Application of Self2Self self-supervised denoising framework to airborne gamma-ray spectrometry data.

Boolean and fuzzy logic operators and multivariate linear regression applied to airborne gamma-ray spectrometry data for regolith mapping in granite-greenstone terrain in Midwest Brazil

Hydrothermal alteration processes are connected to many mineral formations, particularly auriferous deposits. In this study, airborne gamma-ray spectrometry (GRS) data and the analysis of radioactive materials (eU, eTh, and K) are applied to search for regions with hydrothermal alteration activity. An example is presented from Wadi Al-Allaqi, South Eastern Desert, Egypt. GRS was used to analyse various radiometric data to address potential mineral deposit areas, to map regions potentially showing metallic ore mineralisation, and to point out new geological mineral resources. The Kd (potassium deviation), “F” parameter and Th-normalisation of the K and eU anomalies were calculated, and locating new exploratory targets in the study area that exhibit high F-parameter, Kd, and K/eTh values was recommended. Additionally, the research region has a few isolated enriched spots of (K). Therefore, GRS data was used to characterise and estimate potential metallic ores, nonmetallic deposits, and gold ore zones associated with the alteration zones. Results show that most of the known mineral deposits and gold occurrences in the area, according to the metallogenic map of Egypt, are located in zones with a ratio value of (0.25–0.30) (K%/(U or Th ppm)) maps which may suggest a moderate degree of alteration. Also, most mineral deposits and gold occurrences are found in intermediate altered zones, or K-enriched sites, with a Kd% of (0.2. The work represents an attempt to map hydrothermal alteration zones associated with mineral deposits in the Wadi Al-Allaqi area. Generally, natural radiation characteristics and attributes suggest criteria that can be used globally for regional mineral exploration.

Application of Bayesian classifier to magnetic and gamma ray spectrometry data for targeting hydrocarbon microseepages

In this study, we propose a CNN-GAN-based real-time processing technique for filtering images of underwater cables used in power systems. This addresses the excessive interference impurities that are frequently observed in images captured by remotely operated vehicles (ROVs). The process begins with the input of the original image into the convolutional neural network (CNN). Subsequently, the training outcomes, which serve as input parameters for the generative adversarial network (GAN), facilitate the filtering process. The system also calculates both the structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR), performing model updates via backward propagation. This technique utilizes deep learning technologies to achieve rapid, real-time filtering of underwater cable images. The experimental results reveal that the loss function of the CNN reaches 0.16 with an accuracy of 97.5%, while the loss function of the adversarial GAN network approaches 0.05. Compared with traditional methods such as DDN, JORDOR, RESCAN, and PRENet, the proposed CNN-GAN algorithm exhibits superior performance, as evidenced by the higher PSNR and SSIM values. Specifically, for clear water images, the PSNR reaches 29.86 dB and the SSIM is 0.9045. For severely polluted images, the PSNR is 28.67 dB and the SSIM is 0.8965, while for unevenly illuminated images, the PSNR and SSIM values are 24.37 dB and 0.88, respectively. These enhancements significantly benefit the monitoring and maintenance of power systems.

Prediction of pleural invasion in lung cancer is crucial in planning appropriate operating procedures and can be assessed using dynamic chest radiography (DCR). However, this assessment is negatively affected by rib shadows in conventional images. The purpose of this study was to develop a deep learning-based bone suppression technique for DCR in various projection directions. Twenty breathing XCAT phantoms with lung tumors and a pair of phantoms consisting only of bone structures were generated. The XCAT phantoms were virtually projected from six directions, resulting in 54000 multidirectional chest X-ray images and were used for training a pix2pix model to estimate bone images from original images. Bone suppression (BS) images were created by subtracting the bone images from the original images and then evaluated based on the peak signal to noise ratio (PSNR), structural similarity index measure (SSIM), and Fréchet Video Distance (FVD). Clinical cases were processed using the trained model as well. Furthermore, lung tumors on the original and BS images were tracked using OpenCV template matching, and the tracking accuracy was compared. The PSNR, SSIM, and FVD were 0.9966, 52.20 and 136.9, respectively. In the visual evaluation, we confirmed the effect of BS without temporal fluctuation of pixel values in both the resulting images of the XCAT phantom and real patients. Furthermore, precise tracking of the targeted tumor was achieved on the resulting BS images even in oblique directions, without any interruption from the rib shadows. These results indicate that our proposed method can effectively reduce bone shadows as well as temporal variations in the effect of bone suppression.

Blurring facial images can significantly degrade the performance of face recognition and video surveillance applications. Therefore, an effective image restoration method is essential to address this issue. However, existing methods struggle with varying levels of blur severity, limiting their effectiveness. This study proposes a facial image restoration approach that integrates blur severity classification using a Convolutional Neural Network (CNN) with a U-Net-based deblurring model to overcome these challenges. This method ensures that each blurred image is processed using the most suiTable deblurring model, optimizing the restoration process. The dataset used in this study is Flickr-Faces-HQ (FFHQ), to which a Gaussian blur is applied and categorized into five levels: very low, low, medium, high, and very high. They employ the Peak Signal-to-Noise Ratio (PSNR) and the Structural Similarity Index Measure (SSIM) as quantitative metrics to evaluate the model's performance. Experimental results show that the proposed method consistently outperforms fixed kernel size and multi-kernel size deblurring approaches across all blur severities. Specifically, our method achieves a PSNR of 40.001 dB for very low blur severity and an SSIM of 0.990. For low severity, it attains a PSNR of 31.104 dB and an SSIM of 0.946. For medium severity, the results are 27.995 dB (PSNR) and 0.874 (SSIM). At high severity, our model achieves a PSNR of 28.896 dB and an SSIM of 0.855. Finally, for very high severity, the PSNR drops to 26.566 dB, with an SSIM of 0.812. These results demonstrate the effectiveness of the proposed method in enhancing image clarity across different blur levels while preserving facial details. This research contributes to the development of more adaptive and efficient image restoration techniques, particularly for applications that require high-quality facial images, such as face recognition and video surveillance.

Gamma-ray spectrometry method is widely used in various fields. The study area is located in the Central Eastern Desert of Egypt and covers about 1200 square kilometers. The rock units in the study area can be organized in ages from the Precambrian (mainly granite) to Quaternary. Airborne gamma-ray spectrometry data were collected by Aero-Service, 1984. The main objectives of this study are to detect hydrothermal alteration zones and the uranium anomalies in the study area. The processing of the data was conducted to generate radioactive element concentration maps (K, eU and eTh). Hydrothermal alteration zones such as potassic and phyllic altered areas were detected using the Potassium Ternary Composite Image map and the eTh/K map. The F-parameter technique and K ideal method were used to determine the locations of the potassium-enrichment. These locations are characteristically associated with the orogenic gold mineralization in the study area. Fourteen radioactivity zones were classified based on the Interpreted Radio-Spectrometric Zonation (IRSZ) Map. These zones were divided into high, medium and low radioactive response where zones of high radioactive response were revealed as zones 1, 6, 9 and 13. Uranium anomalies have been identified as these anomalies are abundant in eU values and also associated with higher values of eU/eTh and eU/K, where the maximum value of eU is associated with the Younger Granite with a value of 18.831 ppm. These uranium anomalies are associated with Taref Formation, Post-Hammamat felsite, Quaternary deposits and Younger Granite. The locations of uranium anomalies are considered important exploration targets, as these locations are promising and have priority for ground geophysical and geological follow-up.

Airborne gamma-ray spectrometry to quantify chemical erosion processes

The size of pixels in a digital recording device, such as a CCD array, limits the spatial resolution in images obtained by an optical imaging system, thereby degrading the image quality. Digital super-resolution (DSR) techniques are used to reconstruct a high-resolution (HR) image from multiple sub-pixel-shifted low-resolution images in order to improve the image quality. In this article, we formulate a mathematical framework for DSR using the generalized sampling theorem (GST). The GST-based DSR method’s performance is evaluated by comparing it to existing resolution enhancement methods on the basis of evaluation metrics like percentage mean square error (%MSE), structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR). The GST DSR method exhibits an overall superior quality image reconstruction based on quantitative analysis (with a near zero %MSE, SSIM around one, and improved PSNR in dB) and qualitative comparison. The robustness of the GST DSR method is further demonstrated in the presence of frame-to-frame shift estimation error using %MSE and SSIM by comparing it with multi-frame interpolation approaches.

Speckle noise in ultrasound images compromises image quality and hinders diagnostic accuracy. Traditional ultrasound denoising methods often struggle to preserve anatomical details while effectively reducing noise, especially under high-noise conditions. In this study, we propose an innovative approach that integrates a lightweight channel attention mechanism (LCAM) within a convolutional variational autoencoder (CVAE) framework to enhance ultrasound image denoising. The proposed approach efficiently reduces speckle noise while maintaining essential anatomical features. Comprehensive evaluations across six diverse ultrasound datasets demonstrate that the LCAM-CVAE outperforms conventional denoising techniques in both subjective image quality and objective performance metrics, including peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), standard deviation in PSNR (SD-PSNR), standard deviation in SSIM (SD-SSIM), PSNR statistical relevance tests, and computational efficiency (CE). The LCAM-CVAE approach demonstrates exceptional performance, particularly under high-noise conditions, ensuring the preservation of key anatomical structures for accurate diagnosis. These results highlight the LCAM-CVAE approach as a robust and promising solution for ultrasound image denoising, with significant clinical potential to improve diagnostic quality in noisy environments.

The aim of this study is to remove scattered photons and beam hardening effect in cone beam CT (CBCT) images and make an image available for treatment planning. To remove scattered photons and beam hardening effect, a convolutional neural network (CNN) was used, and trained with distorted projection data including scattered photons and beam hardening effect and supervised projection data calculated with monochromatic X-rays. The number of training projection data was 17,280 with data augmentation and that of test projection data was 540. The performance of the CNN was investigated in terms of the number of photons in the projection data used in the training of the network. Projection data of pelvic CBCT images (32 cases) were calculated with a Monte Carlo simulation with six different count levels ranging from 0.5 to 3 million counts/pixel. For the evaluation of corrected images, the peak signal-to-noise ratio (PSNR), the structural similarity index measure (SSIM), and the sum of absolute difference (SAD) were used. The results of simulations showed that the CNN could effectively remove scattered photons and beam hardening effect, and the PSNR, the SSIM, and the SAD significantly improved. It was also found that the number of photons in the training projection data was important in correction accuracy. Furthermore, a CNN model trained with projection data with a sufficient number of photons could yield good performance even though a small number of photons were used in the input projection data.

Compression in image data needs to be done to manage storage space efficiently. In this study, image compression is performed using the Differential Evolution (DE) algorithm as an approach to achieve an optimal threshold for pixel clustering in images. The performance of this image compression method is evaluated based on the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) parameters. The PSNR of DE image compression ranges from 27 dB to 40 dB with threshold levels between 10 to 40. Meanwhile, the SSIM ranges from 0.83 to 0.97 at threshold levels of 10 to 40. The PSNR and SSIM of DE image compression are both at a good level, resulting in compressed image visuals that closely resemble the original image as the threshold levels increase.

The object of research is the adaptive switching weighted median image filter (ASWM) algorithm. This algorithm is one of the most effective in the field of impulse noise suppression. The computational complexity and algorithmic features of this adaptive nonlinear filter make it impossible to implement a filter that works in real time on modern PLD microcircuits. The most problematic areas of the algorithm are the weight coefficient estimation cycle, which has no limit on the number of iterations and contains a large number of division operations. This does not allow implementing the filter on PLDs with a sufficiently effective method. In the course of the research, the programming model of the filter in Python was used. The performance of the algorithm was assessed using the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) metrics. Modeling made it possible to find out empirically the number of iterations of the cycle for estimating the weight coefficients at different levels of noise density and to estimate the effect of artificial limitation of the maximum number of iterations on the filter performance. Regardless of the intensity of the noise impact, the algorithm performs less than 40 iterations of the evaluation cycle. Let’s also simulate the operation of the algorithm with different variants of the division module implementation. The paper considers the main of them and offers the most optimal in terms of the ratio of accuracy/hardware costs for implementation. Thus, a modified algorithm was proposed that does not have these disadvantages. Thanks to modifications of the algorithm, it is possible to implement a pipelined ASWM image filter on modern PLDs. The filter is synthesized for the main families of Intel PLDs. The implementation, which is not inferior in terms of SSIM and PSNR metrics to the original algorithm, requires less than 65,000 FPGA logical cells and allows filtering of monochrome images with FullHD resolution at 48 frames/s at a clock frequency of 100 MHz.

PurposeThe purpose of this study was to evaluate the accuracy and perceptual quality of red-green-blue (RGB) fundus images generated from red-green (RG) images using Pix2Pix-based generative adversarial network (GAN).MethodsRG images were extracted from original RGB images recorded by a newer model of the Optos system by turning off the blue laser source. A conditional GAN based on the Pix2Pix architecture was trained to convert this extracted RG image to an RGB image. The performance was evaluated quantitatively by using a five-fold cross-validation and assessments with structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), learned perceptual image patch similarity (LPIPS), mean absolute error (MAE), and root mean square error (RMSE). To assess the perceptual accuracy, 39 ophthalmologists examined 60 randomly selected images, 30 true and 30 generated RGB images, using Google Form-based interface. They were instructed to classify each image as a true or an artificial intelligence (AI)-generated image.ResultsThe generated images had high similarity to the original RGB images (SSIM = 0.97 ± 0.00; PSNR = 35.74 ± 0.17 decibel [dB]; LPIPS = 0.09 ± 0.00; MAE = 1.76 ± 0.04; and RMSE = 4.25 ± 0.10). The overall correct classification rate was 57.1%. Certain lesion types, such as epiretinal membrane and optic disc cupping, had higher discrimination rates. The receiver operating characteristic (ROC) analysis across all evaluations yielded an area under the curve (AUC) of 0.497 (P = 0.314).ConclusionsPix2Pix-based GANs can generate perceptually and structurally consistent RGB images from RG images without lesion-specific attention mechanisms.Translational RelevanceThis method enables a reinterpretation of legacy RG fundus images and may support future AI diagnostic applications.

Monte Carlo (MC)-based estimation of dose-averaged linear energy transfer (LETd) in proton therapy is accurate but computationally intensive. Although deep learning (DL) models offer efficient alternatives, few have been optimized for multi-source inputs or validated on anatomically complex tumors, such as nasopharyngeal carcinoma (NPC). This study developed and evaluated a multi-source DL framework for voxel-level LETd prediction in proton therapy for NPC, aiming to identify optimal input configurations and assess both prediction accuracy and clinical reliability. This study analyzed 100 patients with NPC, yielding 500 proton therapy fields. Ground-truth LETd maps were generated using GPU-based Monte Carlo simulations. A deep learning model was trained with different combinations of four input features: dose distribution, CT image, beam mask, and radiological depth, resulting in eight configurations (Mdose, Mdose+ct, Mdose+beam, Mdose+depth, Mdose+beam+depth, Mdose+ct+depth, Mdose+ct+beam, and Mdose+ct+beam+depth). The dataset was randomly divided into training (300 fields, 60 patients), validation (100 fields, 20 patients), and testing (100 fields, 20 patients) sets. Model performance was evaluated using mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), structural similarity index measure (SSIM), and 3D Gamma analysis. A 3D U-Net baseline was retrained under identical conditions for comparison. Model interpretability was examined with Gradient-weighted Class Activation Mapping (Grad-CAM), and prediction confidence was assessed using uncertainty maps. The Mdose+ct model achieved the best performance (MAE 0.21±0.03keV/µm, PSNR 28.99±1.33, NCC 0.98±0.01, SSIM 0.971±0.007, Gamma 3%/3mm: 94.69±3.69%), outperforming the 3D U-Net baseline (MAE 0.38±0.52keV/µm). Grad-CAM visualizations confirmed that the model focused on physically relevant regions, while uncertainty maps highlighted high-variability zones at LETd transitions. The model generated voxel-level LETd predictions in approximately 7 s per case, which was substantially faster than GPU-based MC simulations (30 s per case). The proposed multi-source deep learning framework enables rapid and accurate voxel-level LETd prediction for NPC. Combining dose with CT yielded the most reliable input configuration, and incorporating interpretability and uncertainty analyses enhances confidence in its suitability for biologically guided proton therapy planning.