Remote Sensing Image Dehazing through an Unsupervised Generative Adversarial Network.

The degradation of visual quality in remote sensing images caused by haze presents significant challenges in interpreting and extracting essential information. To effectively mitigate the impact of haze on image quality, we propose an unsupervised generative adversarial network specifically designed for remote sensing image dehazing. This network includes two generators with identical structures and two discriminators with identical structures. One generator is focused on image dehazing, while the other generates images with added haze. The two discriminators are responsible for distinguishing whether an image is real or generated. The generator, employing an encoder-decoder architecture, is designed based on the proposed multi-scale feature-extraction modules and attention modules. The proposed multi-scale feature-extraction module, comprising three distinct branches, aims to extract features with varying receptive fields. Each branch comprises dilated convolutions and attention modules. The proposed attention module includes both channel and spatial attention components. It guides the feature-extraction network to emphasize haze and texture within the remote sensing image. For enhanced generator performance, a multi-scale discriminator is also designed with three branches. Furthermore, an improved loss function is introduced by incorporating color-constancy loss into the conventional loss framework. In comparison to state-of-the-art methods, the proposed approach achieves the highest peak signal-to-noise ratio and structural similarity index metrics. These results convincingly demonstrate the superior performance of the proposed method in effectively removing haze from remote sensing images.

Unsupervised Low-Light Image Enhancement Based on Generative Adversarial Network.

Low-light image enhancement aims to improve the perceptual quality of images captured under low-light conditions. This paper proposes a novel generative adversarial network to enhance low-light image quality. Firstly, it designs a generator consisting of residual modules with hybrid attention modules and parallel dilated convolution modules. The residual module is designed to prevent gradient explosion during training and to avoid feature information loss. The hybrid attention module is designed to make the network pay more attention to useful features. A parallel dilated convolution module is designed to increase the receptive field and capture multi-scale information. Additionally, a skip connection is utilized to fuse shallow features with deep features to extract more effective features. Secondly, a discriminator is designed to improve the discrimination ability. Finally, an improved loss function is proposed by incorporating pixel loss to effectively recover detailed information. The proposed method demonstrates superior performance in enhancing low-light images compared to seven other methods.

Results of Sacral Neuromodulation Therapy for Urinary Voiding Dysfunction: Five-Year Experience of a Retrospective, Multicenter Study in China.

PURPOSE: This five-year, retrospective, multicenter study evaluated the long-term safety and efficiency of sacral neuromodulation (SNM) in Chinese patients with urinary voiding dysfunction. PATIENTS AND METHODS: This is a Chinese national, multicenter, retrospective study that included 247 patients (51.2% female) who received an implantable pulse generator (IPG) (InterStim, Medtronic, Minneapolis, MN, USA) between 2012 and 2016. Success was considered if the initial ≥50% improvement in any of primary voiding diary variables persisted compared with baseline. The results were further stratified by identifying patients who showed >50% improvement and those although showed <50% improvement but still wanted to receive IPG; these data were collected and analyzed for general improvement. RESULTS: Following test stimulation, 187 patients (43%) declined implantation and 247 (57%) underwent implantation using InterStim®. Among 247 patients, 34 (13.7%) had overactive bladder (OAB), 59 (23.8%) had interstitial cystitis/bladder pain syndrome (IC/BPS), 47 (19%) had idiopathic urinary retention (IUR), and 107 (44.1%) had neurogenic bladder (NB). IPG efficiency rate for OAB, interstitial cystitis/bladder pain syndrome, idiopathic urinary retention, and neurogenic bladder were 42.5, 72.4, 51.6, and 58.8%, respectively. The mean duration of follow-up was 20.1 ± 12.8 months. CONCLUSIONS: SNM appears effective in the long term, with a total IPG implantation rate of approximately 57% (ranging between 42.5 and 72.4% depending on indication). Interstitial cystitis/bladder pain syndrome appear to be the best indication for stage I testing. Chinese neurogenic bladder patients are most inclined to choose SNM. SNM is relatively safe, with low postoperation adverse events of 16.1% and reoperation rate of 3.2% during the follow-up period.

Perception-oriented generative adversarial network for retinal fundus image super-resolution.

Retinal fundus imaging is a crucial diagnostic tool in ophthalmology, enabling the early detection and monitoring of various ocular diseases. However, capturing high-resolution fundus images often presents challenges due to factors such as defocusing and diffraction in the digital imaging process, limited shutter speed, sensor unit density, and random noise in the image sensor or during image transmission. Super-resolution techniques offer a promising solution to overcome these limitations and enhance the visual details in retinal fundus images. Since the retina has rich texture details, the super-resolution images often introduce artifacts into texture details and lose some fine retinal vessel structures. To improve the perceptual quality of the retinal fundus image, a generative adversarial network that consists of a generator and a discriminator is proposed. The proposed generator mainly comprises 23 multi-scale feature extraction blocks, an image segmentation network, and 23 residual-in-residual dense blocks. These components are employed to extract features at different scales, acquire the retinal vessel grayscale image, and extract retinal vascular features, respectively. The generator has two branches that are mainly responsible for extracting global features and vascular features, respectively. The extracted features from the two branches are fused to better restore the super-resolution image. The proposed generator can restore more details and more accurate fine vessel structures in retinal images. The improved discriminator is proposed by introducing our designed attention modules to help the generator yield clearer super-resolution images. Additionally, an artifact loss function is also introduced to enhance the generative adversarial network, enabling more accurate measurement of the disparity between the high-resolution image and the restored image. Experimental results show that the generated images obtained by our proposed method have a better perceptual quality than the state-of-the-art image super-resolution methods.

Low-light image enhancement using generative adversarial networks.

In low-light environments, the amount of light captured by the camera sensor is reduced, resulting in lower image brightness. This makes it difficult to recognize or completely lose details in the image, which affects subsequent processing of low-light images. Low-light image enhancement methods can increase image brightness while better-restoring color and detail information. A generative adversarial network is proposed for low-quality image enhancement to improve the quality of low-light images. This network consists of a generative network and an adversarial network. In the generative network, a multi-scale feature extraction module, which consists of dilated convolutions, regular convolutions, max pooling, and average pooling, is designed. This module can extract low-light image features from multiple scales, thereby obtaining richer feature information. Secondly, an illumination attention module is designed to reduce the interference of redundant features. This module assigns greater weight to important illumination features, enabling the network to extract illumination features more effectively. Finally, an encoder-decoder generative network is designed. It uses the multi-scale feature extraction module, illumination attention module, and other conventional modules to enhance low-light images and improve quality. Regarding the adversarial network, a dual-discriminator structure is designed. This network has a global adversarial network and a local adversarial network. They determine if the input image is actual or generated from global and local features, enhancing the performance of the generator network. Additionally, an improved loss function is proposed by introducing color loss and perceptual loss into the conventional loss function. It can better measure the color loss between the generated image and a normally illuminated image, thus reducing color distortion during the enhancement process. The proposed method, along with other methods, is tested using both synthesized and real low-light images. Experimental results show that, compared to other methods, the images enhanced by the proposed method are closer to normally illuminated images for synthetic low-light images. For real low-light images, the images enhanced by the proposed method retain more details, are more apparent, and exhibit higher performance metrics. Overall, compared to other methods, the proposed method demonstrates better image enhancement capabilities for both synthetic and real low-light images.

Remote sensing image dehazing using generative adversarial network with texture and color space enhancement.

Remote sensing is gradually playing an important role in the detection of ground information. However, the quality of remote-sensing images has always suffered from unexpected natural conditions, such as intense haze phenomenon. Recently, convolutional neural networks (CNNs) have been applied to deal with dehazing problems, and some important findings have been obtained. Unfortunately, the performance of these classical CNN-based methods still needs further enhancement owing to their limited feature extraction capability. As a critical branch of CNNs, the generative adversarial network (GAN), composed of a generator and discriminator, has become a hot research topic and is considered a feasible approach to solving the dehazing problems. In this study, a novel dehazed generative adversarial network (GAN) is proposed to reconstruct the clean images from the hazy ones. For the generator network of the proposed GAN, the color and luminance feature extraction module and the high-frequency feature extraction module aim to extract multi-scale features and color space characteristics, which help the network to acquire texture, color, and luminance information. Meanwhile, a color loss function based on hue saturation value (HSV) is also proposed to enhance the performance in color recovery. For the discriminator network, a parallel structure is designed to enhance the extraction of texture and background information. Synthetic and real hazy images are used to check the performance of the proposed method. The experimental results demonstrate that the performance can significantly improve the image quality with a significant increment in peak-signal-to-noise ratio (PSNR). Compared with other popular methods, the dehazing results of the proposed method closely resemble haze-free images.

Sacral neuromodulation for overactive bladder using the InterStim and BetterStim systems.

This study aimed to evaluate differences in the clinical outcomes of different sacral neuromodulation systems (InterStim and BetterStim) used in the treatment of overactive bladder. Data from a previously established database of sacral neuromodulation in China (the InterStim system) and a 2020 clinical trial of the BetterStim system were screened. Patients with overactive bladder undergoing stage II implanted pulse generator implantation were selected for analysis and divided into InterStim and BetterStim system groups. Voiding diaries and subjective scores obtained preoperatively, after stage I tined-lead implantation (experience period), and after stage II implanted pulse generator implantation were compared between the two groups. This study included 113 patients with overactive bladder (43, InterStim system group; 70, BetterStim system group). Voiding diaries and subjective scores significantly improved in both the InterStim and BetterStim system groups over the treatment period. Specifically, the urination frequency (all P < 0.001), average voiding volume (all P < 0.001), and average urinary leakage (InterStim, P < 0.05; BetterStim, P < 0.01) in both groups significantly improved at different periods during treatment. At the same time, the urgency perception scale (P < 0.001) and OAB-related quality of life score (InterStim, P < 0.05; BetterStim, P < 0.01) also significantly improved. There was no significant difference in urination frequency at baseline between the two groups (P = 0.169). Urination frequency was significantly higher in the BetterStim system group than in the InterStim group during the experience period and at follow-up (P = 0.031, P = 0.006). There was no significant difference in the number of urinary leakages between the different systems at baseline (P = 0.662), although this was higher in the InterStim system group during the experience period (P = 0.016), and the difference disappeared at the last follow-up (P = 0.565). There were significant differences in baseline urgency perception scale (P = 0.001) and OAB-related quality of life score (P < 0.001) between the two groups; however, these differences were not maintained at follow-up (P = 0.81, P = 0.479). Both sacral neuromodulation systems are safe and effective in treating overactive bladder. The InterStim system may be more beneficial for patients with dry overactive bladder. Satisfactory outcomes may be achieved with the BetterStim system in patients with wet overactive bladder. However, further studies are required to confirm this finding.

Influence of patient sex on the effectiveness of sacral neuromodulation: A cohort study from China.

BACKGROUND: Sacral neuromodulation (SNM) has been widely used to treat lower urinary tract dysfunction. Studies have shown a higher conversion rate among female patients than among male patients. However, the influence of gender on the clinical effectiveness of SNM remains unclear. We aimed to confirm whether patients of both genders show similar benefits after SNM treatment. MATERIALS AND METHODS: Clinical data of patients with lower urinary tract symptoms associated with pelvic floor dysfunction (overactive bladder, neurogenic bladder, interstitial cystitis/painful bladder syndrome, idiopathic urinary retention) treated with SNM in 10 medical centres in China between January 2012 and December 2016 were retrospectively collected. The patients were classified by gender. Variations in objective (voiding diary) and subjective scores in the baseline, testing, and last follow-up periods were compared. Data were analysed using statistical measures. RESULTS: The study included 203 patients (93 males, 110 females). There were no statistical differences in baseline information between the two groups, both groups showed improvement over time. Unsatisfactory improvement was observed in the quality of life and sexual life scores of both groups over the entire treatment period (all p＞0.05). Although there was a difference in the maximum voiding volume between the groups at baseline, no difference was observed at the last follow-up (p = 0.004, p = 0.044, p = 0.124), unlike in the average volume where a difference was noted at the last follow-up (p = 0.085, p = 0.964, p = 0.031). While there were no differences in quality of life, sexual life, or pelvic pain and urinary urgency frequency scores at baseline, a significant difference was observed at the last follow-up, and the degree of improvement was less among female patients (p = 0.836, p = 0.131, p = 0.015; p = 0.294, p = 0.265, p = 0.013; p = 0.299, p = 0.087, p = 0.015). CONCLUSION: SNM treatment elicited a similar effect on patients of both gender; however, a significant difference was observed regarding patient satisfaction with the treatment. Further preoperative patient education, especially, for female patients with interstitial cystitis/painful bladder syndrome may improve patient satisfaction.

Clinical outcomes of sacral neuromodulation in non-neurogenic, non-obstructive dysuria: A 5-year retrospective, multicentre study in China.

BACKGROUND: Management of non-neurogenic, non-obstructive dysuria represents one of the most challenging dilemmas in urological practice. The main clinical symptom is the increase in residual urine. Voiding dysfunction is the main cause of dysuria or urinary retention, mainly due to the decrease in bladder contraction (the decrease in contraction amplitude or duration) or the increase in outflow tract resistance. Sacral neuromodulation (SNM) has been used for > 10 years to treat many kinds of lower urinary tract dysfunction. It has become increasingly popular in China in recent years. Consequently, studies focusing on non-neurogenic, non-obstructive dysuria patients treated by SNM are highly desirable. AIM: To assess the outcome of two-stage SNM in non-neurogenic, non-obstructive dysuria. METHODS: Clinical data of 54 patients (26 men, 28 women) with non-neurogenic, non-obstructive dysuria treated by SNM from January 2012 to December 2016 in ten medical centers in China were retrospectively analyzed. All patients received two or more conservative treatments. The voiding diary, urgency score, and quality of life score before operation, after implantation of tined lead in stage I (test period), and during short-term follow-up (latest follow-up) after implantation of the implanted pulse generator in stage II were compared to observe symptom improvements. RESULTS: Among the 54 study patients, eight refused to implant an implanted pulse generator because of the unsatisfactory effect, and 46 chose to embed the implanted pulse generator at the end of stage I. The conversion rate of stage I to stage II was 85.2%. The average follow-up time was 18.6 mo. There were significant differences between baseline (before stage I) and the test period (after stage I) in residual urine, voiding frequency, average voiding amount, maximum voiding amount, nocturia, urgency score, and quality of life score. The residual urine and urgency score between the test period and the latest follow-up time (after stage II) were also significantly different. No significant differences were observed for other parameters. No wound infection, electrode breakage, or other irreversible adverse events occurred. CONCLUSION: SNM is effective for patients with non-neurogenic, non-obstructive dysuria showing a poor response to traditional treatment. The duration of continuous stimulation may be positively correlated with the improvement of residual urine.

The association of CXCR4 expression with clinicopathological significance and potential drug target in prostate cancer: a meta-analysis and literature review.

CXCR4/CXCL12 axis plays an important role in tumor growth, angiogenesis, metastasis, and therapeutic resistance. The aim of this study is to perform a meta-analysis and literature review to evaluate the association of CXCR4 expression with clinicopathological significance and prognosis in patients with prostate cancer (PCa). A detailed literature search was made in Medline, EMBASE, Web of Science, and Google Scholar for related research publications. The data were extracted and assessed independently. Analysis of pooled data was performed using Review Manager 5.2. Odds ratio (OR) with corresponding confidence intervals were calculated and summarized. The meta-analysis included a total of eleven studies and 630 patients. The rate of CXCR4 protein expression in PCa was significantly higher than in nonmalignant prostate tissues (OR =35.71, P<0.00001). The expression of CXCR4 protein was not significantly associated with Gleason score (P=0.73). However, the frequency of CXCR4 protein expression was significantly higher in T3-4 stage than in T1-2 stage of PCa (OR =2.35, P=0.001). The expression of CXCR4 protein was significantly associated with the presence of lymph node and bone metastasis of PCa: for lymph node metastasis positive versus negative, OR was 5.07 and P=0.0003, and for bone metastasis positive versus negative, OR was 7.03 and P=0.003. Cancer-specific survival of patients with PCa was significantly associated with CXCR4 protein expression, and the pooled Hazard ratio was 0.24 and P=0.002. In conclusion, the high expression of CXCR4 protein is a diagnostic biomarker of PCa, and it is significantly associated with T stages. The increased expression of CXCR4 protein is significantly associated with lymph nodes or bone metastasis, and CXCR4 is a poor prognosis predictor for patients with PCa. Taken together, our findings indicate that CXCR4 could be a target not only for the development of therapeutic intervention but also for the noninvasive monitoring of PCa progression.

[Analysis of the results of percent free prostate specific antigen detection among the men without prostate diseases in Xi'an].

OBJECTIVE: To measure the percentage of free prostate specific antigen (fPSA) among the men without prostate diseases in Xi'an, and understand the age-related changes in percent fPSA and its association with the pathological grade and clinical stage of prostate cancer, so as to determine the reference range of percent fPSA in Chinese men. METHODS: A total of 829 men (including 775 men without prostate disease and 54 with prostate cancer) were enrolled into the study and the serum PSA and fPSA levels were measured for calculating the percent fPSA. The 775 prostate disease-free subjects were divided into 5 age groups to study the relations of PSA, fPSA and percent fPSA with age. In the 54 patients with prostate cancer, the relations of pathological grade and clinical stage with percent fPSA were investigated with reference to the related data of the men without prostate disease to establish the appropriate reference range of percent fPSA for Chinese men. RESULTS: Serum PSA or fPSA levels, but not the percent fPSA, were correlated to age. In patients with prostate cancer, the percent fPSA was also correlated to the pathological grade and clinical stage. Among the men without prostate disease we investigated, the percent fPSA was significantly lower than that reported in other countries. The reference range of percent fPSA for Chinese men was > or =15%. CONCLUSION: The percent fPSA is not correlated to age and has greater clinical value than PSA measurement in the detection of prostate cancer. The percent fPSA decreases in prostate cancer patients with lowered pathological grade, advanced clinical stage, and therefore increased malignancy. The reference range of percent fPSA reported in other countries does not apply to Chinese men, and according to our investigation, the reference range of percent fPSA> or =15% is appropriate for Chinese men.