Saved Queries

Fibrillins (FBNs) are indispensable for plant growth and development, orchestrating multiple physiological processes. However, the precise functional role of FBNs in cotton fiber development remains uncharacterized. This study reports a genome-wide characterization of the FBN gene family in cotton. A total of 28 GhFBN genes were identified in upland cotton, with systematic analyses of their phylogenetic relationships, protein motifs, gene structures, and hormone-responsive cis-regulatory elements. Expression profiling of GhFBN1A during fiber development revealed stage-specific activity across the developmental continuum. Transcriptomic analyses following hormone treatments demonstrated upregulation of GhFBN family members, implicating their involvement in hormone-mediated regulatory networks governing fiber cell development. Collectively, this work presents a detailed molecular characterization of cotton GhFBNs and establishes a theoretical foundation for exploring their potential applications in cotton breeding programs aimed at improving fiber quality. Full article

(This article belongs to the Section Bioinformatics)

►▼ Show Figures

Figure 1

24 pages, 3507 KiB

Open AccessArticle

A Semi-Supervised Wildfire Image Segmentation Network with Multi-Scale Structural Fusion and Pixel-Level Contrastive Consistency

by Yong Sun, Wei Wei, Jia Guo, Haifeng Lin and Yiqing Xu

Fire 2025, 8(8), 313; https://doi.org/10.3390/fire8080313 (registering DOI) - 7 Aug 2025

Abstract

The increasing frequency and intensity of wildfires pose serious threats to ecosystems, property, and human safety worldwide. Accurate semantic segmentation of wildfire images is essential for real-time fire monitoring, spread prediction, and disaster response. However, existing deep learning methods heavily rely on large volumes of pixel-level annotated data, which are difficult and costly to obtain in real-world wildfire scenarios due to complex environments and urgent time constraints. To address this challenge, we propose a semi-supervised wildfire image segmentation framework that enhances segmentation performance under limited annotation conditions by integrating multi-scale structural information fusion and pixel-level contrastive consistency learning. Specifically, a Lagrange Interpolation Module (LIM) is designed to construct structured interpolation representations between multi-scale feature maps during the decoding stage, enabling effective fusion of spatial details and semantic information, and improving the model’s ability to capture flame boundaries and complex textures. Meanwhile, a Pixel Contrast Consistency (PCC) mechanism is introduced to establish pixel-level semantic constraints between CutMix and Flip augmented views, guiding the model to learn consistent intra-class and discriminative inter-class feature representations, thereby reducing the reliance on large labeled datasets. Extensive experiments on two public wildfire image datasets, Flame and D-Fire, demonstrate that our method consistently outperforms other approaches under various annotation ratios. For example, with only half of the labeled data, our model achieves 5.0% and 6.4% mIoU improvements on the Flame and D-Fire datasets, respectively, compared to the baseline. This work provides technical support for efficient wildfire perception and response in practical applications. Full article

►▼ Show Figures

Figure 1

22 pages, 6168 KiB

Open AccessArticle

Valorization of Sugarcane Bagasse in Thailand: An Economic Analysis of Ethanol and Co-Product Recovery via Organosolv Fractionation

by Suphalerk Khaowdang, Nopparat Suriyachai, Saksit Imman, Nathiya Kreetachat, Santi Chuetor, Surachai Wongcharee, Kowit Suwannahong, Methawee Nukunudompanich and Torpong Kreetachat

Sustainability 2025, 17(15), 7145; https://doi.org/10.3390/su17157145 (registering DOI) - 7 Aug 2025

Abstract

A comprehensive techno-economic assessment was undertaken to determine the viability of bioethanol production from sugarcane bagasse in Thailand through organosolv fractionation, incorporating three distinct catalytic systems: sulfuric acid, formic acid, and sodium methoxide. Rigorous process simulations were executed using Aspen Plus, facilitating the derivation of detailed mass and energy balances, which served as the foundational input for downstream cost modeling. Economic performance metrics, including the total annualized cost and minimum ethanol selling price, were systematically quantified for each scenario. Among the evaluated configurations, the formic acid-catalyzed organosolv system exhibited superior techno-economic attributes, achieving the lowest unit production costs of 1.14 USD/L for ethanol and 1.84 USD/kg for lignin, corresponding to an estimated ethanol selling price of approximately 1.14 USD/L. This favorable outcome was attained with only moderate capital intensity, indicating a well-balanced trade-off between operational efficiency and investment burden. Conversely, the sodium methoxide-based process configuration imposed the highest economic burden, with a TAC of 15.27 million USD/year, culminating in a markedly elevated MESP of 5.49 USD/kg (approximately 4.33 USD/L). The sulfuric acid-driven system demonstrated effective delignification performance. Sensitivity analysis revealed that reagent procurement costs exert the greatest impact on TAC variation, highlighting chemical expenditure as the key economic driver. These findings emphasize the critical role of solvent choice, catalytic performance, and process integration in improving the cost-efficiency of lignocellulosic ethanol production. Among the examined options, the formic acid-based organosolv process stands out as the most economically viable for large-scale implementation within Thailand’s bioeconomy. Full article

(This article belongs to the Special Issue Towards Circular Economy: Innovations in Waste Management and Technological Solutions)

►▼ Show Figures

Figure 1

17 pages, 216 KiB

Open AccessArticle

Inference to the Only Explanation: The Case of the Cretaceous/Paleogene Extinction Controversies

by Keith M. Parsons

Philosophies 2025, 10(4), 89; https://doi.org/10.3390/philosophies10040089 (registering DOI) - 7 Aug 2025

Abstract

In the sciences of the deep past, it is taken for granted that the hypothesis that offers the best explanation is the best confirmed. I examine in detail the debate over the K/Pg mass extinctions that began in 1980 with the publication of the paper by Alvarez et al. that proposed the impact extinction hypothesis. I summarize this debate and show how the impact hypothesis eventually achieved consensus as the best explanation. I then consider the relevance of that case study to an evaluation of the employment of inference to the best explanation (IBE) in the earth sciences. I first reject a number of the standard objections to IBE and then strongly endorse John Norton’s claim that no form of ampliative inference can receive a priori justification. Nevertheless, drawing on the case study and other instances, we may identify four “abductive virtues” that characterize many of the most successful instances of IBE, making them attractive and even compelling. Full article

16 pages, 53970 KiB

Open AccessArticle

UNet–Transformer Hybrid Architecture for Enhanced Underwater Image Processing and Restoration

by Jie Ji and Jiaju Man

Mathematics 2025, 13(15), 2535; https://doi.org/10.3390/math13152535 (registering DOI) - 6 Aug 2025

Abstract

Underwater image enhancement is crucial for fields like marine exploration, underwater photography, and environmental monitoring, as underwater images often suffer from reduced visibility, color distortion, and contrast loss due to light absorption and scattering. Despite recent progress, existing methods struggle to generalize across diverse underwater conditions, such as varying turbidity levels and lighting. This paper proposes a novel hybrid UNet–Transformer architecture based on MaxViT blocks, which effectively combines local feature extraction with global contextual modeling to address challenges including low contrast, color distortion, and detail degradation. Extensive experiments on two benchmark datasets, UIEB and EUVP, demonstrate the superior performance of our method. On UIEB, our model achieves a PSNR of 22.91, SSIM of 0.9020, and CCF of 37.93—surpassing prior methods such as URSCT-SESR and PhISH-Net. On EUVP, it attains a PSNR of 26.12 and PCQI of 1.1203, outperforming several state-of-the-art baselines in both visual fidelity and perceptual quality. These results validate the effectiveness and robustness of our approach under complex underwater degradation, offering a reliable solution for real-world underwater image enhancement tasks. Full article

(This article belongs to the Special Issue Algorithms, Models and Proofs: Advancing Scalability, Sustainability and Security in Large-Scale Computation and Cryptography)

►▼ Show Figures

Figure 1

26 pages, 6895 KiB

Open AccessArticle

Generation of Individualized, Standardized, and Electrically Synchronized Human Midbrain Organoids

by Sanae El Harane, Bahareh Nazari, Nadia El Harane, Manon Locatelli, Bochra Zidi, Stéphane Durual, Abderrahim Karmime, Florence Ravier, Adrien Roux, Luc Stoppini, Olivier Preynat-Seauve and Karl-Heinz Krause

Cells 2025, 14(15), 1211; https://doi.org/10.3390/cells14151211 - 6 Aug 2025

Abstract

Organoids allow to model healthy and diseased human tissues. and have applications in developmental biology, drug discovery, and cell therapy. Traditionally cultured in immersion/suspension, organoids face issues like lack of standardization, fusion, hypoxia-induced necrosis, continuous agitation, and high media volume requirements. To address these issues, we developed an air–liquid interface (ALi) technology for culturing organoids, termed AirLiwell. It uses non-adhesive microwells for generating and maintaining individualized organoids on an air–liquid interface. This method ensures high standardization, prevents organoid fusion, eliminates the need for agitation, simplifies media changes, reduces media volume, and is compatible with Good Manufacturing Practices. We compared the ALi method to standard immersion culture for midbrain organoids, detailing the process from human pluripotent stem cell (hPSC) culture to organoid maturation and analysis. Air–liquid interface organoids (3D-ALi) showed optimized size and shape standardization. RNA sequencing and immunostaining confirmed neural/dopaminergic specification. Single-cell RNA sequencing revealed that immersion organoids (3D-i) contained 16% fibroblast-like, 23% myeloid-like, and 61% neural cells (49% neurons), whereas 3D-ALi organoids comprised 99% neural cells (86% neurons). Functionally, 3D-ALi organoids showed a striking electrophysiological synchronization, unlike the heterogeneous activity of 3D-i organoids. This standardized organoid platform improves reproducibility and scalability, demonstrated here with midbrain organoids. The use of midbrain organoids is particularly relevant for neuroscience and neurodegenerative diseases, such as Parkinson’s disease, due to their high incidence, opening new perspectives in disease modeling and cell therapy. In addition to hPSC-derived organoids, the method’s versatility extends to cancer organoids and 3D cultures from primary human cells. Full article

(This article belongs to the Special Issue The Current Applications and Potential of Stem Cell-Derived Organoids)

►▼ Show Figures

Figure 1

20 pages, 6776 KiB

Open AccessArticle

Computational Approaches to Assess Flow Rate Efficiency During In Situ Recovery of Uranium: From Reactive Transport to Streamline- and Trajectory-Based Methods

by Maksat Kurmanseiit, Nurlan Shayakhmetov, Daniar Aizhulov, Banu Abdullayeva and Madina Tungatarova

Minerals 2025, 15(8), 835; https://doi.org/10.3390/min15080835 - 6 Aug 2025

Abstract

This study presents a comprehensive computational analysis of flow rate efficiency during uranium extraction via the In Situ Recovery method. Using field data from a deposit located in Southern Kazakhstan, a series of mathematical models were developed to evaluate the distribution and balance of leaching solution. A reactive transport model incorporating uranium dissolution kinetics and acid–rock interactions were utilized to assess the accuracy of both traditional and proposed methods. The results reveal a significant spatial imbalance in sulfuric acid distribution, with up to 239.1 tons of acid migrating beyond the block boundaries. To reduce computational demands while maintaining predictive accuracy, two alternative methods, a streamline-based and a trajectory-based approach were proposed and verified. The streamline method showed close agreement with reactive transport modeling and was able to effectively identify the presence of intra-block reagent imbalance. The trajectory-based method provided detailed insight into flow dynamics but tended to overestimate acid overflow outside the block. Both alternative methods outperformed the conventional approach in terms of accuracy by accounting for geological heterogeneity and well spacing. The proposed methods have significantly lower computational costs, as they do not require solving complex systems of partial differential equations involved in reactive transport simulations. The proposed approaches can be used to analyze the efficiency of mineral In Situ Recovery at both the design and operational stages, as well as to determine optimal production regimes for reducing economic expenditures in a timely manner. Full article

(This article belongs to the Section Mineral Processing and Extractive Metallurgy)

►▼ Show Figures

Figure 1

35 pages, 3289 KiB

Open AccessReview

Applications of Machine Learning Algorithms in Geriatrics

by Adrian Stancu, Cosmina-Mihaela Rosca and Emilian Marian Iovanovici

Appl. Sci. 2025, 15(15), 8699; https://doi.org/10.3390/app15158699 (registering DOI) - 6 Aug 2025

Abstract

The increase in the elderly population globally reflects a change in the population’s mindset regarding preventive health measures and necessitates a rethinking of healthcare strategies. The integration of machine learning (ML)-type algorithms in geriatrics represents a direction for optimizing prevention, diagnosis, prediction, monitoring, and treatment. This paper presents a systematic review of the scientific literature published between 1 January 2020 and 31 May 2025. The paper is based on the applicability of ML techniques in the field of geriatrics. The study is conducted using the Web of Science database for a detailed discussion. The most studied algorithms in research articles are Random Forest, Extreme Gradient Boosting, and support vector machines. They are preferred due to their performance in processing incomplete clinical data. The performance metrics reported in the analyzed papers include the accuracy, sensitivity, F1-score, and Area under the Receiver Operating Characteristic Curve. Nine search categories are investigated through four databases: WOS, PubMed, Scopus, and IEEE. A comparative analysis shows that the field of geriatrics, through an ML approach in the context of elderly nutrition, is insufficiently explored, as evidenced by the 61 articles analyzed from the four databases. The analysis highlights gaps regarding the explainability of the models used, the transparency of cross-sectional datasets, and the validity of the data in real clinical contexts. The paper highlights the potential of ML models in transforming geriatrics within the context of personalized predictive care and outlines a series of future research directions, recommending the development of standardized databases, the integration of algorithmic explanations, the promotion of interdisciplinary collaborations, and the implementation of ethical norms of artificial intelligence in geriatric medical practice. Full article

(This article belongs to the Special Issue Diet, Nutrition and Human Health)

►▼ Show Figures

Figure 1

14 pages, 1950 KiB

Open AccessArticle

Ancient Ritual Behavior as Reflected in the Imagery at Picture Cave, Missouri, USA

by Carol Diaz-Granados and James R. Duncan

Arts 2025, 14(4), 88; https://doi.org/10.3390/arts14040088 (registering DOI) - 6 Aug 2025

Abstract

Since 1992, we have promoted the use of descriptions from ethnographic data, including ancient, surviving oral traditions, to aid in explaining the iconography portrayed in pictographs and petroglyphs found in Missouri, particularly those at Picture Cave. The literature to which we refer is from American Indian groups related linguistically and connected to the pre-Columbian inhabitants of Missouri. In addition, we have had on-going conversations with many elder tribal members of the Dhegiha Sioux language group (including the Osage, Quapaw, and Kansa (the Ponca and Omaha are also part of this cognate linguistic group)). With the copious collections of southern Siouan ethnographic accounts, we have been able to explain salient features in the iconography of several of the detailed rock art motifs and vignettes, and propose interpretations. This Midwest region is part of the Cahokia interaction sphere, an area that displays western Mississippian symbolism associated with that found in Missouri rock art as well as on pottery, shell, and copper. Full article

(This article belongs to the Special Issue Advances in Rock Art Studies)

►▼ Show Figures

Figure 1

31 pages, 17555 KiB

Open AccessArticle

Evaluating Performance of Friction Stir Lap Welds Made at Ultra-High Speeds

by Todd Lainhart, Joshua Sheffield, Jeremy Russell, Jeremy Coyne and Yuri Hovanski

J. Manuf. Mater. Process. 2025, 9(8), 263; https://doi.org/10.3390/jmmp9080263 - 6 Aug 2025

Abstract

Friction stir lap welding has been utilized across research and industry for over a decade. However, difficulties in welding in the lap configuration without an interface-related defect have prevented the process from moving beyond low feed rates (generally less than 1.5 m per minute). As a means of making a huge leap in welding productivity, this study will evaluate friction stir welds made at 10 m per minute (mpm), detailing the changes to tool geometries and weld parameters that result in fully consolidated welds. Characterization of the subsequent material properties, namely through optical microscopy, CT scanning, microhardness testing, tensile and fatigue testing, hermetic seal pressure tests, and electron backscattered diffraction, is presented as a means of demonstrating the quality and repeatability of friction stir lap welds made at 10 mpm. Fully consolidated welds were produced at spindle speeds 5.5% faster and 2.9% slower than nominal values and weld depths ranging from 1% shallower to 8.2% deeper than nominal values. Additionally, the loading direction of the weld had a significant impact on tensile properties, with the advancing side of the weld measured to be 16% stronger in lap-shear tensile and 289% fatigue life improvement under all loading conditions measured when compared to the retreating side. Full article

►▼ Show Figures

Figure 1

31 pages, 334 KiB

Open AccessArticle

Enhancing Discoverability: A Metadata Framework for Empirical Research in Theses

by Giannis Vassiliou, George Tsamis, Stavroula Chatzinikolaou, Thomas Nipurakis and Nikos Papadakis

Algorithms 2025, 18(8), 490; https://doi.org/10.3390/a18080490 - 6 Aug 2025

Abstract

Despite the significant volume of empirical research found in student-authored academic theses—particularly in the social sciences—these works are often poorly documented and difficult to discover within institutional repositories. A key reason for this is the lack of appropriate metadata frameworks that balance descriptive richness with usability. General standards such as Dublin Core are too simplistic to capture critical research details, while more robust models like the Data Documentation Initiative (DDI) are too complex for non-specialist users and not designed for use with student theses. This paper presents the design and validation of a lightweight, web-based metadata framework specifically tailored to document empirical research in academic theses. We are the first to adapt existing hybrid Dublin Core–DDI approaches specifically for thesis documentation, with a novel focus on cross-methodological research and non-expert usability. The model was developed through a structured analysis of actual student theses and refined to support intuitive, structured metadata entry without requiring technical expertise. The resulting system enhances the discoverability, classification, and reuse of empirical theses within institutional repositories, offering a scalable solution to elevate the visibility of the gray literature in higher education. Full article

(This article belongs to the Special Issue AI-Driven Solutions for Smart Systems in Engineering, Computing, Education, and Society)

16 pages, 6256 KiB

Open AccessArticle

Influence of Alpha/Gamma-Stabilizing Elements on the Hot Deformation Behaviour of Ferritic Stainless Steel

by Andrés Núñez, Irene Collado, Marta Muratori, Andrés Ruiz, Juan F. Almagro and David L. Sales

J. Manuf. Mater. Process. 2025, 9(8), 265; https://doi.org/10.3390/jmmp9080265 - 6 Aug 2025

Abstract

This study investigates the hot deformation behaviour and microstructural evolution of two AISI 430 ferritic stainless steel variants: 0A (basic) and 1C (modified). These variants primarily differ in chemical composition, with 0A containing higher austenite-stabilizing elements (C, N) compared to 1C, which features lower interstitial content and slightly higher Si and Cr. This research aimed to optimize hot rolling conditions for enhanced forming properties. Uniaxial hot compression tests were conducted using a Gleeble thermo-mechanical system between 850 and 990

^{\circ}

C at a strain rate of 3.3 s

^{- 1}

, simulating industrial finishing mill conditions. Analysis of flow curves, coupled with detailed microstructural characterization using electron backscatter diffraction, revealed distinct dynamic restoration mechanisms influencing each material’s response. Thermodynamic simulations confirmed significant austenite formation in both materials within the tested temperature range, notably affecting their deformation behaviour despite their initial ferritic state. Material 0A consistently exhibited a strong tendency towards dynamic recrystallization (DRX) across a wider temperature range, particularly at 850

^{\circ}

C. DRX led to a microstructure with a high concentration of low-angle grain boundaries and sharp deformation textures, actively reorienting grains towards energetically favourable configurations. However, under this condition, DRX did not fully complete the recrystallization process. In contrast, material 1C showed greater activity of both dynamic recovery and DRX, leading to a much more advanced state of grain refinement and recrystallization compared to 0A. This indicates that the composition of 1C helps mitigate the strong influence of the deformation temperature on the crystallographic texture, leading to a weaker texture overall than 0A. Full article

24 pages, 1471 KiB

Open AccessArticle

WDM-UNet: A Wavelet-Deformable Gated Fusion Network for Multi-Scale Retinal Vessel Segmentation

by Xinlong Li and Hang Zhou

Sensors 2025, 25(15), 4840; https://doi.org/10.3390/s25154840 - 6 Aug 2025

Abstract

Retinal vessel segmentation in fundus images is critical for diagnosing microvascular and ophthalmologic diseases. However, the task remains challenging due to significant vessel width variation and low vessel-to-background contrast. To address these limitations, we propose WDM-UNet, a novel spatial-wavelet dual-domain fusion architecture that integrates spatial and wavelet-domain representations to simultaneously enhance the local detail and global context. The encoder combines a Deformable Convolution Encoder (DCE), which adaptively models complex vascular structures through dynamic receptive fields, and a Wavelet Convolution Encoder (WCE), which captures the semantic and structural contexts through low-frequency components and hierarchical wavelet convolution. These features are further refined by a Gated Fusion Transformer (GFT), which employs gated attention to enhance multi-scale feature integration. In the decoder, depthwise separable convolutions are used to reduce the computational overhead without compromising the representational capacity. To preserve fine structural details and facilitate contextual information flow across layers, the model incorporates skip connections with a hierarchical fusion strategy, enabling the effective integration of shallow and deep features. We evaluated WDM-UNet in three public datasets: DRIVE, STARE, and CHASE_DB1. The quantitative evaluations demonstrate that WDM-UNet consistently outperforms state-of-the-art methods, achieving 96.92% accuracy, 83.61% sensitivity, and an 82.87% F1-score in the DRIVE dataset, with superior performance across all the benchmark datasets in both segmentation accuracy and robustness, particularly in complex vascular scenarios. Full article

(This article belongs to the Section Sensing and Imaging)

►▼ Show Figures

Figure 1

16 pages, 655 KiB

Open AccessReview

Seeing Opportunity in Virtual Reality: A Rapid Review of the Use of VR as a Tool in Vision Care

by Kiana Masoudi, Madeline Wong, Danielle Tchao, Ani Orchanian-Cheff, Michael Reber and Lora Appel

Technologies 2025, 13(8), 342; https://doi.org/10.3390/technologies13080342 (registering DOI) - 6 Aug 2025

Abstract

(1) Virtual reality (VR) technologies have shown significant potential for diagnosing and treating vision-related impairments. This rapid review evaluates and characterizes the existing literature on VR technologies for diagnosing and treating vision-based diseases. (2) Methods: A systematic search was conducted across Ovid MEDLINE, Ovid Embase, the Cochrane Database of Systematic Reviews (Ovid), and the Cochrane Central Register of Controlled Trials (Ovid). Abstracts were screened using Rayyan QCRI, followed by full-text screening and data extraction. Eligible studies were published in peer-reviewed journals, written in English, focused on human participants, used immersive and portable VR devices as the primary intervention, and reported on the clinical effectiveness of VR for therapeutic, diagnostic, or screening purposes for vision or auditory–visual impairments. Various study characteristics, including design and participant details, were extracted, and the MMAT assessment tool was used to evaluate study quality. (3) Results: Seventy-six studies met the inclusion criteria. Among these, sixty-four (84.2%) were non-randomized studies exploring VR’s effectiveness, while twenty-two (15.8%) were randomized-controlled trials. Of the included studies, 38.2% focused on diagnosing, 21.0% on screening, and 38.2% on treating vision impairments. Glaucoma and amblyopia were the most commonly studied visual impairments. (4) Conclusions: The use of standalone, remotely controlled VR headsets for screening and diagnosing visual diseases represents a promising advancement in ophthalmology. With ongoing technological developments, VR has the potential to revolutionize eye care by improving accessibility, efficiency, and personalization. Continued research and innovation in VR applications for vision care are expected to further enhance patient outcomes. Full article

(This article belongs to the Section Assistive Technologies)

►▼ Show Figures

Figure 1

18 pages, 1253 KiB

Open AccessArticle

Leveraging Synthetic Degradation for Effective Training of Super-Resolution Models in Dermatological Images

by Francesco Branciforti, Kristen M. Meiburger, Elisa Zavattaro, Paola Savoia and Massimo Salvi

Electronics 2025, 14(15), 3138; https://doi.org/10.3390/electronics14153138 - 6 Aug 2025

Abstract

Teledermatology relies on digital transfer of dermatological images, but compression and resolution differences compromise diagnostic quality. Image enhancement techniques are crucial to compensate for these differences and improve quality for both clinical assessment and AI-based analysis. We developed a customized image degradation pipeline simulating common artifacts in dermatological images, including blur, noise, downsampling, and compression. This synthetic degradation approach enabled effective training of DermaSR-GAN, a super-resolution generative adversarial network tailored for dermoscopic images. The model was trained on 30,000 high-quality ISIC images and evaluated on three independent datasets (ISIC Test, Novara Dermoscopic, PH²) using structural similarity and no-reference quality metrics. DermaSR-GAN achieved statistically significant improvements in quality scores across all datasets, with up to 23% enhancement in perceptual quality metrics (MANIQA). The model preserved diagnostic details while doubling resolution and surpassed existing approaches, including traditional interpolation methods and state-of-the-art deep learning techniques. Integration with downstream classification systems demonstrated up to 14.6% improvement in class-specific accuracy for keratosis-like lesions compared to original images. Synthetic degradation represents a promising approach for training effective super-resolution models in medical imaging, with significant potential for enhancing teledermatology applications and computer-aided diagnosis systems. Full article

(This article belongs to the Section Computer Science & Engineering)

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 1314.

Go to page 1 2 3 4 5

Search Results (65,666)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI