Journal of imaging informatics in medicine最新文献

{"title":"Res-TransNet: A Hybrid deep Learning Network for Predicting Pathological Subtypes of lung Adenocarcinoma in CT Images.","authors":"Yue Su, Xianwu Xia, Rong Sun, Jianjun Yuan, Qianjin Hua, Baosan Han, Jing Gong, Shengdong Nie","doi":"10.1007/s10278-024-01149-z","DOIUrl":"10.1007/s10278-024-01149-z","url":null,"abstract":"<p><p>This study aims to develop a CT-based hybrid deep learning network to predict pathological subtypes of early-stage lung adenocarcinoma by integrating residual network (ResNet) with Vision Transformer (ViT). A total of 1411 pathologically confirmed ground-glass nodules (GGNs) retrospectively collected from two centers were used as internal and external validation sets for model development. 3D ResNet and ViT were applied to investigate two deep learning frameworks to classify three subtypes of lung adenocarcinoma namely invasive adenocarcinoma (IAC), minimally invasive adenocarcinoma and adenocarcinoma in situ, respectively. To further improve the model performance, four Res-TransNet based models were proposed by integrating ResNet and ViT with different ensemble learning strategies. Two classification tasks involving predicting IAC from Non-IAC (Task1) and classifying three subtypes (Task2) were designed and conducted in this study. For Task 1, the optimal Res-TransNet model yielded area under the receiver operating characteristic curve (AUC) values of 0.986 and 0.933 on internal and external validation sets, which were significantly higher than that of ResNet and ViT models (p < 0.05). For Task 2, the optimal fusion model generated the accuracy and weighted F1 score of 68.3% and 66.1% on the external validation set. The experimental results demonstrate that Res-TransNet can significantly increase the classification performance compared with the two basic models and have the potential to assist radiologists in precision diagnosis.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"2883-2894"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612082/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141302293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiomics Nomogram Derived from Gated Myocardial Perfusion SPECT for Identifying Ischemic Cardiomyopathy.","authors":"Chunqing Zhou, Yi Xiao, Longxi Li, Yanyun Liu, Fubao Zhu, Weihua Zhou, Xiaoping Yi, Min Zhao","doi":"10.1007/s10278-024-01145-3","DOIUrl":"10.1007/s10278-024-01145-3","url":null,"abstract":"<p><p>Personalized management involving heart failure (HF) etiology is crucial for better prognoses. We aim to evaluate the utility of a radiomics nomogram based on gated myocardial perfusion imaging (GMPI) in distinguishing ischemic from non-ischemic origins of HF. A total of 172 heart failure patients with reduced left ventricular ejection fraction (HFrEF) who underwent GMPI scan were divided into training (n = 122) and validation sets (n = 50) based on chronological order of scans. Radiomics features were extracted from the resting GMPI. Four machine learning algorithms were used to construct radiomics models, and the model with the best performances were selected to calculate the Radscore. A radiomics nomogram was constructed based on the Radscore and independent clinical factors. Finally, the model performance was validated using operating characteristic curves, calibration curve, decision curve analysis, integrated discrimination improvement values (IDI), and the net reclassification index (NRI). Three optimal radiomics features were used to build a radiomics model. Total perfusion deficit (TPD) was identified as the independent factors of conventional GMPI metrics for building the GMPI model. In the validation set, the radiomics nomogram integrating the Radscore, age, systolic blood pressure, and TPD significantly outperformed the GMPI model in distinguishing ischemic cardiomyopathy (ICM) from non-ischemic cardiomyopathy (NICM) (AUC 0.853 vs. 0.707, p = 0.038). IDI analysis indicated that the nomogram improved diagnostic accuracy by 28.3% compared to the GMPI model in the validation set. By combining radiomics signatures with clinical indicators, we developed a GMPI-based radiomics nomogram that helps to identify the ischemic etiology of HFrEF.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"2784-2793"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612043/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141163134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DSTAN: A Deformable Spatial-temporal Attention Network with Bidirectional Sequence Feature Refinement for Speckle Noise Removal in Thyroid Ultrasound Video.","authors":"Jianning Chi, Jian Miao, Jia-Hui Chen, Huan Wang, Xiaosheng Yu, Ying Huang","doi":"10.1007/s10278-023-00935-5","DOIUrl":"10.1007/s10278-023-00935-5","url":null,"abstract":"<p><p>Thyroid ultrasound video provides significant value for thyroid diseases diagnosis, but the ultrasound imaging process is often affected by the speckle noise, resulting in poor quality of the ultrasound video. Numerous video denoising methods have been proposed to remove noise while preserving texture details. However, existing methods still suffer from the following problems: (1) relevant temporal features in the low-contrast ultrasound video cannot be accurately aligned and effectively aggregated by simple optical flow or motion estimation, resulting in the artifacts and motion blur in the video; (2) fixed receptive field in spatial features integration lacks the flexibility of aggregating features in the global region of interest and is susceptible to interference from irrelevant noisy regions. In this work, we propose a deformable spatial-temporal attention denoising network to remove speckle noise in thyroid ultrasound video. The entire network follows the bidirectional feature propagation mechanism to efficiently exploit the spatial-temporal information of the whole video sequence. In this process, two modules are proposed to address the above problems: (1) a deformable temporal attention module (DTAM) is designed after optical flow pre-alignment to further capture and aggregate relevant temporal features according to the learned offsets between frames, so that inter-frame information can be better exploited even with the imprecise flow estimation under the low contrast of ultrasound video; (2) a deformable spatial attention module (DSAM) is proposed to flexibly integrate spatial features in the global region of interest through the learned intra-frame offsets, so that irrelevant noisy information can be ignored and essential information can be precisely exploited. Finally, all these refined features are rectified and merged through residual convolution blocks to recover the clean video frames. Experimental results on our thyroid ultrasound video (US-V) dataset and the DDTI dataset demonstrate that our proposed method exceeds 1.2 <math><mo>∼</mo></math> 1.3 dB on PSNR and has clearer texture detail compared to other state-of-the-art methods. In the meantime, the proposed model can also assist thyroid nodule segmentation methods to achieve more accurate segmentation effect, which provides an important basis for thyroid diagnosis. In the future, the proposed model can be improved and extended to other medical image sequence datasets, including CT and MRI slice denoising. The code and datasets are provided at https://github.com/Meta-MJ/DSTAN .</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"3264-3281"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612081/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141263524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identification of Prolactinoma in Pituitary Neuroendocrine Tumors Using Radiomics Analysis Based on Multiparameter MRI.","authors":"Hongxia Li, Zhiling Liu, Fuyan Li, Yuwei Xia, Tong Zhang, Feng Shi, Qingshi Zeng","doi":"10.1007/s10278-024-01153-3","DOIUrl":"10.1007/s10278-024-01153-3","url":null,"abstract":"<p><p>This study aims to investigate the feasibility of preoperatively predicting histological subtypes of pituitary neuroendocrine tumors (PitNETs) using machine learning and radiomics based on multiparameter MRI. Patients with PitNETs from January 2016 to May 2022 were retrospectively enrolled from four medical centers. A cfVB-Net network was used to automatically segment PitNET multiparameter MRI. Radiomics features were extracted from the MRI, and the radiomics score (Radscore) of each patient was calculated. To predict histological subtypes, the Gaussian process (GP) machine learning classifier based on radiomics features was performed. Multi-classification (six-class histological subtype) and binary classification (PRL vs. non-PRL) GP model was constructed. Then, a clinical-radiomics nomogram combining clinical factors and Radscores was constructed using the multivariate logistic regression analysis. The performance of the models was evaluated using receiver operating characteristic (ROC) curves. The PitNET auto-segmentation model eventually achieved the mean Dice similarity coefficient of 0.888 in 1206 patients (mean age 49.3 ± SD years, 52% female). In the multi-classification model, the GP of T2WI got the best area under the ROC curve (AUC), with 0.791, 0.801, and 0.711 in the training, validation, and external testing set, respectively. In the binary classification model, the GP of T2WI combined with CE T1WI demonstrated good performance, with AUC of 0.936, 0.882, and 0.791 in training, validation, and external testing sets, respectively. In the clinical-radiomics nomogram, Radscores and Hardy' grade were identified as predictors for PRL expression. Machine learning and radiomics analysis based on multiparameter MRI exhibited high efficiency and clinical application value in predicting the PitNET histological subtypes.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"2865-2873"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612092/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141285815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Myocardial Transit Time Mapping by CMR: A Novel Indicator of Microcirculatory Dysfunction in Cardiac Amyloidosis.","authors":"Jinxiu Yang, Zhen Wang, Huimin Wang, Peiyang Zheng, Wei Deng, Hui Gao, Kaixuan Yao, Yong Cheng, Mingkuan Wu, Rong He, Xiuzheng Yue, Yongqiang Yu, Ren Zhao, Xiaohu Li","doi":"10.1007/s10278-024-01179-7","DOIUrl":"10.1007/s10278-024-01179-7","url":null,"abstract":"<p><p>Cardiac amyloidosis (CA) is characterized by the deposition of amyloid fibrils within the myocardium, resulting in a restrictive physiology. Although microvascular dysfunction is a common feature, it is difficult to assess. This study aimed to explore myocardial transit time (MyoTT) by cardiovascular magnetic resonance (CMR) as a potential novel parameter of microcirculatory dysfunction in CA. This prospective study enrolled 20 CA patients and 20 control subjects. CMR acquisition included cine imaging, pre- and post-contrast T1 mapping, and MyoTT assessment, which was calculated from the time delay in contrast agent arrival between the aortic root and coronary sinus (CS). Compared to the control group, patients with CA exhibited significantly reduced left ventricular (LV) ejection fraction and myocardial strain, an increase in LV global peak wall thickness (LVGPWT), extracellular volume fraction (ECV), and prolonged MyoTT (14.4 ± 3.8 s vs. 7.7 ± 1.5 s, p < 0.001). Moreover, patients at Mayo stage III had a significantly longer MyoTT compared to those at stage I/II. MyoTT showed a positive correlation with the ECV, LVGPWT, and LV global longitudinal strain (LV-GLS) (p < 0.05). The area under the curve (AUC) for MyoTT was 0.962, demonstrating diagnostic performance comparable to that of the ECV (AUC 0.995) and LV-GLS (AUC 0.950) in identifying CA. MyoTT is significantly prolonged in patients with CA, correlating with fibrosis markers, remodeling, and dysfunction. As a novel parameter of coronary microvascular dysfunction (CMD), MyoTT has the potential to be an integral biomarker in multiparametric CMR assessment of CA.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"3049-3056"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612126/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141474484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Invariant Content Representation for Generalizable Medical Image Segmentation.","authors":"Zhiming Cheng, Shuai Wang, Yuhan Gao, Zunjie Zhu, Chenggang Yan","doi":"10.1007/s10278-024-01088-9","DOIUrl":"10.1007/s10278-024-01088-9","url":null,"abstract":"<p><p>Domain generalization (DG) for medical image segmentation due to privacy preservation prefers learning from a single-source domain and expects good robustness on unseen target domains. To achieve this goal, previous methods mainly use data augmentation to expand the distribution of samples and learn invariant content from them. However, most of these methods commonly perform global augmentation, leading to limited augmented sample diversity. In addition, the style of the augmented image is more scattered than the source domain, which may cause the model to overfit the style of the source domain. To address the above issues, we propose an invariant content representation network (ICRN) to enhance the learning of invariant content and suppress the learning of variability styles. Specifically, we first design a gamma correction-based local style augmentation (LSA) to expand the distribution of samples by augmenting foreground and background styles, respectively. Then, based on the augmented samples, we introduce invariant content learning (ICL) to learn generalizable invariant content from both augmented and source-domain samples. Finally, we design domain-specific batch normalization (DSBN) based style adversarial learning (SAL) to suppress the learning of preferences for source-domain styles. Experimental results show that our proposed method improves by 8.74% and 11.33% in overall dice coefficient (Dice) and reduces 15.88 mm and 3.87 mm in overall average surface distance (ASD) on two publicly available cross-domain datasets, Fundus and Prostate, compared to the state-of-the-art DG methods. The code is available at https://github.com/ZMC-IIIM/ICRN-DG .</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"3193-3207"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612095/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140961414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving Laryngoscopy Image Analysis Through Integration of Global Information and Local Features in VoFoCD Dataset.","authors":"Thao Thi Phuong Dao, Tuan-Luc Huynh, Minh-Khoi Pham, Trung-Nghia Le, Tan-Cong Nguyen, Quang-Thuc Nguyen, Bich Anh Tran, Boi Ngoc Van, Chanh Cong Ha, Minh-Triet Tran","doi":"10.1007/s10278-024-01068-z","DOIUrl":"10.1007/s10278-024-01068-z","url":null,"abstract":"<p><p>The diagnosis and treatment of vocal fold disorders heavily rely on the use of laryngoscopy. A comprehensive vocal fold diagnosis requires accurate identification of crucial anatomical structures and potential lesions during laryngoscopy observation. However, existing approaches have yet to explore the joint optimization of the decision-making process, including object detection and image classification tasks simultaneously. In this study, we provide a new dataset, VoFoCD, with 1724 laryngology images designed explicitly for object detection and image classification in laryngoscopy images. Images in the VoFoCD dataset are categorized into four classes and comprise six glottic object types. Moreover, we propose a novel Multitask Efficient trAnsformer network for Laryngoscopy (MEAL) to classify vocal fold images and detect glottic landmarks and lesions. To further facilitate interpretability for clinicians, MEAL provides attention maps to visualize important learned regions for explainable artificial intelligence results toward supporting clinical decision-making. We also analyze our model's effectiveness in simulated clinical scenarios where shaking of the laryngoscopy process occurs. The proposed model demonstrates outstanding performance on our VoFoCD dataset. The accuracy for image classification and mean average precision at an intersection over a union threshold of 0.5 (mAP50) for object detection are 0.951 and 0.874, respectively. Our MEAL method integrates global knowledge, encompassing general laryngoscopy image classification, into local features, which refer to distinct anatomical regions of the vocal fold, particularly abnormal regions, including benign and malignant lesions. Our contribution can effectively aid laryngologists in identifying benign or malignant lesions of vocal folds and classifying images in the laryngeal endoscopy process visually.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"2794-2809"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612113/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141163133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CSAMDT: Conditional Self Attention Memory-Driven Transformers for Radiology Report Generation from Chest X-Ray.","authors":"Iqra Shahzadi, Tahir Mustafa Madni, Uzair Iqbal Janjua, Ghanwa Batool, Bushra Naz, Muhammad Qasim Ali","doi":"10.1007/s10278-024-01126-6","DOIUrl":"10.1007/s10278-024-01126-6","url":null,"abstract":"<p><p>A radiology report plays a crucial role in guiding patient treatment, but writing these reports is a time-consuming task that demands a radiologist's expertise. In response to this challenge, researchers in the subfields of artificial intelligence for healthcare have explored techniques for automatically interpreting radiographic images and generating free-text reports, while much of the research on medical report creation has focused on image captioning methods without adequately addressing particular report aspects. This study introduces a Conditional Self Attention Memory-Driven Transformer model for generating radiological reports. The model operates in two phases: initially, a multi-label classification model, utilizing ResNet152 v2 as an encoder, is employed for feature extraction and multiple disease diagnosis. In the second phase, the Conditional Self Attention Memory-Driven Transformer serves as a decoder, utilizing self-attention memory-driven transformers to generate text reports. Comprehensive experimentation was conducted to compare existing and proposed techniques based on Bilingual Evaluation Understudy (BLEU) scores ranging from 1 to 4. The model outperforms the other state-of-the-art techniques by increasing the BLEU 1 (0.475), BLEU 2 (0.358), BLEU 3 (0.229), and BLEU 4 (0.165) respectively. This study's findings can alleviate radiologists' workloads and enhance clinical workflows by introducing an autonomous radiological report generation system.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"2825-2837"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612068/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141238906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prediction of Tumor-Associated Macrophages and Immunotherapy Benefits Using Weakly Supervised Contrastive Learning in Breast Cancer Pathology Images.","authors":"Guobang Yu, Yi Zuo, Bin Wang, Hui Liu","doi":"10.1007/s10278-024-01166-y","DOIUrl":"10.1007/s10278-024-01166-y","url":null,"abstract":"<p><p>The efficacy of immune checkpoint inhibitors is significantly influenced by the tumor immune microenvironment (TIME). RNA sequencing of tumor tissue can offer valuable insights into TIME, but its high cost and long turnaround time seriously restrict its utility in routine clinical examinations. Several recent studies have suggested that ultrahigh-resolution pathology images can infer cellular and molecular characteristics. However, few study pay attention to the quantitative estimation of various tumor infiltration immune cells from pathology images. In this paper, we integrated contrastive learning and weakly supervised learning to infer tumor-associated macrophages and potential immunotherapy benefit from whole slide images (WSIs) of H &E stained pathological sections. We split the high-resolution WSIs into tiles and then apply contrastive learning to extract features of each tile. After aggregating the features at the tile level, we employ weak supervisory signals to fine-tune the encoder for various downstream tasks. Comprehensive experiments on two independent breast cancer cohorts and spatial transcriptomics data demonstrate that the computational pathological features accurately predict the proportion of tumor-infiltrating immune cells, particularly the infiltration level of macrophages, as well as the immune subtypes and potential immunotherapy benefit. These findings demonstrate that our model effectively captures pathological features beyond human vision, establishing a mapping relationship between cellular compositions and histological morphology, thus expanding the clinical applications of digital pathology images.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"3090-3100"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612040/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141422417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Removing Adversarial Noise in X-ray Images via Total Variation Minimization and Patch-Based Regularization for Robust Deep Learning-based Diagnosis.","authors":"Burhan Ul Haque Sheikh, Aasim Zafar","doi":"10.1007/s10278-023-00919-5","DOIUrl":"10.1007/s10278-023-00919-5","url":null,"abstract":"<p><p>Deep learning has significantly advanced the field of radiology-based disease diagnosis, offering enhanced accuracy and efficiency in detecting various medical conditions through the analysis of complex medical images such as X-rays. This technology's ability to discern subtle patterns and anomalies has proven invaluable for swift and accurate disease identification. The relevance of deep learning in radiology has been particularly highlighted during the COVID-19 pandemic, where rapid and accurate diagnosis is crucial for effective treatment and containment. However, recent research has uncovered vulnerabilities in deep learning models when exposed to adversarial attacks, leading to incorrect predictions. In response to this critical challenge, we introduce a novel approach that leverages total variation minimization to combat adversarial noise within X-ray images effectively. Our focus narrows to COVID-19 diagnosis as a case study, where we initially construct a classification model through transfer learning designed to accurately classify lung X-ray images encompassing no pneumonia, COVID-19 pneumonia, and non-COVID pneumonia cases. Subsequently, we extensively evaluated the model's susceptibility to targeted and un-targeted adversarial attacks by employing the fast gradient sign gradient (FGSM) method. Our findings reveal a substantial reduction in the model's performance, with the average accuracy plummeting from 95.56 to 19.83% under adversarial conditions. However, the experimental results demonstrate the exceptional efficacy of the proposed denoising approach in enhancing the performance of diagnosis models when applied to adversarial examples. Post-denoising, the model exhibits a remarkable accuracy improvement, surging from 19.83 to 88.23% on adversarial images. These promising outcomes underscore the potential of denoising techniques to fortify the resilience and reliability of AI-based COVID-19 diagnostic systems, laying the foundation for their successful deployment in clinical settings.</p>","PeriodicalId":516858,"journal":{"name":"Journal of imaging informatics in medicine","volume":" ","pages":"3282-3303"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11639383/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141422418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}