European Radiology最新文献

{"title":"Predicting lesion reversal in acute cerebral ischaemia via apparent diffusion coefficient threshold on diffusion-weighted MRI.","authors":"Thor Håkon Skattør, Atle Bjørnerud, Terje Nome, Kine Mari Bakke, Brian Anthony Enriquez, Ingrid Digernes, Cecilie Mørck Offersen, Mona Kristiansen Beyer, Geir Ringstad, Anne Hege Aamodt","doi":"10.1007/s00330-025-12014-0","DOIUrl":"https://doi.org/10.1007/s00330-025-12014-0","url":null,"abstract":"<p><strong>Objectives: </strong>Diffusion-weighted imaging (DWI) quickly detects early ischaemic changes, but does not necessarily signify irreversible tissue damage, as DWI lesion reversal (DWI-R) can occur. Apparent diffusion coefficient (ADC) thresholds have been proposed to distinguish salvageable from irreversibly damaged tissue. This study aimed to evaluate the predictive value of a single ADC threshold for DWI-R following rapid and successful recanalization with voxel-level methodology.</p><p><strong>Materials and methods: </strong>In this cohort study, we retrospectively analysed consecutive patients examined with DWI before and the day after endovascular therapy, with successful recanalization within 120 min of baseline MRI. DWI-R was assessed voxel-wise for ADC values between 200 mm²/s and 760 × 10^-6 mm²/s. Predictive accuracy of ADC thresholds was evaluated using receiver operating characteristic (ROC) analyses.</p><p><strong>Results: </strong>Seventy-one patients with a mean baseline DWI lesion volume of 18.13 mL (IQR: 6.15, 26.25) were included. Median time from MRI to recanalization was 84.0 min (IQR: 72.0, 95.5). On average, 37.5% of voxels demonstrated reversal. The area under the curve for predicting reversal based on ADC was 0.708, and the optimal threshold was 555 × 10^-6 mm²/s (sensitivity 73.8%, specificity 58.6%). The voxel-wise probability of reversal declined with lower ADC, but even low ADC values exhibited some degree of reversal.</p><p><strong>Conclusion: </strong>This study reinforces existing concerns about using fixed ADC thresholds to define irreversible injury. A single ADC cut-off showed only modest sensitivity and poor specificity for predicting DWI-R. No definitive lower ADC boundary was identified across clinically relevant ranges below which the likelihood of DWI-R became negligible.</p><p><strong>Key points: </strong>Question Identifying penumbra from permanent ischaemic damage remains challenging, and the role of ADC evaluation before thrombectomy in predicting tissue viability is debated. Findings An ADC threshold of 555 × 10-6 mm²/s yielded moderate sensitivity (73.8%) and low specificity (58.6%) for predicting DWI-R (AUC 0.708). Clinical relevance A uniform ADC threshold has limited utility in identifying salvageable brain tissue in thrombectomy triage. Our findings emphasise the need for caution when excluding acute ischaemic stroke patients from recanalization therapy based on restrictive diffusion.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145191412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of hepatic iron quantification by MRI between 1.5 and 3 T, using different methods and conversion formulas.","authors":"Yves Gandon, Thi Hien Trang Chau, Paul Borde, Elise Bannier, Thibault Lapotre, Mustapha Azahaf, Edouard Bardou-Jacquet","doi":"10.1007/s00330-025-12027-9","DOIUrl":"https://doi.org/10.1007/s00330-025-12027-9","url":null,"abstract":"<p><strong>Objectives: </strong>This study aimed to compare liver iron concentration (LIC) estimates obtained from 1.5-T and 3-T MRI systems, using multiple quantification methods.</p><p><strong>Materials and methods: </strong>Seventy-eight participants from two centers were prospectively included and underwent liver MRI scans at both 1.5 T and 3 T. Two multi-echo gradient recalled echo (ME-GRE) sequences-one vendor-specific and one non-specific-were used to quantify liver fat and iron. Proton density fat fraction (PDFF) and R2* were obtained from vendor sequences via map reconstruction, and from non-specific sequences using MRQuantif software. LIC was derived from R2* using European or American calibrations, but also from signal intensity ratio (SIR) method at 1.5 T. Statistical analysis included linear correlation regression and Bland-Altman plots to evaluate agreement between LIC values across field strengths and methods.</p><p><strong>Results: </strong>For LIC calculated from R2*, both sequences demonstrated a strong correlation between 1.5-T and 3-T results, with R = 0.84 and 0.99 for the vendor and MRQuantif methods, respectively. The LIC bias between field strengths remained low (± 7% maximum) when using the same calibration group or when comparing, at 1.5 T, the SIR method and European calibrations. However, this bias increased to 26% when comparing LIC values obtained from American and European calibrations.</p><p><strong>Conclusion: </strong>LIC values are highly comparable when similar calibration methods are applied, but notable bias arises when comparing results from different calibration groups. For clinical convenience, a rapid and intermediate LIC (in µmol/g) can be estimated by dividing R2*values by 2 at 1.5 T or by 4 at 3 T.</p><p><strong>Key points: </strong>Question There are differences in the estimation of liver iron concentration depending on the magnetic field, method and calibration used. Findings Liver iron concentration (LIC) reproducibility is high across methods or between 1.5-T and 3-T devices-only when identical calibration groups are used. Clinical relevance LIC determination from R2 is highly reproducible, but the conversion formula must align with the referring physicians' standard references.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145174237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI demonstrates comparable diagnostic performance to radiologists in MRI detection of anterior cruciate ligament tears: a systematic review and meta-analysis.","authors":"Saran Singh Gill, Taha Haq, Yi Zhao, Mihailo Ristic, Dimitri Amiras, Chinmay Madhukar Gupte","doi":"10.1007/s00330-025-12020-2","DOIUrl":"https://doi.org/10.1007/s00330-025-12020-2","url":null,"abstract":"<p><strong>Introduction: </strong>Anterior cruciate ligament (ACL) injuries are among the most common knee injuries, affecting 1 in 3500 people annually. With rising rates of ACL tears, particularly in children, timely diagnosis is critical. This study evaluates artificial intelligence (AI) effectiveness in diagnosing and classifying ACL tears on MRI through a systematic review and meta-analysis, comparing AI performance with clinicians and assessing radiomic and non-radiomic models.</p><p><strong>Methods: </strong>Major databases were searched for AI models diagnosing ACL tears via MRIs. 36 studies, representing 52 models, were included. Accuracy, sensitivity, and specificity metrics were extracted. Pooled estimates were calculated using a random-effects model. Subgroup analyses compared MRI sequences, ground truths, AI versus clinician performance, and radiomic versus non-radiomic models. This study was conducted in line with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocols.</p><p><strong>Results: </strong>AI demonstrated strong diagnostic performance, with pooled accuracy, sensitivity, and specificity of 87.37%, 90.73%, and 91.34%, respectively. Classification models achieved pooled metrics of 90.46%, 88.68%, and 94.08%. Radiomic models outperformed non-radiomic models, and AI demonstrated comparable performance to clinicians in key metrics. Three-dimensional (3D) proton density fat suppression (PDFS) sequences with < 2 mm slice depth yielded the most promising results, despite small sample sizes, favouring arthroscopic benchmarks. Despite high heterogeneity (I² > 90%).</p><p><strong>Conclusion: </strong>AI models demonstrate diagnostic performance comparable to clinicians and may serve as valuable adjuncts in ACL tear detection, pending prospective validation. However, substantial heterogeneity and limited interpretability remain key challenges. Further research and standardised evaluation frameworks are needed to support clinical integration.</p><p><strong>Key points: </strong>Question Is AI effective and accurate in diagnosing and classifying anterior cruciate ligament (ACL) tears on MRI? Findings AI demonstrated high accuracy (87.37%), sensitivity (90.73%), and specificity (91.34%) in ACL tear diagnosis, matching or surpassing clinicians. Radiomic models outperformed non-radiomic approaches. Clinical relevance AI can enhance the accuracy of ACL tear diagnosis, reducing misdiagnoses and supporting clinicians, especially in resource-limited settings. Its integration into clinical workflows may streamline MRI interpretation, reduce diagnostic delays, and improve patient outcomes by optimising management.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"More than meets the eye: decoding glioma genomes in MRI pixel patterns.","authors":"Zhi-Cheng Li, Dong Liang, Hairong Zheng","doi":"10.1007/s00330-025-12041-x","DOIUrl":"https://doi.org/10.1007/s00330-025-12041-x","url":null,"abstract":"","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning reconstruction for temporomandibular joint MRI: diagnostic interchangeability, image quality, and scan time reduction.","authors":"Gyu-Dong Jo, Kug Jin Jeon, Yoon Joo Choi, Chena Lee, Sang-Sun Han","doi":"10.1007/s00330-025-12029-7","DOIUrl":"https://doi.org/10.1007/s00330-025-12029-7","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate the diagnostic interchangeability, image quality, and scan time of deep learning (DL)-reconstructed magnetic resonance imaging (MRI) compared with conventional MRI for the temporomandibular joint (TMJ).</p><p><strong>Materials and methods: </strong>Patients with suspected TMJ disorder underwent sagittal proton density-weighted (PDW) and T2-weighted fat-suppressed (T2W FS) MRI using both conventional and DL reconstruction protocols in a single session. Three oral radiologists independently assessed disc shape, disc position, and joint effusion. Diagnostic interchangeability for these findings was evaluated by comparing interobserver agreement, with equivalence defined as a 95% confidence interval (CI) within ±5%. Qualitative image quality (sharpness, noise, artifacts, overall) was rated on a 5-point scale. Quantitative image quality was assessed by measuring the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in the condyle, disc, and background air. Image quality scores were compared using the Wilcoxon signed-rank test, and SNR/CNR using paired t-tests. Scan times were directly compared.</p><p><strong>Results: </strong>A total of 176 TMJs from 88 patients (mean age, 37 ± 16 years; 43 men) were analyzed. DL-reconstructed MRI demonstrated diagnostic equivalence to conventional MRI for disc shape, position, and effusion (equivalence indices < 3%; 95% CIs within ±5%). DL reconstruction significantly reduced noise in PDW and T2W FS sequences (p < 0.05) while maintaining sharpness and artifact levels. SNR and CNR were significantly improved (p < 0.05), except for disc SNR in PDW (p = 0.189). Scan time was reduced by 49.2%.</p><p><strong>Conclusion: </strong>DL-reconstructed TMJ MRI is diagnostically interchangeable with conventional MRI, offering improved image quality with a shorter scan time.</p><p><strong>Key points: </strong>Question Long MRI scan times in patients with temporomandibular disorders can increase pain and motion-related artifacts, often compromising image quality in diagnostic settings. Findings DL reconstruction is diagnostically interchangeable with conventional MRI for assessing disc shape, disc position, and effusion, while improving image quality and reducing scan time. Clinical relevance DL reconstruction enables faster and more tolerable TMJ MRI workflows without compromising diagnostic accuracy, facilitating broader adoption in clinical settings where long scan times and motion artifacts often limit diagnostic efficiency.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Acute pulmonary embolism: a paradigm shift in interventional treatment and interdisciplinary care?","authors":"Willie M Lüdemann, Federico Collettini, Uli Fehrenbach, Timo A Auer, Maximilian de Bucourt, Bernhard Gebauer","doi":"10.1007/s00330-025-11805-9","DOIUrl":"10.1007/s00330-025-11805-9","url":null,"abstract":"","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical deployment and prospective validation of an AI model for limb-length discrepancy measurements using an open-source platform.","authors":"Andy Tsai, Sandip Samal, Paul Lamonica, Nicole Morris, John McNeil, Rudolph Pienaar","doi":"10.1007/s00330-025-12022-0","DOIUrl":"https://doi.org/10.1007/s00330-025-12022-0","url":null,"abstract":"<p><strong>Objectives: </strong>To deploy an AI model to measure limb-length discrepancy (LLD) and prospectively validate its performance.</p><p><strong>Materials and methods: </strong>We encoded the inference of an LLD AI model into a docker container, incorporated it into a computational platform for clinical deployment, and conducted two prospective validation studies: a shadow trial (07/2024-9/2024) and a clinical trial (11/2024-01/2025). During each trial period, we queried for LLD EOS scanograms to serve as inputs to our model. For the shadow trial, we hid the AI-annotated outputs from the radiologists, and for the clinical trial, we displayed the AI-annotated output to the radiologists at the time of study interpretation. Afterward, we collected the bilateral femoral and tibial lengths from the radiology reports and compared them against those generated by the AI model. We used median absolute difference (MAD) and interquartile range (IQR) as summary statistics to assess the performance of our model.</p><p><strong>Results: </strong>Our shadow trial consisted of 84 EOS scanograms from 84 children, with 168 femoral and tibial lengths. The MAD (IQR) of the femoral and tibial lengths were 0.2 cm (0.3 cm) and 0.2 cm (0.3 cm), respectively. Our clinical trial consisted of 114 EOS scanograms from 114 children, with 228 femoral and tibial lengths. The MAD (IQR) of the femoral and tibial lengths were 0.3 cm (0.4 cm) and 0.2 cm (0.3 cm), respectively.</p><p><strong>Conclusion: </strong>We successfully employed a computational platform for seamless integration and deployment of an LLD AI model into our clinical workflow, and prospectively validated its performance.</p><p><strong>Key points: </strong>Question No AI models have been clinically deployed for limb-length discrepancy (LLD) assessment in children, and the prospective validation of these models is unknown. Findings We deployed an LLD AI model using a homegrown platform, with prospective trials showing a median absolute difference of 0.2-0.3 cm in estimating bone lengths. Clinical relevance An LLD AI model with performance comparable to that of radiologists can serve as a secondary reader in increasing the confidence and accuracy of LLD measurements.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145130244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ultrashort echo time bi-component analysis of tibiotalar articular cartilage after long-distance running.","authors":"Dantian Zhu, Yiyin Hu, Feier Deng, Wenhao Wu, Wei Li, Jiaxin Feng, Yajun Ma, Shaolin Li, Yijie Fang","doi":"10.1007/s00330-025-12016-y","DOIUrl":"https://doi.org/10.1007/s00330-025-12016-y","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate tibiotalar cartilage changes in amateur marathon runners pre- and post-marathon using 3D ultrashort echo time (UTE) bi-component analysis.</p><p><strong>Materials and methods: </strong>Amateur runners were prospectively enrolled and underwent ankle MRI at three time points: pre-marathon, 2 days post-marathon, and 4 weeks post-marathon. UTE component analysis was used to obtain single-component values (T2*M), and bi-component values (short (T2*S) and long T2* component values (T2*L), and short T2* fractions) of cartilage. Sagittal images were analyzed by segmenting tibial and talus cartilage into 12 subregions (medial/lateral, anterior/middle/posterior).</p><p><strong>Results: </strong>Thirty-two runners (26 men, 6 women; mean age, 39.80 ± 6.00 years) were evaluated. UTE component analysis parameters increased in most subregions after running, with T2*M increasing further at 4 weeks, while T2*S, T2*L, and short T2* fractions decreased. Repeated-measures analysis of variance (RM-ANOVA) revealed significant T2*S differences in the middle and posterior medial tibia (MTiM, MTiP), the middle medial talus (MTaM), the anterior, middle, and posterior lateral tibia (LTiA, LTiM, and LTiP), and the middle and posterior lateral talus (LTaM and LTaP) (p < 0.05). Short T2* fractions exhibited significant changes in MTiM, MTiP, MTaM, LTiM, and LTaM (RM-ANOVA, p < 0.05). MTiP, MTaP, LTiM, and LTaM showed significant T2*M changes (RM-ANOVA, p < 0.05). Only LTaM showed significant T2*L changes (Friedman's rank test, p < 0.05).</p><p><strong>Conclusions: </strong>The T2*S and short T2* fractions from UTE bi-component analysis may be more sensitive than T2*M, offering a promising method for detecting dynamic changes in ankle cartilage following long-distance running.</p><p><strong>Key points: </strong>Question Long-distance running causes changes in the tibiotalar articular cartilage. Can UTE component analysis of T2* monitor dynamic changes of tibiotalar articular cartilage non-invasively? Findings T2*S and short T2* fractions of UTE bi-component analysis were superior to single-component analysis in monitoring dynamic changes in ankle cartilage. Clinical relevance This study suggests that the UTE bi-component T2* analysis detects exercise-induced cartilage changes, allowing early matrix assessment in at-risk populations and supporting prevention strategies for athletes.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145130317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Letter to the Editor: Node-RADS and nodal staging-promise, pitfalls, and pathways forward.","authors":"Deniz Esin Tekcan Sanli, Ahmet Necati Sanli","doi":"10.1007/s00330-025-12017-x","DOIUrl":"https://doi.org/10.1007/s00330-025-12017-x","url":null,"abstract":"","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145130269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of delays in diagnosis of lung cancer in interstitial lung disease.","authors":"Tician Schnitzler, Ali Nowroozi, Maya Vella, Jonathan Liu, Shravan Sridhar, Erica Farrand, Jae Ho Sohn","doi":"10.1007/s00330-025-12015-z","DOIUrl":"https://doi.org/10.1007/s00330-025-12015-z","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate diagnostic delays in lung cancer among patients with interstitial lung disease (ILD), assess tumor characteristics and outcomes, and identify radiologic factors contributing to missed or delayed detection.</p><p><strong>Materials and methods: </strong>In this retrospective study, 54 ILD patients with concurrent lung cancer from two tertiary ILD referral centers were reviewed. A panel of four thoracic radiologists determined the earliest callable date-when the majority would have raised concern about a nodule-compared to the date of first clinical suspicion. Tumor size, morphology, staging, growth rate (volume doubling time), treatment regimens, and survival were analyzed. For delayed cases, radiologic causes of missed detection were categorized by consensus.</p><p><strong>Results: </strong>Diagnostic delays occurred in 50% of cases (27/54), with a mean delay of 3.18 years (IQR: 0.94-4.51). Delayed cases showed significantly slower tumor growth (mean volume doubling time: 1.42 vs. 0.22 years, p < 0.0001). The most common radiologic factors contributing to missed detection were obscuration by fibrosis, ground-glass opacity, or bronchovascular bundles (63%), misinterpretation as infection or exacerbation (15%), minimal interval growth (11%), and invisibility on CT (15%). Tumor size, morphology, stage, and histologic subtype-most commonly adenocarcinoma (56%)-were similar between groups. No significant difference in survival was observed between delayed and non-delayed cases (χ² = 0.4, p = 0.5).</p><p><strong>Conclusion: </strong>Half of the lung cancer cases in ILD patients were diagnosed with significant delay, primarily due to parenchymal masking and subtle imaging findings, rather than biologically aggressive disease. Despite delays, survival outcomes and tumor characteristics did not differ significantly, suggesting a less aggressive tumor phenotype in delayed cases.</p><p><strong>Key points: </strong>Question To evaluate diagnostic delays in lung cancer among patients with interstitial lung disease (ILD), assess tumor characteristics and outcomes, and identify radiologic factors contributing to missed or delayed detection. Findings Half of lung cancer cases in ILD patients had diagnostic delays, mainly due to parenchymal masking. Delayed cases showed slower growth but no survival difference; tumor size and histology were indifferent. Clinical relevance This study shows that lung cancer diagnosis is often delayed in ILD patients (avg. > 3 years). Clinicians should stay alert for malignancy in peripheral fibrotic regions with slow-growing nodules. Despite no clear survival impact, future work should assess the risks and benefits of earlier detection.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145130168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}