performed using three acquisition methods (step-and-shoot, continuous, and acquire during the step), and the sampling angles were set to 2, 3, 4, 5, and 6°. The xSPECT-reconstruction technology which is used with ordered subset-conjugated gradient minimization was used for image reconstruction. Full width of half maximum, an evaluation index of spatial resolution, varied up to 2.73 mm with different sampling angles and up to 2.06 mm with different acquisition methods. Uniformity, as assessed by the coefficient of variation, improved with increasing sampling angles. The accuracy of the quantification of the hot sphere showed an error rate of approximately 10% depending on the sampling angle, and an error rate of approximately 5% depending on the different acquisition methods. In xSPECT-reconstructed images, the difference in sampling angle has a greater impact on image quality and quantitativity than the difference in the acquisition method. For tests in which uniformity is important, a larger sampling angle is recommended.

Tumor spheroid elasticity estimation using mechano-microscopy combined with a conditional generative adversarial network

Background and ObjectivesTechniques for imaging the mechanical properties of cells are needed to study how cell mechanics influence cell function and disease progression. Mechano-microscopy (a high-resolution variant of compression optical coherence elastography) generates elasticity images of a sample undergoing compression from the phase difference between optical coherence microscopy (OCM) B-scans. However, the existing mechano-microscopy signal processing chain (referred to as the algebraic method) assumes the sample stress is uniaxial and axially uniform, such that violation of these assumptions reduces the accuracy and precision of elasticity images. Furthermore, it does not account for prior information regarding the sample geometry or mechanical property distribution. In this study, we investigate the feasibility of training a conditional generative adversarial network (cGAN) to generate elasticity images from phase difference images of samples containing a cell spheroid embedded in a hydrogel. MethodsTo construct the cGAN training and simulated test sets, we generated 30,000 artificial elasticity images using a parametric model and computed the corresponding phase difference images using finite element analysis to simulate compression applied to the artificial samples. We also imaged real MCF7 breast tumor spheroids embedded in hydrogel using mechano-microscopy to construct the experimental test set and evaluated the cGAN using the algebraic elasticity images and co-registered OCM and confocal fluorescence microscopy (CFM) images. ResultsComparison with the simulated test set ground truth elasticity images shows the cGAN produces a lower root mean square error (median: 3.47 kPa, 95 % confidence interval (CI) [3.41, 3.52]) than the algebraic method (median: 4.91 kPa, 95 % CI [4.85, 4.97]). For the experimental test set, the cGAN elasticity images contain features resembling stiff nuclei at locations corresponding to nuclei seen in the algebraic elasticity, OCM, and CFM images. Furthermore, the cGAN elasticity images are higher resolution and more robust to noise than the algebraic elasticity images. ConclusionsThe cGAN elasticity images exhibit better accuracy, spatial resolution, sensitivity, and robustness to noise than the algebraic elasticity images for both simulated and real experimental data.

Role of Cine-Magnetic Resonance Imaging in the Assessment of Mediastinal Masses with Uncertain/Equivocal Findings from Pre-Operative Computed Tomography Scanning.

Malignant neoplasms originating from or involving the mediastinum represent a diagnostic and therapeutic challenge when they are in contact with nearby cardiovascular structures. We aimed to test the diagnostic accuracy of cine-magnetic resonance imaging (cine-MRI) in detecting the infiltration of cardiovascular structures in cases with uncertain or equivocal findings from contrast-enhanced Computed Tomography (CT) scanning. Fifty patients affected by tumors with a suspected invasion of mediastinal cardiovascular structures at the pre-operative chest CT scan stage underwent cine-MRI before surgery at our Institution. Intraoperative findings and the histological post-surgical report were used as a reference standard to define infiltration. Inter- and intra-observer agreement for CT scans and cine-MRI were also computed over a homogenous sample of 14 patients. Cine-MRI had a higher negative predictive value (93% vs. 54%, p < 0.001) than CT scans, higher sensitivity (91% vs. 16%, p < 0.001), as well as greater accuracy (66% vs. 50%, p < 0.001) in detecting cardiovascular invasion. Cine-MRI also showed better inter- and intra-observer agreement for infiltration detection. Cine-MRI outperforms conventional contrast-enhanced chest CT scans in the preoperative assessment of cardiovascular infiltration by mediastinal or pulmonary tumors, making it a useful imaging modality in the preoperative staging and evaluation of patients with equivocal findings at the chest CT scan stage.

Towards High-Quality MRI Reconstruction With Anisotropic Diffusion-Assisted Generative Adversarial Networks And Its Multi-Modal Images Extension.

Recently, fast Magnetic Resonance Imaging reconstruction technology has emerged as a promising way to improve the clinical diagnostic experience by significantly reducing scan times. While existing studies have used Generative Adversarial Networks to achieve impressive results in reconstructing MR images, they still suffer from challenges such as blurred zones/boundaries and abnormal spots caused by inevitable noise in the reconstruction process. To this end, we propose a novel deep framework termed Anisotropic Diffusion-Assisted Generative Adversarial Networks, which aims to maximally preserve valid high-frequency information and structural details while minimizing noises in reconstructed images by optimizing a joint loss function in a unified framework. In doing so, it enables more authentic and accurate MR image generation. To specifically handle unforeseeable noises, an Anisotropic Diffused Reconstruction Module is developed and added aside the backbone network as a denoise assistant, which improves the final image quality by minimizing reconstruction losses between targets and iteratively denoised generative outputs with no extra computational complexity during the testing phase. To make the most of valuable MRI data, we extend its application to support multi-modal learning to boost reconstructed image quality by aggregating more valid information from images of diverse modalities. Extensive experiments on public datasets show that the proposed framework can achieve superior performance in polishing up the quality of reconstructed MR images. For example, the proposed method obtains average PSNR and mSSIM values of 35.785dB and 0.9765 on the MRNet dataset, which are at least about 2.9dB and 0.07 higher than those from the baselines.

A Review of Medical Image Registration for Different Modalities.

Medical image registration has become pivotal in recent years with the integration of various imaging modalities like X-ray, ultrasound, MRI, and CT scans, enabling comprehensive analysis and diagnosis of biological structures. This paper provides a comprehensive review of registration techniques for medical images, with an in-depth focus on 2D-2D image registration methods. While 3D registration is briefly touched upon, the primary emphasis remains on 2D techniques and their applications. This review covers registration techniques for diverse modalities, including unimodal, multimodal, interpatient, and intra-patient. The paper explores the challenges encountered in medical image registration, including geometric distortion, differences in image properties, outliers, and optimization convergence, and discusses their impact on registration accuracy and reliability. Strategies for addressing these challenges are highlighted, emphasizing the need for continual innovation and refinement of techniques to enhance the accuracy and reliability of medical image registration systems. The paper concludes by emphasizing the importance of accurate medical image registration in improving diagnosis.

Abstract We016: Generative Artificial Intelligence for hPSC-derived Cardiac Organoid Florescence Generation

Background: Human pluripotent stem cell (hPSC)-derived cardiac organoids (COs) are extensively applied in modeling heart development and disease. COs are daily monitored by bright-field microscopic imaging for tracking organoid differentiation, morphology, and size, however, it does not provide important information on cell distribution and network in COs. Then, fluorescence microscopic imaging is required with extra steps of experiments when multiple hPSC lines are required. Recently, generative artificial intelligence (AI) has been increasingly applied for image colorization, which extends the possibility of colorizing COs with fluorescence from phase/contrast images. While the phase/contrast image is routinely applied and conveniently used in every biomedical lab daily, is it possible to generate accurate and useful fluorescence information or image colorization from the phase/contrast images of cardiac organoids under the same differentiation protocol is crucial for a broader characterization and analysis of hPSC-derived cardiac organoids and further applications? Hypothesis: We hypothesize that generative AI will generate accurate fluorescence images from phase/contrast images of COs. Approach: 1300 images from COs (differentiated from one reporter hPSC line) were used as the training dataset for the Pix2Pix Conditional Generative Adversarial Networks (GANs) model ( Fig.1a ) with the Convolution Block Attention Module (CBAM) in the U-Net generator ( Fig.1b ). Three models were designed: U-Net (Model 1), U-Net generator with CBAM (Model 2), and U-Net with CBAM and Generator Iteration (Model 3). Three different metrics were included: Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Weighted Patch Histogram to evaluate the prediction outcomes. Results: The predicted fluorescence images from phase/contrast images of different COs are very similar to the ground truths ( Fig.1c ). The similarities evaluated by all three metrics ( Fig.1d ) have the scores of PSNR over 30, SSIM over 0.92, and a Weighted Patch Histogram score over 0.75, which are the most accurate and similar to the ground truth. Conclusions: A generative AI model was established to colorize phase images of COs using cGANs and CBAM, which capture intricate fluorescence details to support the biomedical research of COs with enhanced interpretability and analysis of cardiac organoid morphology and structure in biomedical research and applications.

The Feasibility of a Model-Based Iterative Reconstruction Technique Tuned for the Myocardium on Myocardial Computed Tomography Late Enhancement.

This study evaluated the feasibility of a model-based iterative reconstruction technique (MBIR) tuned for the myocardium on myocardial computed tomography late enhancement (CT-LE). Twenty-eight patients who underwent myocardial CT-LE and late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) within 1 year were retrospectively enrolled. Myocardial CT-LE was performed using a 320-row CT with low tube voltage (80 kVp). Myocardial CT-LE images were scanned 7 min after CT angiography (CTA) without additional contrast medium. All myocardial CT-LE images were reconstructed with hybrid iterative reconstruction (HIR), conventional MBIR (MBIR_cardiac), and new MBIR tuned for the myocardium (MBIR_myo). Qualitative (5-grade scale) scores and quantitative parameters (signal-to-noise ratio [SNR] and contrast-to-noise ratio [CNR]) were assessed as image quality. The sensitivity, specificity, and accuracy of myocardial CT-LE were evaluated at the segment level using an American Heart Association (AHA) 16-segment model, with LGE-MRI as a reference standard. These results were compared among the different CT image reconstructions. In 28 patients with 448 segments, 160 segments were diagnosed with positive by LGE-MRI. In the qualitative assessment of myocardial CT-LE, the mean image quality scores were 2.9 ± 1.2 for HIR, 3.0 ± 1.1 for MBIR_cardiac, and 4.0 ± 1.0 for MBIR_myo. MBIR_myo showed a significantly higher score than HIR (P < 0.001) and MBIR_cardiac (P = 0.018). In the quantitative image quality assessment of myocardial CT-LE, the median image SNR was 10.3 (9.1-11.1) for HIR, 10.8 (9.8-12.1) for MBIR_cardiac, and 16.8 (15.7-18.4) for MBIR_myo. The median image CNR was 3.7 (3.0-4.6) for HIR, 3.8 (3.2-5.1) for MBIR_cardiac, and 6.4 (5.0-7.7) for MBIR_myo. MBIR_myo significantly improved the SNR and CNR of CT-LE compared to HIR and MBIR_cardiac (P < 0.001). The sensitivity, specificity, and accuracy for the detection of myocardial CT-LE were 70%, 92%, and 84% for HIR; 71%, 92%, and 85% for MBIR_cardiac; and 84%, 92%, and 89% for MBIR_myo, respectively. MBIR_myo showed significantly higher image quality, sensitivity, and accuracy than the others (P < 0.05). MBIR tuned for myocardium improved image quality and diagnostic performance for myocardial CT-LE assessment.

Van der Waals mid-wavelength infrared detector linear array for room temperature passive imaging.

Passive imaging for mid-wave infrared (MWIR) is resistant to atmospheric pollutants, guaranteeing image clarity and accuracy. Arrayed photodetectors can simultaneously perform radiation sensing to improve efficiency. Room temperature van der Waals (vdWs) photodetectors without lattice matching have evolved rapidly with optimized stacking methods, primarily for single-pixel devices. The urgent need to implement arrayed devices aligns with practical demands. Here, we present an 8 by 1 black phosphorus/molybdenum sulfide (BP/MoS2) vdWs photodetector linear array with a fill-factor of ~77%, fabricated using a temperature-assisted sloping transfer method. The flat interface and uniform thickness facilitate carrier transport and minimize pixel nonuniformities, showing an average peak detectivity (D*) of 2.34 × 109 cm·Hz1/2·W-1 in the mid-wave infrared region. Compared to a single pixel, push-broom scanning passive imaging is eight times more efficient and further enhanced through mean filtering and fast Fourier transform filtering for strip noise correction. Our study offers guidance on vdWs arrayed devices for engineering applications.

Ultrasound-Compatible Electrode for Functional Electrical Stimulation.

Functional electrical stimulation (FES) is a vital method in neurorehabilitation used to reanimate paralyzed muscles, enhance the size and strength of atrophied muscles, and reduce spasticity. FES often leads to increased muscle fatigue, necessitating careful monitoring of the patient's response. Ultrasound (US) imaging has been utilized to provide valuable insights into FES-induced fatigue by assessing changes in muscle thickness, stiffness, and strain. Current commercial FES electrodes lack sufficient US transparency, hindering the observation of muscle activity beneath the skin where the electrodes are placed. US-compatible electrodes are essential for accurate imaging and optimal FES performance, especially given the spatial constraints of conventional US probes and the need to monitor muscle areas directly beneath the electrodes. This study introduces specially designed body-conforming US-compatible FES (US-FES) electrodes constructed with a silver nanowire/polydimethylsiloxane (AgNW/PDMS) composite. We compared the performance of our body-conforming US-FES electrode with a commercial hydrogel electrode. The findings revealed that our US-FES electrode exhibited comparable conductivity and performance to the commercial one. Furthermore, US compatibility was investigated through phantom and in vivo tests, showing significant compatibility even during FES, unlike the commercial electrode. The results indicated that US-FES electrodes hold significant promise for the real-time monitoring of muscle activity during FES in clinical rehabilitative applications.

SSCFormer: Revisiting ConvNet-Transformer Hybrid Framework From Scale-Wise and Spatial-Channel-Aware Perspectives for Volumetric Medical Image Segmentation.

Accurate and robust medical image segmentation is crucial for assisting disease diagnosis, making treatment plan, and monitoring disease progression. Adaptive to different scale variations and regions of interest is essential for high accuracy in automatic segmentation methods. Existing methods based on the U-shaped architecture respectively tackling intra- and inter-scale problem with a hierarchical encoder, however, are restricted by the scope of multi-scale modeling. In addition, global attention and scaling attention in regions of interest have not been appropriately adopted, especially for the salient features. To address these two issues, we propose a ConvNet-Transformer hybrid framework named SSCFormer for accurate and versatile medical image segmentation. The intra-scale ResInception and inter-scale transformer bridge are designed to collaboratively capture the intra- and inter-scale features, facilitating the interaction of small-scale disparity information at a single stage with large-scale from multiple stages. Global attention and scaling attention are cleverly integrated from a spatial-channel-aware perspective. The proposed SSCFormer is tested on four different medical image segmentation tasks. Comprehensive experimental results show that SSCFormer outperforms the current state-of-the-art methods.

A Convolutional Neural Network with Multifrequency and Structural Similarity Loss Functions for Electromagnetic Imaging.

In this paper, artificial intelligence (AI) technology is applied to the electromagnetic imaging of anisotropic objects. Advances in magnetic anomaly sensing systems and electromagnetic imaging use electromagnetic principles to detect and characterize subsurface or hidden objects. We use measured multifrequency scattered fields to calculate the initial dielectric constant distribution of anisotropic objects through the backpropagation scheme (BPS). Later, the estimated multifrequency permittivity distribution is input to a convolutional neural network (CNN) for the adaptive moment estimation (ADAM) method to reconstruct a more accurate image. In the meantime, we also improve the definition of loss function in the CNN. Numerical results show that the improved loss function unifying the structural similarity index measure (SSIM) and root mean square error (RMSE) can effectively enhance image quality. In our simulation environment, noise interference is considered for both TE (transverse electric) and TM (transverse magnetic) waves to reconstruct anisotropic scatterers. Lastly, we conclude that multifrequency reconstructions are more stable and precise than single-frequency reconstructions.

A deep learning model to assist visually impaired in pothole detection using computer vision

Visually impaired individuals encounter numerous impediments when traveling, such as navigating unfamiliar routes, accessing information, and transportation, which can limit their mobility and restrict their access to opportunities. However, assistive technologies and infrastructure solutions such as tactile paving, audio cues, voice announcements, and smartphone applications have been developed to mitigate these challenges. Visually impaired individuals also face difficulties when encountering potholes while traveling. Potholes can pose a significant safety hazard, as they can cause individuals to trip and fall, potentially leading to injury. For visually impaired individuals, identifying and avoiding potholes can be particularly challenging. The solutions ensure that all individuals can travel safely and independently, regardless of their visual abilities. An innovative approach that leverages the You Only Look Once (YOLO) algorithm to detect potholes and provide auditory or haptic feedback to visually impaired individuals has been proposed in this paper. The dataset of pothole images was trained and integrated into an application for detecting potholes in real-time image data using a camera. The app provides feedback to the user, allowing them to navigate potholes and increasing their mobility and safety. This approach highlights the potential of YOLO for pothole detection and provides a valuable tool for visually impaired individuals. According to the testing, the model achieved 82.7% image accuracy and 30 Frames Per Second (FPS) accuracy in live video. The model is trained to detect potholes close to the user, but it may be hard to detect potholes far away from the user. The current model is only trained to detect potholes, but visually impaired people face other challenges. The proposed technology is a portable option for visually impaired people.

A Hessian-Based Technique for Specular Reflection Detection and Inpainting in Colonoscopy Images.

In the field of Computer-Aided Detection (CADx), the use of AI-based algorithms for disease detection in endoscopy images, especially colonoscopy images, is on the rise. However, these algorithms often encounter performance issues due to obstructions like specular reflection, resulting in false positives. This paper presents a novel algorithm specifically designed to tackle the challenges posed by high specular reflection regions in colonoscopy images. The proposed algorithm identifies these regions and applies precise inpainting for restoration. The process entails converting the input image from RGB to HSV color space and focusing on the Saturation (S) component in convex regions detected using a Hessian-based method. This step creates a binary mask that pinpoints areas of specular reflection. The inpainting function then uses this mask to guide the restoration of these identified regions and their borders. To ensure a seamless blend of the restored regions with the background and adjacent pixels, a feathering process is applied to the repaired regions. This enhances both the accuracy and aesthetic coherence of the inpainted images. The performance of our algorithm was rigorously tested on five unique colonoscopy datasets and various endoscopy images from the Kvasir dataset, using an extensive set of evaluation metrics and a comparative analysis with existing methods consistently highlighted the superior performance of our algorithm.