European Journal of Cardiovascular Medicine

Background: Diabetic Retinopathy (DR) is one of the serious complications of diabetes, which can be diagnosed early to prevent vision loss and blindness. Deep learning has progressed rapidly over the past few years, leading to effective image synthesis methods for many medical imaging applications. In the study, we propose an Efficient Deep Convolutional Generative Adversarial Network (EDC-GAN) framework for predicting and generating high-resolution diabetic retinopathy fundus images. Through an attention-based feature extraction method and an improved conditional GAN structure, the proposed model directly synthesizes high-fidelity, diverse, and realistic retinal images. We also employ a hybrid loss function composed of perceptual loss and adversarial loss to improve the quality of synthetic images even more. Large-scale contrastive learning with an instance noise module is also proposed to retain fine-grained details in the synthesized images, as well as a new component, Feature-Preserving Adaptive Normalization (FPAN), in the model. We train the model on publicly available DR datasets and validate it using several evaluation metrics such as Freshet Inception Distance (FID), Structural Similarity Index (SSIM), and Peak Signal-to-Noise Ratio (PSNR). Experimental results show that EDC-GAN outperforms existing image synthesis methods in terms of generating higher quality and more realistic synthetic fundus images and improving the modeling performance in terms of predictive power. Therefore, the proposed model presents a very effective approach to augment a scarce DR dataset and to assure robust computer-aided diagnosis (CAD) for DR detection.

Background: Post-COVID sequelae in young adults have garnered significant attention, particularly regarding cardiopulmonary recovery. This study aims to evaluate the impact of COVID-19 on resting oxygen saturation and exercise tolerance in young adults. Methods: An observational study was conducted between March 2020 and June 2020 involving 100 young adults (aged 18–35 years) who had recovered from mild to moderate COVID-19. Baseline demographic data, resting oxygen saturation (SpO₂), and 6-minute walk test (6MWT) performance were recorded. Post-exercise desaturation (≥4% drop in SpO₂), fatigue scores, and heart rate changes were analyzed. Symptomatology was assessed via self-reported outcomes. Results: The mean age was 26.8 ± 4.9 years with 58% males. Mean BMI was 24.6 ± 3.2 kg/m². Average resting SpO₂ was 96.4% ± 1.8; 12 participants (12%) had SpO₂ < 95%. The mean 6MWT distance was 465.3 ± 54.7 meters. A ≥4% SpO₂ drop was observed in 28% of participants. These individuals exhibited lower resting SpO₂, reduced walk distance (430.6 ± 48.1 meters vs. 478.2 ± 50.3 meters, p < 0.01), and higher fatigue scores (6.3 ± 1.7 vs. 4.5 ± 1.5, p < 0.01). Persistent fatigue and exertional dyspnea were reported in 37% and 29% respectively. Conclusion: A significant proportion of young adults exhibit post-COVID impairments in oxygen saturation and exercise tolerance, even after mild to moderate infection. These findings highlight the need for post-recovery monitoring and rehabilitation strategies in this population.

Background: Continuous Tretinoin therapy has gained attention for its potential in treating various dermatological conditions, yet its underlying biochemical mechanisms remain insufficiently explored. This study aimed to elucidate the molecular changes induced by continuous retinoid therapy and identify potential biomarkers for treatment response. Methods: A randomized, double-blind, placebo-controlled trial was conducted with 120 participants from a tertiary care hospital. Molecular analyses, including transcriptomics and epigenomics, characterized gene expression and epigenetic modifications. Cytokine profiles were assessed, and clinical responses were rigorously evaluated using validated scoring systems. Data were analyzed using bioinformatics pipelines, statistical tests, and machine learning algorithms. Results: Significant upregulation of the "Keratinocyte Proliferation" pathway indicated cellular turnover enhancement. A moderate rise in the "Inflammation" pathway suggested immunomodulatory effects. IL-6 levels decreased significantly (p = 0.042), suggesting inflammation suppression. Clinical response was markedly higher in the treatment group at Week 4 (p < 0.001) and Week 8 (p = 0.018). Potential biomarkers, such as "Gene A" (p < 0.001), demonstrated predictive capabilities. Conclusion: Continuous Tretinoin therapy induces molecular changes related to enhanced cellular turnover and immunomodulation. IL-6 reduction aligns with observed clinical improvements. The study underscores the potential of biomarkers for treatment stratification and offers insights into the mechanisms of continuous retinoid therapy. Further investigations are warranted to validate and extend these findings.

Background: Alopecia areata is an autoimmune disorder characterized by sudden hair loss. The management of this condition is challenging due to its unpredictable nature and limited therapeutic options. Platelet-rich plasma (PRP) and intralesional triamcinolone acetonide (TA) have emerged as potential treatments, but a comprehensive comparison is lacking. Methods: A prospective, randomized, controlled clinical trial was conducted at a tertiary care hospital. Participants (n=120) aged 18-60 years with alopecia areata were randomized into PRP and intralesional TA groups. Primary outcomes included hair regrowth assessed through standardized photographs and the Severity of Alopecia Tool (SALT) score at baseline, post-third treatment session, and 6-month follow-up. Secondary outcomes comprised response rates, safety profiles, treatment response duration, and relapse rates. Statistical analysis employed appropriate tests. Results: Participants' baseline characteristics were comparable between groups. PRP demonstrated significantly higher hair regrowth rates at post-third session (55.3% vs. 41.2%) and 6-month follow-up (68.9% vs. 53.7%). Response rates were notably greater in the PRP group (83.3% vs. 67.5%). Safety profiles were similar. PRP treatment showed a longer treatment response duration (12.6 vs. 9.8 months) and comparable relapse rates (15% vs. 23.3%). Conclusion: This study highlights the superior efficacy of PRP over intralesional TA in alopecia areata treatment. PRP exhibited higher hair regrowth and response rates, longer treatment response duration, and comparable safety and relapse rates. PRP presents as a promising therapeutic option for managing alopecia areata.