SLAS Technology最新文献

{"title":"An Integrated Deep Learning Framework Using Adaptive Enhanced Vision Fusion and Modified MobileNet Architecture for Precision Classification of Skin Diseases with Enhanced Diagnostic Performance.","authors":"Ahsan Bilal Tariq, Muhammad Zaheer Sajid, Nauman Ali Khan, Muhammad Fareed Hamid, Anwaar UlHaq, Jarrar Amjad","doi":"10.1016/j.slast.2025.100331","DOIUrl":"https://doi.org/10.1016/j.slast.2025.100331","url":null,"abstract":"<p><p>Due to challenges such as illumination variability, noise, and visual distortions, machine learning (ML) and deep learning (DL) approaches for skin disease evaluation remain complex. Traditional methods often neglect these issues, leading to skewed predictions and poor performance. This research leverages a diverse dataset and robust image processing techniques to enhance diagnostic accuracy under such demanding conditions. We propose Dermo-Transfer, a novel architecture that combines MobileNet with dense blocks and residual connections to improve skin disease severity classification by addressing problems such as vanishing gradients and overfitting. Our method incorporates multi-scale Retinex, gamma correction, and histogram equalization to enhance image quality and visibility. Furthermore, a quantum support vector machine (QSVM) classifier is employed to improve classification performance, providing confidence scores and effectively handling multi-class problems. The proposed approach significantly enhances diagnostic accuracy and outperforms previous models. Dermo-Transfer not only improves pattern recognition and classification accuracy but also robustly handles varying image quality and lighting conditions. Dermo-Transfer was trained on 77,314 images covering skin conditions such as molluscum, warts, eczema, psoriasis, lichen planus, seborrheic keratoses, atopic dermatitis, melanoma, basal cell carcinoma (BCC), melanocytic nevi (NV), benign keratosis, and other benign tumors. The Dermo-Transfer classification method achieved accuracies of 99%, 98.5%, 97.5%, and 89% across four datasets, demonstrating its effectiveness and potential utility for clinical diagnostics. Additionally, Dermo-Transfer outperformed SkinLesNet and MobileNet V2-LSTM in terms of classification accuracy. Experimental results also highlight how IoT devices and mobile applications can enhance the computational efficiency and practical deployment of the Dermo-Transfer model.</p>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":" ","pages":"100331"},"PeriodicalIF":2.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144669002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Construction and Validation of a Nomogram Model for Predicting Rebleeding in High-risk Peptic Ulcer Bleeding Patients Based on Lasso Regression: A Single Center Retrospective Research.","authors":"Qingrong Chen, Xuefeng Chen, Rongna You, Huaxin Huang","doi":"10.1016/j.slast.2025.100332","DOIUrl":"https://doi.org/10.1016/j.slast.2025.100332","url":null,"abstract":"<p><strong>Objective: </strong>To construct a Nomogram prediction model for high-risk Peptic Ulcer Bleeding (PUB) rebleeding using Lasso regression analysis and verify its predictive performance.</p><p><strong>Methods: </strong>Retrospective research was performed on 279 cases with PUB admitted from January 2020 to December 2023 in a hospital's medical record information system. Clinical data were collected and randomly separated into a modeling group and a validation group in a 7:3. The overfitting in the constructed model was verified by comparing the clinical data. According to the clinical data of the modeling group, Lasso regression analysis was used to screen variables and conduct multiple factor analysis. A Nomogram model was constructed accordingly, and its predictive performance was validated.</p><p><strong>Results: </strong>Among 279 patients included in this study, 45 cases had rebleeding, with an incidence rate of 16.13%. The Lasso regression analysis demonstrated that a total of 15 variables were screened, taking λ_min as the standard. Multivariate analysis showed that diastolic blood pressure, hematocrit, blood transfusion volume, GBS score, endoscopic examination, and mechanical hemostasis were all independent risk factors for rebleeding in PUB cases. The Nomogram model based on multiple factor analysis demonstrated that the AUC of the modeling group and the validation group were 0.832 (95%CI=0.744-0.921) and 0.814 (95%CI=0.672-0.956), and Hosmer-Lemeshow χ²=13.520 (P=0.095). The DCA and CIC curve analysis results showed that using this model for patient intervention achieved positive benefits and relatively accurately predicted the rebleeding in PUB patients.</p><p><strong>Conclusion: </strong>This research constructs a Nomogram model based on Lasso regression analysis that can effectively predict the rebleeding in PUB patients, providing reference for early prevention of clinical PUB rebleeding.</p>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":" ","pages":"100332"},"PeriodicalIF":2.5,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144661070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and development of a rotating hotel as an enabler for mobile robotics integration in the lab of the future","authors":"Richard Rösch ,&nbsp;Tobias Sauer ,&nbsp;Christina Mavreas,&nbsp;Stefan Scheuermann,&nbsp;Andreas Traube","doi":"10.1016/j.slast.2025.100330","DOIUrl":"10.1016/j.slast.2025.100330","url":null,"abstract":"<div><div>The development of the rotating hotel system marks a significant advancement in laboratory automation. It is particularly impactful for integrating mobile robots with closed systems like high-throughput screening (HTS) workstations, which are central to the Lab of the Future. This technical brief presents the design and integration of a modular rotating hotel unit, specifically engineered to support autonomous, flexible sample handling. With up to four SBS-compatible plate nests, customizable configurations, and built-in presence sensors, the rotating hotel enables for the first time, mobile robots to transfer samples and labware seamlessly between HTS workstations automatically, enhancing workflow efficiency and reducing manual handling.</div><div>Key use cases for the rotating hotel include continuous sample loading and unloading during HTS runs, cross-station sample retrieval and contamination-free handling for sensitive assays. The system’s compact, removable design allows for straightforward maintenance and cleaning, supporting safe and contamination-free environments. By eliminating process interruptions and enabling 24/7 operation, the rotating hotel can significantly improve throughput and flexibility, essential features for next-generation labs. This modular storage solution demonstrates how tailored automation can transform workflows, positioning the rotating hotel as a key enabling component in the adaptive, high-efficiency Lab of the Future.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100330"},"PeriodicalIF":2.5,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144651182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Magnetic Nanoparticles-Based Targeted Drug Delivery System in Tumor Pain Management.","authors":"Xiaoli Lv, Fei Wang, Xiaomei Liu, Ting Xu, Xiaofeng Zhou","doi":"10.1016/j.slast.2025.100333","DOIUrl":"https://doi.org/10.1016/j.slast.2025.100333","url":null,"abstract":"<p><strong>Background: </strong>the precise management of tumor-related pain is a critical challenge in improving the quality of life (QoL) of cancer patients. This study aimed to develop a magnetic nanoparticle-based transdermal drug delivery system (MNPs-TDDS) using green nanotechnology, incorporating folic acid targeting and magnetic-controlled release mechanisms, to achieve efficient and low-toxicity pain intervention.</p><p><strong>Methods: </strong>folic acid-modified magnetic nanocomplexes (catHEC·FA@SPIO) were synthesized via a water-phase co-precipitation method. The structural and morphological characteristics were verified using Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), and transmission electron microscopy (TEM). A total of 70 nasopharyngeal carcinoma (NPC) patients were enrolled and randomly divided into the experimental group (EG, MNPs-TDDS) and the control group (CG, conventional analgesia). Pain scores (NRS), psychological status (SAS/SDS), activities of daily living (Barthel index), and cancer cell apoptosis rates were assessed.</p><p><strong>Results: </strong>the particle size of catHEC·FA@SPIO was 150±20 nm, exhibiting pH-responsive release properties (82.4% cumulative release over 72 hours at pH 5.5). The NRS scores (2.35±0.47 vs. 4.47±0.87), SAS (41.46±1.13 vs. 55.32±1.24), and SDS (40.06±0.75 vs. 54.11±1.52) in the EG were significantly lower than those in the CG (P<0.05), with cytotoxicity to normal cells being under 10%. The nursing satisfaction rate in the EG was 94.29%, significantly higher than the 68.57% in the CG (P<0.001).</p><p><strong>Conclusion: </strong>this study is the first to combine a green synthesis strategy with multidimensional clinical evaluation, demonstrating the comprehensive advantages of MNPs-TDDS in pain relief, improving psychological state, and enhancing activities of daily living. This approach provides an innovative solution for the precise management of tumor-related pain. Future research should further validate its long-term safety and applicability across various cancer types.</p>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":" ","pages":"100333"},"PeriodicalIF":2.5,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144651191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PRIMDEx: Prototyping rapid innovation of microfluidics devices for experimentation","authors":"Patrick B. Kruk,&nbsp;Jose A. Wippold","doi":"10.1016/j.slast.2025.100326","DOIUrl":"10.1016/j.slast.2025.100326","url":null,"abstract":"<div><div>Microfluidics has quickly become an established technology in the transformative fields that make up broader biotechnology. Microfluidics has applications spanning the entire breadth of the discipline, from chemical synthesis, environmental monitoring, biomedical diagnostics, to lab- and organ-on-a-chip. New demands for novel microfluidic chips have outpaced their contemporary manufacturing methods, thus limiting their scientific applicability. This predicament is particularly accentuated for R&amp;D and research laboratories where resources (time &amp; money) are limited. Manufacturing a microfluidic device (MFD) for mass production typically involves outsourcing a design for CNC machining of the negative mold, followed by Injection Molding (IM) the positive-feature consumables or MFDs. This process can cost ∼$1000-$5000 depending on complexity and can require a 1–2-week lead time. In comparison, 3D Printing (3DP) is limited by long print times, limited resolutions, and higher per unit material cost. This leaves traditional commercial fabrication processes impractical to implement into a typical biotech experimental procedure, where they could be subjected to constantly changing experimental demands and redesigns. Each redesign and subsequent round of fabrication demands greater cost and time investments. Here, we present PRIMDEx, or <u>P</u>rototyping <u>R</u>apid <u>I</u>nnovation of <u>M</u>icrofluidic <u>D</u>evices for <u>Ex</u>perimentation, to address this by integrating both 3DP and rapid IM into a single manufacturing workflow. PRIMDEx implemented the advantages of both manufacturing methods to establish an approach more conducive to the design-test-build cycles of biotech and biomedical research regimes.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100326"},"PeriodicalIF":2.5,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144576957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data acquisition of exercise and fitness pressure measurement based on artificial intelligence technology","authors":"Ru Liu ,&nbsp;Wenxi Shen","doi":"10.1016/j.slast.2025.100328","DOIUrl":"10.1016/j.slast.2025.100328","url":null,"abstract":"<div><div>This project aims to improve the accuracy of fitness and physical pressure ratings, focusing on basketball, by integrating artificial intelligence (AI) into data collection and training. Athletes and fitness fanatics can benefit greatly from the data collected using complex AI algorithms to determine stress levels. This study employs the Intelligent Physiological Monitoring Framework for Exercise and Fitness Pressure Measurement (IPM-EFPM) to perform automated stress tests that employ AI to enhance the precision of exercise and fitness pressure measurements. Basketball training programs can benefit from this framework's utilization of state-of-the-art technology, meticulous monitoring of exercise-induced stress, and continuous validation and improvement. The IPM-EFPM system gathers data from wearable sensors, uses real-time location systems, and employs artificial intelligence's Long Short-Term Memory (LSTM) and machine learning algorithms to uncover new insights in healthcare and sports. To accurately record fitness strain, physical activity, exercise-induced stress, and sports like basketball, this system employs cutting-edge artificial intelligence technologies, such as wearable sensors and current gathering data methods. Placement of sensors, real-time data collecting, data preprocessing and integrating, evaluation of stress by artificial intelligence algorithms, discovery and application of new information, validation and improvement are all parts of an iterative method that has been fine-tuned for use in sports and fitness settings by the IPM-EFPM. Examining the intricate relationship between AI, physical activity, and psychological stress is the main objective of this research. This could have real-world uses tailored to the sports world, particularly for basketball players.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100328"},"PeriodicalIF":2.5,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144576956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and application of a customized online affinity selection mass spectrometry screening platform","authors":"Jun Zhang,&nbsp;Jerry Barney,&nbsp;Qingqing Shen,&nbsp;Anthony Paiva,&nbsp;Wilson Shou,&nbsp;Chris Barbieri,&nbsp;Nancy Huynh,&nbsp;Heidi L. Perez,&nbsp;Andrew F. Donnell,&nbsp;Cullen L. Cavallaro","doi":"10.1016/j.slast.2025.100329","DOIUrl":"10.1016/j.slast.2025.100329","url":null,"abstract":"<div><div>With the evolving landscape of small molecule modalities and drug target portfolios, affinity selection mass spectrometry (ASMS) has emerged as a preferred high-throughput screening approach, driven by the growing chemical space capable of regulating historically undruggable targets. To meet the ever-increasing demand for binding screens, the ASMS platform requires continuous enhancements in efficiency and scalability. In this study, we developed a customized online ASMS platform, featuring multiplexed two-dimensional LC/MS for sample analysis, ultrafast acoustic ejection mass spectrometry for reaction product scouting of high-throughput chemistry before subjecting them to direct ASMS screens, and an integrated tool for real-time data processing and cross-hit evaluation. With the quality controls we implemented to ensure operational robustness and quality, the multiplexed system has demonstrated significantly enhanced analytical speed, efficiency, and overall capacity. The integrated data processing tool enhanced data quality by incorporating advanced cross-hit assessments to reduce false positives, and streamlined the screening workflow through real-time data analysis, effectively eliminating the traditional bottleneck in ASMS screening. The established ASMS platform has been utilized for our internal screening campaigns generating tractable binding hits, and providing direct affinity ranking of hits from nanoscale synthesis, enabling hit optimization and chemotype expansion with demonstrated high platform performance.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100329"},"PeriodicalIF":2.5,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144561960","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect and toxicity of PF chemotherapy combined with radiotherapy in the treatment of advanced cervical cancer: Medical thermography test","authors":"Caixia Gou ,&nbsp;Wang Qi ,&nbsp;Pengbing Han ,&nbsp;Chunlin Zhang","doi":"10.1016/j.slast.2025.100327","DOIUrl":"10.1016/j.slast.2025.100327","url":null,"abstract":"<div><div>Cancer is one of the leading causes of death worldwide, and treatment options and prognosis for patients with advanced cancer are particularly challenging. PF chemotherapy regimen (the combination of cisplatin and 5-fluorouracil) has been widely used in the treatment of a variety of cancers, and radiotherapy as a local treatment can effectively control tumor growth. This is a prospective clinical trial in which patients were treated with PF chemotherapy combined with radiotherapy. Medical thermal imaging was performed on all patients before, during and after treatment. The examination process involves recording the patient's body surface temperature using a highly sensitive infrared camera and analyzing temperature changes in the tumor area and surrounding tissue. Clinical data on patients were also collected, including treatment response, quality of life scores, and reports of toxic and side effects. Preliminary results showed that PF chemotherapy combined with radiotherapy showed a positive effect in controlling tumor growth, and most patients experienced a reduction in tumor volume. Medical thermal image examination revealed significant changes in tumor area temperature during treatment, which correlated with tumor reactivity. In some cases, thermal imagery shows potential skin and mucosal damage in advance, suggesting the need for early intervention. Thermal imagery also helped assess the impact of treatment on patients' quality of life, such as pain and discomfort by looking at changes in the patient's body surface temperature distribution.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100327"},"PeriodicalIF":2.5,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144561961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Classifying kidney disease using a dense layers deep learning model","authors":"Amal Al-Rasheed ,&nbsp;Sheikh Muhammad Saqib ,&nbsp;Muhammad Zubair Asghar ,&nbsp;Tehseen Mazhar ,&nbsp;Asim Seedahmed Ali Osman ,&nbsp;Mohammad Shahid ,&nbsp;Muhammad Iqbal ,&nbsp;Muhammad Amir Khan","doi":"10.1016/j.slast.2025.100324","DOIUrl":"10.1016/j.slast.2025.100324","url":null,"abstract":"<div><div>Early diagnosis and thorough management techniques are crucial for people with chronic kidney disease (CKD), a crippling and potentially fatal condition. Research has focused a lot on machine learning and deep learning systems for the detection of kidney diseases. Deep learning platforms like hidden layers, activation functions, optimizers, and epochs are also necessary for the automatic detection of these diseases. The proposed model achieved 99 % accuracy, with a precision, recall, and F1 score of 0.99, indicating highly reliable performance. Additionally, the model demonstrated strong agreement and robustness, as reflected in metrics such as the ROC AUC score of 0.9821 and Matthews Correlation Coefficient of 0.9727. The experiment used a publicly accessible dataset with 24 independent fields and independent values as chronic or not-chronic classes, building dense-layered deep neural networks based on an optimized architecture. The outcomes demonstrated that, when compared to the other models, the proposed model was the most accurate.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100324"},"PeriodicalIF":2.5,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144531106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging FastViT based knowledge distillation with EfficientNet-B0 for diabetic retinopathy severity classification","authors":"Jyotirmayee Rautaray ,&nbsp;Ali B.M. Ali ,&nbsp;Meenakshi Kandpal ,&nbsp;Pranati Mishra ,&nbsp;Rzgar Farooq Rashid ,&nbsp;Farzona Alimova ,&nbsp;Mohamed Kallel ,&nbsp;Nadia Batool","doi":"10.1016/j.slast.2025.100325","DOIUrl":"10.1016/j.slast.2025.100325","url":null,"abstract":"<div><div>Diabetic retinopathy (DR) remains a key contributor to eye impairment worldwide, requiring the development of efficient and accurate deep learning models for automated diagnosis. This study presents FastEffNet, a novel framework that leverages transformer-based knowledge distillation (KD) to enhance DR severity classification while reducing computational complexity. The proposed approach employs FastViT-MA26 as the teacher model and EfficientNet-B0 as the student model, striking the ideal mix between accuracy and computational efficiency. APTOS blindness detection dataset comprising 3662 images across five severity classes is collected, pre-processed, normalized, split and augmented to address class imbalance. The teacher model undergoes training and validation before transferring its knowledge to the student model, enabling the latter to approximate the teacher’s performance while maintaining a lightweight architecture. To comprehensively assess the efficacy of the proposed framework, additional student models—including HGNet, ResNet50, MobileNetV3, and DeiT—are analysed for comparative assessment. Model interpretability is enhanced through Grad-CAM++ visualizations, which highlight critical retinal regions influencing DR severity classification. Several measures are used to evaluate performance, including accuracy, precision, recall, F1-score, Cohen’s Kappa Score (CKS), Weighted Kappa Score (WKS), and Matthews Correlation Coefficient (MCC), ensuring a robust assessment. Among all student models, EfficientNet-B0 achieves the highest classification accuracy of 95.39 %, 95.43 % precision, recall of 95.39 %, F1-score of 95.37 %, CKS of 0.94, WKS of 0.97, MCC of 0.94, AUC of 0.99, and a KD loss of 0.17, with a computational cost of 0.38 G FLOPs. These results demonstrate its effectiveness as an optimized lightweight model for DR detection. The findings emphasize the potential of KD-based lightweight models in attaining high diagnostic accuracy while reducing computational complexity, paving the way for scalable and cost-effective DR screening solutions.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"33 ","pages":"Article 100325"},"PeriodicalIF":2.5,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144523640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}