Diagnostics最新文献

{"title":"Diagnostic Accuracy of Pre-Biopsy MRI and CT Features for Predicting Vertebral Biopsy Yield in Suspected Vertebral Discitis Osteomyelitis: A Retrospective Single-Center Study.","authors":"Amirmasoud Negarestani, Andrew Pasion, Caleb Bhatnagar, Zuhaib Khokhar, Ashima Kundu, Samantha Diulus, Jorge P Parada, Emad Allam","doi":"10.3390/diagnostics15141760","DOIUrl":"10.3390/diagnostics15141760","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Vertebral discitis osteomyelitis (VDO) is a serious infection involving the vertebral bodies and intervertebral discs, often requiring biopsy for pathogen identification. However, biopsy yields are variable, and guidance on patient selection remains limited. This study aimed to assess how biopsy culture results influence clinical management and to develop imaging-based scoring systems to predict biopsy outcomes. <b>Methods</b>: In this retrospective study, 70 patients who underwent image-guided vertebral biopsy for suspected VDO between 2013 and 2022 were reviewed. Pre-biopsy MRI and CT findings were scored using novel, simplified criteria. MRI was graded based on soft tissue involvement, while CT evaluated the presence or absence of a vacuum phenomenon. Culture results were correlated with imaging scores and subsequent changes in antibiotic management. Statistical analysis included logistic regression, ROC analysis, and interobserver agreement using Cohen's Kappa. <b>Results</b>: Of the 70 patients, 27 (38.6%) had positive cultures, and 20 (28.5%) experienced changes in management. Among the 48 patients with both MRI and CT imaging, MRI scores indicating soft tissue involvement and absence of the vacuum sign on CT were independent predictors of positive culture (<i>p</i> = 0.022 and <i>p</i> = 0.047, respectively). The combined predictive model showed an AUC of 0.76. Interobserver agreement was excellent (κ = 0.90 for MRI, κ = 0.95 for CT). <b>Conclusions</b>: MRI and CT features can be used to predict biopsy yield and guide clinical decisions in suspected VDO. These scoring systems may help clinicians identify patients most likely to benefit from biopsy, potentially improving outcomes and minimizing unnecessary procedures.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293688/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Overall Survival and Complication Rates in the Treatment of Liver Carcinoma: A Comparative Study of Ultrasound, Computed Tomography, and Combined Ultrasound and Computed Tomography Guidance for Radiofrequency Ablation.","authors":"Chia-Hsien Chien, Chia-Ling Chiang, Huei-Lung Liang, Jer-Shyung Huang, Chia-Jung Tsai","doi":"10.3390/diagnostics15141754","DOIUrl":"10.3390/diagnostics15141754","url":null,"abstract":"<p><p><b>Background:</b> Liver cancer is a major health concern worldwide. Radiofrequency ablation is a safe treatment option that can be guided by either ultrasound, computer tomography (CT), or fluoroscopy. Although ultrasound-guided radiofrequency ablation is commonly used in clinical practice, radiofrequency ablation guided by CT is more precise but requires more time and does not offer real-time monitoring, which may result in complications such as pneumothorax or organ damage. <b>Objectives:</b> In this study, we investigated the effect of ultrasound, CT, and combined ultrasound/CT guidance on patient survival and complication development. <b>Methods:</b> A total of 982 radiofrequency ablation sessions conducted on 553 patients were analyzed. Clinical outcomes were assessed during follow-up to determine the survival and recurrence rates of malignant tumors. <b>Results:</b> Overall, the three guidance approaches exhibited significant differences in terms of tumor size, number, complication development, and treatment duration. However, no significant differences were observed in survival rate. A comparison of the effect of CT guidance and ultrasound guidance on complication development revealed a higher odds ratio for CT guidance in some cases. A comparison of combined ultrasound/CT guidance and ultrasound guidance revealed nonsignificant differences in complication development. A comparison of CT guidance and combined ultrasound/CT guidance revealed a higher odds ratio for CT guidance in some cases. Radiofrequency ablation is a safe and effective treatment for liver tumors. However, CT has an increased incidence of complications. <b>Conclusions:</b> Combined ultrasound/computer tomography guidance is recommended for patients with multiple or large tumors or tumors near the hepatic dome or diaphragm.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293419/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unsupervised Clustering Successfully Predicts Prognosis in NSCLC Brain Metastasis Cohorts.","authors":"Emre Uysal, Gorkem Durak, Ayse Kotek Sedef, Ulas Bagci, Tanju Berber, Necla Gurdal, Berna Akkus Yildirim","doi":"10.3390/diagnostics15141747","DOIUrl":"10.3390/diagnostics15141747","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Current developments in computer-aided systems rely heavily on complex and computationally intensive algorithms. However, even a simple approach can offer a promising solution to reduce the burden on clinicians. Addressing this, we aim to employ unsupervised cluster analysis to identify prognostic subgroups of non-small-cell lung cancer (NSCLC) patients with brain metastasis (BM). Simple-yet-effective algorithms designed to identify similar group characteristics will assist clinicians in categorizing patients effectively. <b>Methods</b>: We retrospectively collected data from 95 NSCLC patients with BM treated at two oncology centers. To identify clinically distinct subgroups, two types of unsupervised clustering methods-two-step clustering (TSC) and hierarchical cluster analysis (HCA)-were applied to the baseline clinical data. Patients were categorized into prognostic classes according to the Diagnosis-Specific Graded Prognostic Assessment (DS-GPA). Survival curves for the clusters and DS-GPA classes were generated using Kaplan-Meier analysis, and the differences were assessed with the log-rank test. The discriminative ability of three categorical variables on survival was compared using the concordance index (C-index). <b>Results</b>: The mean age of the patients was 61.8 ± 0.9 years, and the majority (77.9%) were men. Extracranial metastasis was present in 71.6% of the patients, with most (63.2%) having a single BM. The DS-GPA classification significantly divided the patients into prognostic classes (<i>p</i> < 0.001). Furthermore, statistical significance was observed between clusters created by TSC (<i>p</i> < 0.001) and HCA (<i>p</i> < 0.001). HCA showed the highest discriminatory power (C-index = 0.721), followed by the DS-GPA (C-index = 0.709) and TSC (C-index = 0.650). <b>Conclusions</b>: Our findings demonstrated that the TSC and HCA models were comparable in prognostic performance to the DS-GPA index in NSCLC patients with BM. These results suggest that unsupervised clustering may offer a data-driven perspective on patient stratification, though further validation is needed to clarify its role in prognostic modeling.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293156/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AΚtransU-Net: Transformer-Equipped U-Net Model for Improved Actinic Keratosis Detection in Clinical Photography.","authors":"Panagiotis Derekas, Charalampos Theodoridis, Aristidis Likas, Ioannis Bassukas, Georgios Gaitanis, Athanasia Zampeta, Despina Exadaktylou, Panagiota Spyridonos","doi":"10.3390/diagnostics15141752","DOIUrl":"10.3390/diagnostics15141752","url":null,"abstract":"<p><p><b>Background:</b> Integrating artificial intelligence into clinical photography offers great potential for monitoring skin conditions such as actinic keratosis (AK) and skin field cancerization. Identifying the extent of AK lesions often requires more than analyzing lesion morphology-it also depends on contextual cues, such as surrounding photodamage. This highlights the need for models that can combine fine-grained local features with a comprehensive global view. <b>Methods:</b> To address this challenge, we propose AKTransU-net, a hybrid U-net-based architecture. The model incorporates Transformer blocks to enrich feature representations, which are passed through ConvLSTM modules within the skip connections. This configuration allows the network to maintain semantic coherence and spatial continuity in AK detection. This global awareness is critical when applying the model to whole-image detection via tile-based processing, where continuity across tile boundaries is essential for accurate and reliable lesion segmentation. <b>Results:</b> The effectiveness of AKTransU-net was demonstrated through comparative evaluations with state-of-the-art segmentation models. A proprietary annotated dataset of 569 clinical photographs from 115 patients with actinic keratosis was used to train and evaluate the models. From each photograph, crops of 512 × 512 pixels were extracted using translation lesion boxes that encompassed lesions in different positions and captured different contexts. AKtransU-net exhibited a more robust context awareness and achieved a median Dice score of 65.13%, demonstrating significant progress in whole-image assessments. <b>Conclusions:</b> Transformer-driven context modeling offers a promising approach for robust AK lesion monitoring, supporting its application in real-world clinical settings where accurate, context-aware analysis is crucial for managing skin field cancerization.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293264/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728594","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Expanded Performance Comparison of the Oncuria 10-Plex Bladder Cancer Urine Assay Using Three Different Luminex xMAP Instruments.","authors":"Sunao Tanaka, Takuto Shimizu, Ian Pagano, Wayne Hogrefe, Sherry Dunbar, Charles J Rosser, Hideki Furuya","doi":"10.3390/diagnostics15141749","DOIUrl":"10.3390/diagnostics15141749","url":null,"abstract":"<p><p><b>Background/Objectives</b>: The clinically validated multiplex Oncuria bladder cancer (BC) assay quickly and noninvasively identifies disease risk and tracks treatment success by simultaneously profiling 10 protein biomarkers in voided urine samples. Oncuria uses paramagnetic bead-based fluorescence multiplex technology (xMAP^®; Luminex, Austin, TX, USA) to simultaneously measure 10 protein analytes in urine [angiogenin, apolipoprotein E, carbonic anhydrase IX (CA9), interleukin-8, matrix metalloproteinase-9 and -10, alpha-1 anti-trypsin, plasminogen activator inhibitor-1, syndecan-1, and vascular endothelial growth factor]. <b>Methods</b>: In a pilot study (N = 36 subjects; 18 with BC), Oncuria performed essentially identically across three different common analyzers (the laser/flow-based FlexMap 3D and 200 systems, and the LED/image-based MagPix system; Luminex). The current study compared Oncuria performance across instrumentation platforms using a larger study population (N = 181 subjects; 51 with BC). <b>Results</b>: All three analyzers assessed all 10 analytes in identical samples with excellent concordance. The percent coefficient of variation (%CV) in protein concentrations across systems was ≤2.3% for 9/10 analytes, with only CA9 having %CVs > 2.3%. In pairwise correlation plot comparisons between instruments for all 10 biomarkers, R² values were 0.999 for 15/30 comparisons and R² ≥ 0.995 for 27/30 comparisons; CA9 showed the greatest variability (R² = 0.948-0.970). Standard curve slopes were statistically indistinguishable for all 10 biomarkers across analyzers. <b>Conclusions</b>: The Oncuria BC assay generates comprehensive urinary protein signatures useful for assisting BC diagnosis, predicting treatment response, and tracking disease progression and recurrence. The equivalent performance of the multiplex BC assay using three popular analyzers rationalizes test adoption by CLIA (Clinical Laboratory Improvement Amendments) clinical and research laboratories.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12294033/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Role of Musculoskeletal Ultrasound in Biologic Drug Tapering and Relapse Monitoring: Findings from a One-Year Prospective Study in a Cohort of Rheumatoid Arthritis Patients in Sustained Clinical Remission.","authors":"Zguro Batalov, Tanya Sapundzhieva, Konstantin Batalov, Rositsa Karalilova, Anastas Batalov","doi":"10.3390/diagnostics15141753","DOIUrl":"10.3390/diagnostics15141753","url":null,"abstract":"<p><p><b>Objectives:</b> To assess the role of musculoskeletal ultrasound (MSUS) in selecting patients with rheumatoid arthritis (RA) in sustained clinical remission, suitable for tapering of biologic therapy (BT), and monitoring for a subclinical relapse. <b>Methods:</b> In this prospective study, seventy-eight patients with RA in sustained Disease Activity for twenty-eight joints (DAS28) clinical remission underwent ultrasound (US) examination of twenty-two joints (bilaterally wrists and metacarpophalangeal and proximal interphalangeal joints). US assessment was performed on gray scale ultrasound (GSUS) and power Doppler US (PDUS) to select patients in imaging remission, defined as a total PD score of synovitis = 0. Group 1 consisted of patients in clinical and imaging remission, in which tapering of BT was done through spacing of the Tumour Necrosis Factor Alpha (TNF-α) blocker. Group 2 consisted of patients only in clinical remission (PDUS > 0), who continued standard therapy. Clinical and US assessment was done at months 6 and 12, and the rate of a clinical (defined as DAS28 ≥ 2.6) and an US relapse (PDUS score ≥ 1) was recorded. <b>Results:</b> Thirty-eight patients were in clinical and US remission (group 1) and forty patients only in clinical remission (group 2). At month 6, 26% of patients in group 1 and 10% in group 2 experienced a clinical and an US relapse, whereas 20% and 15% of them, respectively, only an US relapse. At month 12, 26% of patients in group 1 and 20% of patients in group 2 experienced a clinical and an US relapse, whereas 35% and 22% of them, respectively, only an US relapse. <b>Conclusions:</b> Real-world data show that MSUS is a useful tool to identify RA patients in sustained clinical remission appropriate for BT tapering. US monitoring could predict a clinical relapse and the need to re-escalate treatment in patients with subclinical US relapse during BT tapering.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293901/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Breast Cancer Classification with Various Optimized Deep Learning Methods.","authors":"Mustafa Güler, Gamze Sart, Ömer Algorabi, Ayse Nur Adıguzel Tuylu, Yusuf Sait Türkan","doi":"10.3390/diagnostics15141751","DOIUrl":"10.3390/diagnostics15141751","url":null,"abstract":"<p><p><b>Background/Objectives:</b> In recent years, there has been a significant increase in the number of women with breast cancer. Breast cancer prediction is defined as a medical data analysis and image processing problem. Experts may need artificial intelligence technologies to distinguish between benign and malignant tumors in order to make decisions. When the studies in the literature are examined, it can be seen that applications of deep learning algorithms in the field of medicine have achieved very successful results. <b>Methods:</b> In this study, 11 different deep learning algorithms (Vanilla, ResNet50, ResNet152, VGG16, DenseNet152, MobileNetv2, EfficientB1, NasNet, DenseNet201, ensemble, and Tuned Model) were used. Images of pathological specimens from breast biopsies consisting of two classes, benign and malignant, were used for classification analysis. To limit the computational time and speed up the analysis process, 10,000 images, 6172 IDC-negative and 3828 IDC-positive, were selected. Of the images, 80% were used for training, 10% were used for validation, and 10% were used for testing the trained model. <b>Results:</b> The results demonstrate that DenseNet201 achieved the highest classification accuracy of 89.4%, with a precision of 88.2%, a recall of 84.1%, an F1 score of 86.1%, and an AUC score of 95.8%. <b>Conclusions:</b> In conclusion, this study highlights the potential of deep learning algorithms in breast cancer classification. Future research should focus on integrating multi-modal imaging data, refining ensemble learning methodologies, and expanding dataset diversity to further improve the classification accuracy and real-world clinical applicability.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293705/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Non-Coding RNAs in Asthma: Regulators of Eosinophil Biology and Airway Inflammation.","authors":"Eglė Vasylė, Andrius Januškevičius, Kęstutis Malakauskas","doi":"10.3390/diagnostics15141750","DOIUrl":"10.3390/diagnostics15141750","url":null,"abstract":"<p><p>Asthma is a complex and heterogeneous disease characterized by chronic airway inflammation, bronchial hyperresponsiveness, and reversible airflow obstruction. Despite extensive research, its underlying molecular mechanisms remain incompletely understood. Among the key immune cells involved, eosinophils play a central role in asthma pathophysiology through their contributions to Type 2 inflammation, tissue remodeling, and immune regulation. Recent studies have shown that non-coding RNAs (ncRNAs) play a crucial role in regulating eosinophil biology and contribute to the molecular mechanisms underlying asthma progression. This review consolidates the current understanding of ncRNAs in the development of eosinophils, their involvement in asthma pathogenesis, and the mechanisms underlying this process.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293544/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Study on the Diagnosis of Lumbar Spinal Malposition in Chuna Manual Therapy Using X-Ray Images Based on Digital Markers.","authors":"Min-Su Ju, Tae-Yong Park, Minho Choi, Younseok Ko, Young Cheol Na, Yeong Ha Jeong, Jun-Su Jang, Jin-Hyun Lee","doi":"10.3390/diagnostics15141748","DOIUrl":"10.3390/diagnostics15141748","url":null,"abstract":"<p><p><b>Background/Objectives:</b> This study aimed to evaluate digital markers and establish quantitative diagnostic criteria for spinal malpositions in Chuna manual therapy using lumbar X-rays. <b>Methods:</b> A total of 2000 X-ray images were collected from adult patients at the International St. Mary's Hospital of Catholic Kwandong University. Five Chuna manual medicine experts annotated anatomical landmarks using a digital marker labeling program and diagnosed three types of spinal malpositions: flexion/extension, lateral bending, and rotation. Diagnostic accuracy was evaluated using weighted F1 (F1_W) scores, and the optimal threshold values for each malposition type were determined based on maximum F1_W performance. <b>Results:</b> The results showed high diagnostic performance, with average maximum F1_W scores of 0.76 for flexion/extension, 0.85 for lateral bending, and 0.71 for rotation. Based on this analysis, threshold angles for each type of spinal malposition in Chuna manual diagnosis were determined. <b>Conclusions:</b> This study demonstrates the diagnostic validity of digital marker-based X-ray analysis in Chuna manual therapy and is the first to propose quantitative diagnostic thresholds for spinal malpositions. These findings may serve as a foundation for clinical application in spinal assessment and treatment planning, with further validation studies warranted.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293895/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning-Based Detection of Separated Root Canal Instruments in Panoramic Radiographs Using a U²-Net Architecture.","authors":"Nildem İnönü, Umut Aksoy, Dilan Kırmızı, Seçil Aksoy, Nurullah Akkaya, Kaan Orhan","doi":"10.3390/diagnostics15141744","DOIUrl":"10.3390/diagnostics15141744","url":null,"abstract":"<p><p><b>Background:</b> Separated endodontic instruments are a significant complication in root canal treatment, affecting disinfection and long-term prognosis. Their detection on panoramic radiographs is challenging, particularly in complex anatomy or for less experienced clinicians. <b>Objectives:</b> This study aimed to develop and evaluate a deep learning model using the U²-Net architecture for automated detection and segmentation of separated instruments in panoramic radiographs from multiple imaging systems. <b>Methods:</b> A total of 36,800 panoramic radiographs were retrospectively reviewed, and 191 met strict inclusion criteria. Separated instruments were manually segmented using the Computer Vision Annotation Tool. The U²-Net model was trained and evaluated using standard performance metrics: Dice coefficient, IoU, precision, recall, and F1 score. <b>Results:</b> The model achieved a Dice coefficient of 0.849 (95% CI: 0.840-0.857) and IoU of 0.790 (95% CI: 0.781-0.799). Precision was 0.877 (95% CI: 0.869-0.884), recall was 0.847 (95% CI: 0.839-0.855), and the F1-score was 0.861 (95% CI: 0.853-0.869). <b>Conclusions:</b> These results demonstrate a strong overlap between predictions and ground truth, indicating high segmentation accuracy. The U²-Net model showed robust performance across radiographs from various systems, suggesting its clinical utility in aiding detection and treatment planning. Further multicenter studies are recommended to confirm generalizability.</p>","PeriodicalId":11225,"journal":{"name":"Diagnostics","volume":"15 14","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12293869/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144728608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}