Current Medical Imaging Reviews最新文献

{"title":"Integration of Three-dimensional Visualization Reconstruction Technology with Problem-Based Learning in the Clinical Training of Resident Physicians Specialized in Pheochromocytoma.","authors":"Dong Wang","doi":"10.2174/0115734056327236250101052226","DOIUrl":"10.2174/0115734056327236250101052226","url":null,"abstract":"<p><strong>Objective: </strong>We examined the effectiveness of integrating three-dimensional (3D) visualization reconstruction technology with Problem-Based Learning (PBL) in the clinical teaching of resident physicians focusing on pheochromocytoma.</p><p><strong>Methods: </strong>Fifty resident physicians specializing in urology at Peking Union Medical College Hospital were randomly divided into two groups over the period spanning January 2022 to January 2024: an experimental group and a control group. The experimental group underwent instruction utilizing a pedagogical approach that integrated 3D visualization reconstruction technology with PBL, while the control group used a traditional teaching model. A comparative analysis of examination performance and teaching satisfaction between both groups of resident physicians was conducted to assess the efficacy of the integrated 3D visualization and PBL teaching methods in clinical instruction.</p><p><strong>Results: </strong>The experimental group demonstrated superior performance in both theoretical assessment and clinical skills evaluation, along with heightened levels of teaching satisfaction compared to the control group, with statistically significant differences (p < 0.05). Additionally, the experimental group exhibited markedly higher scores in both theoretical examinations and practical assessments compared to their counterparts in the control group (p < 0.05). The results of theoretical examinations for the experimental group and the control group were 92.15±3.22 and 81.09±4.46, respectively (< 0.0001). The results of practical examinations for the experimental group and the control group were 90.17±3.48 and 70.75±5.11, respectively (< 0.0001).</p><p><strong>Conclusion: </strong>In the clinical teaching of training resident physicians specializing in urology for the management of pheochromocytoma, the integration of 3D visualization reconstruction technology with the PBL method significantly enhanced the teaching efficacy, improving both the quality of instruction and the level of satisfaction among participants.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":" ","pages":"e15734056327236"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143588084","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":"Intracranial Structural Malformations in Children in Tibet: CT and MRI Findings in a Single Tertiary Center.","authors":"Xuan Yin, Dawa Ciren, Ciren Guojie, Guofu Zhang, Jimei Wang, He Zhang","doi":"10.2174/0115734056321642241213103658","DOIUrl":"10.2174/0115734056321642241213103658","url":null,"abstract":"<p><strong>Objectives: </strong>The objective of this study was to summarize the findings of children's intracranial congenital or developmental malformations found during imaging procedures in the Tibetan plateau.</p><p><strong>Methods: </strong>We retrospectively reviewed the imaging data of the suspected patients who presented with the central nervous system (CNS) malformations and were enrolled either through the clinic or after ultrasound examinations between June 2019 and June 2023 in our institution. All imaging data were interpreted by two experienced radiologists through consensus reading.</p><p><strong>Results: </strong>In this study, we recruited 36 patients, including two neonates, 17 infants and 17 children. Seven cases underwent an MRI examination, while the others had a CT scan. Polygyria and pachygyria malformation were the most common type of congenital neurological malformations (7 cases, 31.8%), followed by cystic changes of the cerebral parenchyma (3 cases, 13.6%). Cerebral atrophy was the most common type of secondary CNS abnormality(8 cases, 57.1%), followed by communicative hydrocephalus (3 cases, 21.4%). Five patients in the congenital group and 4 patients in the secondary group had complex malformations. In the current study group, there were 8 deaths, 12 cases with neurological sequelae, 1 case with normal development, and 15 cases lost to follow-up. There were no significant differences between the primary and secondary CNS groups in terms of the outcome for both the infants and children groups.</p><p><strong>Conclusions: </strong>CNS malformations in the Tibetan Plateau are associated with high mortality and morbidity rates. Better utilization of imaging modalities could help design tailored treatments as early as possible.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":" ","pages":"e15734056321642"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142933650","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":"Preoperative CT-based Intratumoral and Peritumoral Radiomics Prediction for Vasculogenic Mimicry in Lung Adenocarcinoma.","authors":"Shuhua Li, Yang Li, Ying Meng, Jingcheng Huang, Yihong Gu, Yan Song, Shuni Zhang, Zhiya Zhang, Weiming Zhao, Zongyu Xie","doi":"10.2174/0115734056383032250320041531","DOIUrl":"https://doi.org/10.2174/0115734056383032250320041531","url":null,"abstract":"<p><strong>Objective: </strong>This study seeks to assess vasculogenic mimicry (VM) occurrence in lung adenocarcinoma (LUAD) by delineating intratumoral and peritumoral characteristics using preoperative CT-based radiomics and a nomogram for enhanced precision.</p><p><strong>Materials and methods: </strong>Our retrospective analysis enrolled 150 LUAD patients, ascertained their VM status, and stratified them randomly into development (n=105) and validation cohorts. We extracted radiomics features from intra- and peritumoral zones, delineating 3, 5, and 7mm expansions on thin-section chest CT images. We formulated logistic models encompassing a clinical model (CM), intratumoral radiomics model (TRM), peritumoral radiomics models at 3, 5, and 7 mm (PRMs), and a composite model integrating both intra- and peritumoral zones (CRM). A radiomics nomogram model (RNM) was devised, amalgamating the Rad-scores from intra- and peritumoral regions with clinical-radiological traits to forecast VM. The models' efficacy was gauged via the receiver operating characteristic (ROC) curve analysis, calibration assessment, and decision curve analysis (DCA).</p><p><strong>Results: </strong>The CRM outperformed its counterparts, the TRM, PRM_3mm, PRM_5mm, and PRM_7mm models, with AUCs reaching 0.859 and 0.860 in the development and validation cohorts. Within the CM, tumor size and spiculation emerged as significant predictive covariates. The RNM, integrating independent predictors with the CRM-Rad-score, demonstrated clinical utility, achieving AUCs of 0.903 and 0.931 in the respective cohorts.</p><p><strong>Conclusion: </strong>Our findings underscore the potential of CT-based radiomics characteristics derived from intratumoral and peritumoral regions to assess VM presence in LUAD patients. Combining radiomics signatures with clinicoradiological parameters within a nomogram framework significantly enhances predictive accuracy.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":"21 ","pages":"e15734056383032"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144040061","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":"Enhancing Medical Image Classification through Transfer Learning and CLAHE Optimization.","authors":"Kamal Halloum, Hamid Ez-Zahraouy","doi":"10.2174/0115734056342623241119061744","DOIUrl":"https://doi.org/10.2174/0115734056342623241119061744","url":null,"abstract":"<p><strong>Introduction: </strong>This paper examines the impact of transfer learning and CLAHE (Contrast Limited Adaptive Histogram Equalization) optimization on the classification of medical images, specifically brain images.</p><p><strong>Methods: </strong>Four different setups were tested: normal images without data augmentation, normal images with data augmentation, CLAHE-processed images without data augmentation, and CLAHE-processed images with data augmentation.</p><p><strong>Results: </strong>The results show that using CLAHE combined with data augmentation significantly improves classification accuracy. Specifically, the combination of CLAHE and data augmentation achieved a precision of 0.90, a recall of 0.87, an F1-score of 0.89, and an accuracy of 0.86, outperforming the other setups.</p><p><strong>Conclusion: </strong>These findings highlight the effectiveness of CLAHE optimization in the context of transfer learning, particularly when data augmentation techniques are also applied, leading to an overall improvement in the performance of brain image classification models.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":"21 ","pages":"e15734056342623"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144053444","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":"Automated 3D Quantitative Analysis of Intrapulmonary Vessel Volume on Noncontrast CT in Healthy Individuals.","authors":"Ying Ming, Yu Zhang, Ran Xiao, Ruijie Zhao, Jiaru Wang, Sirong Piao, Lan Song, Yinghao Xu, Xin Sui, Wei Song","doi":"10.2174/0115734056354924241115102310","DOIUrl":"https://doi.org/10.2174/0115734056354924241115102310","url":null,"abstract":"<p><strong>Objective: </strong>This study aimed to compare automated three-dimensional Intrapulmonary Vessel Volume (IPVV) differences between lung and mediastinal windows in healthy individuals using quantitative measurements obtained from chest Computed Tomography (CT) plain scans.</p><p><strong>Methods: </strong>A total of 258 participants (aged 21-83 years) with negative chest CT scans from routine physical examinations conducted between January to November 2023 were retrospectively enrolled. For each healthy participant, an algorithm was used to automatically extract total lung IPVVs as well as IPVVs for vessels of specific diameter. Differences in IPVVs were then compared between those extracted using the lung window and those extracted using the mediastinal window.</p><p><strong>Results: </strong>The IPVVs for the entire lung, intrapulmonary arteries, intrapulmonary veins, and small pulmonary vessels (categorized by different diameters) extracted from the lung window were significantly higher than those extracted from the mediastinal window (p<0.01). No significant sex-based differences in IPVV were observed for pulmonary arteries and veins with diameters between 0.8 and 1.6 mm, as well as pulmonary veins with diameters between 2.4 and 3.2 mm. However, in pulmonary arteries and veins with diameters between 1.6 and 2.4 mm, females had significantly higher IPVVs than males. In all other cases, IPVVs were larger in males than in females.</p><p><strong>Conclusion: </strong>This method of automatic IPVV extraction and quantitative assessment has been proven to be feasible. Automated IPVV expression effectively identified morphological characteristics of intrapulmonary vessels. The study has concluded IPVVs extracted from the lung window to be generally larger than those extracted from the mediastinal window.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":"21 ","pages":"e15734056354924"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144009617","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":"Attention-aware Deep Learning Models for Dermoscopic Image Classification for Skin Disease Diagnosis.","authors":"Malliga Subramanian, Kogilavani Shanmugavadivel, Sudha Thangaraj, Jaehyuk Cho, Sathishkumar Ve","doi":"10.2174/0115734056332443241129113146","DOIUrl":"https://doi.org/10.2174/0115734056332443241129113146","url":null,"abstract":"<p><strong>Background: </strong>The skin, being the largest organ in the human body, plays a vital protective role. Skin lesions are changes in the appearance of the skin, such as bumps, sores, lumps, patches, and discoloration. If not identified and treated promptly, skin lesion diseases would become a serious and worrisome problem for society due to their detrimental effects. However, visually inspecting skin lesions during medical examinations can be challenging due to their similarities.</p><p><strong>Objective: </strong>The proposed research aimed at leveraging technological advancements, particularly deep learning methods, to analyze dermoscopic images of skin lesions and make accurate predictions, thereby aiding in diagnosis.</p><p><strong>Methods: </strong>The proposed study utilized four pre-trained CNN architectures, RegNetX, EfficientNetB3, VGG19, and ResNet-152, for the multi-class classification of seven types of skin diseases based on dermoscopic images. The significant finding of this study was the integration of attention mechanisms, specifically channel-wise and spatial attention, into these CNN variants. These mechanisms allowed the models to focus on the most relevant regions of the dermoscopic images, enhancing feature extraction and improving classification accuracy. Hyperparameters of the models were optimized using Bayesian optimization, a probabilistic model-based technique that efficiently uses the hyperparameter space to find the optimal configuration for the developed models.</p><p><strong>Results: </strong>The performance of these models was evaluated, and it was found that RegNetX outperformed the other models with an accuracy of 98.61%. RegNetX showed robust performance when integrated with both channel-wise and spatial attention mechanisms, indicating its effectiveness in focusing on significant features within the dermoscopic images.</p><p><strong>Conclusion: </strong>The results demonstrated the effectiveness of attention-aware deep learning models in accurately classifying various skin diseases from dermoscopic images. By integrating attention mechanisms, these models can focus on the most relevant features within the images, thereby improving their classification accuracy. The results confirmed that RegNetX, integrated with optimized attention mechanisms, can provide robust, accurate diagnoses, which is critical for early detection and treatment of skin diseases.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":"21 ","pages":"e15734056332443"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144018389","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":"Detection of Sub-acute Brain Injury and Hypoxic-ischemic Encephalopathy using I2C2-WGO and CO-GW-RNN.","authors":"Priyan Malarvizhi Kumar, Wael Korani, Tayyaba Shahwar, Gokulnath C","doi":"10.2174/0115734056352573241118122026","DOIUrl":"https://doi.org/10.2174/0115734056352573241118122026","url":null,"abstract":"<p><strong>Background: </strong>Hypoxic-ischemic encephalopathy (HIE) is a brain injury that is caused by improper oxygen/blood flow. None of the existing works have concentrated on detecting HIE based on the sub-acute injury in the brain.</p><p><strong>Objective: </strong>To enhance the accuracy and specificity of HIE detection, a comprehensive approach that includes SAI identification, BGT segmentation, and volume calculation will be utilized.</p><p><strong>Methods: </strong>This study addresses the critical challenge of detecting hypoxic-schemic encephalopathy (HIE) through advanced image processing techniques applied to brain MRI data. The primary research questions focus on the effectiveness of the proposed CO-GW-RNN method in accurately identifying HIE and the impact of incorporating segmentation and clustering processes on detection performance.</p><p><strong>Results: </strong>The proposed method achieved remarkable results, demonstrating an accuracy of 98.98% and a specificity of 98.17%, significantly outperforming existing techniques such as the RUB classifier (84.6% accuracy) and the DTL method (94.00% accuracy).</p><p><strong>Conclusion: </strong>These findings validate the effectiveness of the proposed methodology in improving HIE detection in brain MRI images.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":"21 ","pages":"e15734056352573"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144055658","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":"Lightweight Lung-nodule Detection Model Combined with Multidimensional Attention Convolution.","authors":"He-He Huang, Yuetao Zhao, Sen-Yu Wei, Chen Zhao, Yu Shi, Yuan Li, Weijia Huang, Yifei Yang, Jianhua Xu","doi":"10.2174/0115734056310722241210055412","DOIUrl":"10.2174/0115734056310722241210055412","url":null,"abstract":"<p><strong>Background: </strong>Early and timely detection of pulmonary nodules and initiation treatment can substantially improve the survival rate of lung carcinoma. However, current detection methods based on convolutional neural networks (CNNs) cannot easily detect pulmonary nodules owing to low detection accuracy and the difficulty in detecting small-sized pulmonary nodules; meanwhile, more accurate CNN-based models are slow and require high hardware specifications.</p><p><strong>Objective: </strong>The aim of this study is to develop a detection model that achieves both high accuracy and real-time performance, ensuring effective and timely results.</p><p><strong>Methods: </strong>In this study, based on YOLOv5s, a concentrated-comprehensive convolution (C3_ODC) module with multidimensional attention is designed in the convolutional layer of the original backbone network for enhancing the feature-extraction capabilities of the model. Moreover, lightweight convolution is combined with weighted bidirectional feature pyramid networks (BiFPNs) to form a GS-BiFPN structure that enhances the fusion of multiscale features while reducing the number of model parameters. Finally, Focal Loss is combined with the normalized Wasserstein distance (NWD) to optimize the loss function. Focal loss focuses on carcinoma-positive samples to mitigate class imbalance, whereas the NWD enhances the detection performance of small lung nodules.</p><p><strong>Results: </strong>In comparison experiments against the YOLOv5s, the proposed model improved the average precision by 8.7% and reduced the number of parameters and floating-point operations by 5.4% and 8.2%, respectively, while achieving 116.7 frames per second.</p><p><strong>Conclusion: </strong>The proposed model balances high detection accuracy against real-time requirements.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":" ","pages":"e15734056310722"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142933651","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":"Sound Touch Viscosity (STVi) for Thyroid Gland Evaluation in Healthy Individuals: A Pilot Study : STVi for Thyroid Gland Evaluation.","authors":"Feng Mao, Yuemingming Jiang, Yunzhong Wang, Zhenbin Xu, Zhuo Wei, Xueli Zhu, Libin Chen, Shengmin Zhang","doi":"10.2174/0115734056335791241202115022","DOIUrl":"10.2174/0115734056335791241202115022","url":null,"abstract":"<p><strong>Objective: </strong>This prospective study aimed to establish the typical viscosity range of the thyroid gland in healthy individuals using a new method called the Sound Touch Viscosity (STVi) technique with a linear array transducer.</p><p><strong>Methods: </strong>Seventy-eight healthy volunteers were enrolled between March, 2023 and April, 2023. Thyroid viscosity was measured using the Resona R9 ultrasound system equipped with a linear array transducer (L15-3WU). Each patient had three valid viscosity measurements taken for each thyroid lobe, and the average values were analyzed. Thyroid gland stiffness was measured and analyzed simultaneously.</p><p><strong>Results: </strong>The study included 51 women and 27 men with an average age of 48 years. The mean viscosity measurement for a normal thyroid gland was 1.10 ± 0.41 Pa.s (ranging from 0.38 to 2.25 Pa.s). There were no significant differences in viscosity between the left and right lobes of the thyroid gland. We found no significant variations in viscosity based on gender, age, or body mass index (BMI). There was a notable positive correlation between thyroid viscosity and stiffness measurements (r = 0.717, p < 0.001).</p><p><strong>Conclusion: </strong>Our findings suggest that STVi is a highly reliable method for assessing the thyroid. This technique holds promise as a new, non-invasive approach to evaluating thyroid parenchyma viscosity.</p>","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":" ","pages":""},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142933596","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":"Enhanced Pneumonia Detection in Chest X-Rays Using Hybrid Convolutional and Vision Transformer Networks.","authors":"Benzorgat Mustapha, Yatong Zhou, Chunyan Shan, Zhitao Xiao","doi":"10.2174/0115734056326685250101113959","DOIUrl":"10.2174/0115734056326685250101113959","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;The objective of this research is to enhance pneumonia detection in chest X-rays by leveraging a novel hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with modified Swin Transformer blocks. This study aims to significantly improve diagnostic accuracy, reduce misclassifications, and provide a robust, deployable solution for underdeveloped regions where access to conventional diagnostics and treatment is limited.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The study developed a hybrid model architecture integrating CNNs with modified Swin Transformer blocks to work seamlessly within the same model. The CNN layers perform initial feature extraction, capturing local patterns within the images. At the same time, the modified Swin Transformer blocks handle long-range dependencies and global context through window-based self-attention mechanisms. Preprocessing steps included resizing images to 224x224 pixels and applying Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance image features. Data augmentation techniques, such as horizontal flipping, rotation, and zooming, were utilized to prevent overfitting and ensure model robustness. Hyperparameter optimization was conducted using Optuna, employing Bayesian optimization (Tree-structured Parzen Estimator) to fine-tune key parameters of both the CNN and Swin Transformer components, ensuring optimal model performance.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The proposed hybrid model was trained and validated on a dataset provided by the Guangzhou Women and Children's Medical Center. The model achieved an overall accuracy of 98.72% and a loss of 0.064 on an unseen dataset, significantly outperforming a baseline CNN model. Detailed performance metrics indicated a precision of 0.9738 for the normal class and 1.0000 for the pneumonia class, with an overall F1-score of 0.9872. The hybrid model consistently outperformed the CNN model across all performance metrics, demonstrating higher accuracy, precision, recall, and F1-score. Confusion matrices revealed high sensitivity and specificity with minimal misclassifications.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;The proposed hybrid CNN-ViT model, which integrates modified Swin Transformer blocks within the CNN architecture, provides a significant advancement in pneumonia detection by effectively capturing both local and global features within chest X-ray images. The modifications to the Swin Transformer blocks enable them to work seamlessly with the CNN layers, enhancing the model's ability to understand complex visual patterns and dependencies. This results in superior classification performance. The lightweight design of the model eliminates the need for extensive hardware, facilitating easy deployment in resource-constrained settings. This innovative approach not only improves pneumonia diagnosis but also has the potential to enhance patient outcomes and support healthcare providers in underdeveloped regions. Fu","PeriodicalId":54215,"journal":{"name":"Current Medical Imaging Reviews","volume":" ","pages":"e15734056326685"},"PeriodicalIF":1.1,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142980835","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}