Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12798-9. Online ahead of print.
NO ABSTRACT
PMID:42593502 | DOI:10.1007/s00330-026-12798-9
Category: Statistics
Category Added in a WPeMatico Campaign
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12786-z. Online ahead of print.
ABSTRACT
OBJECTIVES: The sensitivity of mammographic screening is lower for women with mammographically dense vs fatty breasts. We aimed to explore automated mammographic breast density and malignancy risk scores generated by an artificial intelligence (AI) model for breast cancer detection, stratified by mammography vendor.
MATERIALS AND METHODS: This retrospective study included information from 200,000 examinations within BreastScreen Norway. An automated volumetric breast density (VBD) assessment and risk scores were obtained from a commercial AI model. The continuous VBD output from the AI model was categorized into four groups, VBD 1-4, with 15%, 40%, 40%, and 5% of the examinations in each group for exam level and for right and left breast. Area under the receiver operating characteristic curve (AUC) for breast cancer detection was calculated using a continuous AI risk score. Analyses were performed for exam-level and breast-level measures.
RESULTS: At exam level, the difference in AUC estimates between the VBD groups was statistically significant, with the highest AUC observed for VBD 1 (0.955, 95% CI: 0.937-0.992) and lowest for VBD 4 (0.857, 95% CI: 0.803-0.910). Stratified by vendor, 3.3% and 6.7% were classified as VBD 4 for A vs B. At breast level, 18.5% of the examinations were classified with different VBD groups for the right and left breast.
CONCLUSION: Despite high AUC estimates across all VBD groups and the potential advantages of an automated density assessment, several factors, such as vendor and breast vs exam level, must be carefully considered before implementing automated density measures in screening programs.
KEY POINTS: Question What are the benefits and drawbacks of mammographic density assessment using an AI model for breast cancer detection? Findings The AI model’s performance for cancer detection was promising but decreased with increasing density category. Density distribution varied by equipment vendor and examination level. Clinical relevance AI risk assessment at the current examination was promising despite the variation by mammographic density. However, discrepancies in density by equipment vendors and examination level highlight the need for careful consideration before use in risk-stratified screening.
PMID:42593501 | DOI:10.1007/s00330-026-12786-z
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12792-1. Online ahead of print.
ABSTRACT
OBJECTIVES: To evaluate the prognostic significance of whole-body diffusion-weighted imaging (WBDWI)-derived metrics, including total diffusion volume (TDV) and global apparent diffusion coefficient (gADC), in patients with metastatic castration-resistant prostate cancer (mCRPC).
MATERIALS AND METHODS: This retrospective single-centre study included patients with mCRPC and bone-predominant disease who underwent whole-body MRI (WBMRI) prior to the start of a new treatment. Bone lesions were segmented on b900 DWI using an AI-assisted tool with radiologist refinement, and TDV and gADC were extracted. Automated bone scan index (aBSI) was calculated when bone scintigraphy was available. Clinical and laboratory parameters were recorded. Overall survival (OS) was the primary endpoint. Kaplan-Meier analysis and Cox regression models were used to assess prognostic associations.
RESULTS: Among 360 patients, higher TDV was associated with significantly shorter OS (median OS: 13.2 vs 33.4 months; p < 0.001), whereas gADC showed no significant association. In multivariable analysis including patients with WBMRI and bone scintigraphy (n = 182), log-TDV (HR 1.20; p = 0.027), LDH > 500 U/L (HR 7.79; p < 0.001) and ALP between 200 and 400 U/L (HR 1.78; p = 0.019) were independently associated with OS, while aBSI was not. In patients without aBSI data (n = 85), log-TDV remained prognostic (HR 1.60; p < 0.001), whereas LDH and ALP were not.
CONCLUSION: TDV derived from WBDWI independently predicts OS in mCRPC and demonstrates stronger prognostic performance than aBSI and conventional laboratory markers in multivariable models, whereas gADC was not prognostic.
KEY POINTS: QuestionThere is no widely adopted quantitative imaging biomarker for prognostic stratification in mCRPC. Can whole-body diffusion-weighted MRI-derived metrics improve prognostic risk stratification in patients with bone metastases? FindingsHigher TDV of bone disease on baseline whole-body diffusion-weighted MRI was independently associated with shorter OS in patients with mCRPC. Clinical relevance Whole-body diffusion-weighted MRI-derived tumour volume may provide a practical quantitative imaging biomarker for prognostic stratification in mCRPC, potentially supporting individualised treatment planning.
PMID:42593500 | DOI:10.1007/s00330-026-12792-1
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12793-0. Online ahead of print.
ABSTRACT
OBJECTIVE: To compare reading time, cancer detection rate (CDR), and abnormal interpretation rate (AIR) between AI-assisted and non-AI-assisted periods in screening and diagnostic mammography performed in routine clinical practice.
MATERIALS AND METHODS: We prospectively collected reading times for consecutive two-view full-field digital mammography interpreted by four radiologists between August 2023 and July 2024. Both screening and diagnostic examinations were included. A commercially available AI system was integrated into the clinical workflow, with results displayed or hidden on a monthly basis. Reading time, CDR, and AIR were compared between two periods. For reading time analysis, a subset of 2917 examinations with times ≤ 5 min was included to minimize the impact of non-interpretive interruptions. Reading time was extracted from the PACS log.
RESULTS: Among 4577 mammography examinations (mean age 51.7 ± 10.4 years), the overall CDR was higher during the AI-assisted period (22.3 vs 11.5 per 1000; p = 0.005). AIR did not differ for screening mammography (9.5% vs 8.4%; p = 0.293) but was higher with AI assistance for diagnostic mammography (18.7% vs 12.1%; p = 0.026). Mean reading times were comparable between AI-assisted and non-AI-assisted periods (65.0 vs 64.3 s; p = 0.723).
CONCLUSION: AI assistance in routine mammography interpretation was not associated with prolonged reading time and was associated with a higher overall CDR.
KEY POINTS: Question Does artificial intelligence assistance in routine mammography affect radiologists’ reading time, cancer detection, or abnormal interpretation in daily practice? Findings Artificial intelligence assistance was not associated with prolonged reading time and was associated with a higher overall cancer detection rate. Abnormal interpretation increased only for diagnostic examinations, with no difference in screening examinations. Clinical relevance AI assistance may support screening mammography by aiding cancer detection without disrupting workflow, whereas diagnostic use should be applied carefully due to higher abnormal interpretation rates.
PMID:42593499 | DOI:10.1007/s00330-026-12793-0
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12712-3. Online ahead of print.
ABSTRACT
OBJECTIVES: To analyze the factors for predicting 1-year outcomes and optimize surgical planning for Meso-Rex bypass (MRB) in children with extrahepatic portal vein obstruction (EHPVO).
MATERIALS AND METHODS: From October 2014 to July 2024, children with EHPVO after MRB were retrospectively analyzed. The number of MRB, the type of bypass vessel, and the position of the lower anastomosis were collected. Predictive variables include the diameter of the connection between the left and right portal veins, the Rex vein (RV) diameter, the number of RV branches, the diameter of the thickest RV branch, and the bypass-to-superior mesenteric vein (SMV) angle, which were measured by Contrast-enhanced CT (CECT) one week after MRB. The upper and lower anastomotic stenosis were diagnosed by Ultrasound and CECT one year after MRB.
RESULTS: A total of 153 children (median age 72 months, 87 males) were included in this study. Univariate factors include the number of MRBs, the type of bypass vessel, and CECT variables (except the diameter of the thickest RV branch). Numbers of MRB and RV branches were retained in the final model for upper anastomotic stenosis. The combination of MRB > 1 and RV branches ≤ 4 predicts it post-MRB with 75% sensitivity, 86.4% specificity, and an area under the receiver operator characteristic curve (AUC) of 0.832. The bypass-to-SMV angle ≤ 110° predicts lower anastomotic stenosis after MRB with 90% sensitivity, 73.39% specificity, and AUC 0.847.
CONCLUSIONS: Associated factors are useful for early postoperative risk stratification. Moreover, the type of bypass vessel and the bypass-to-SMV angle could be used to inform future surgical technique refinement.
KEY POINTS: QuestionWhich factors predict 1-year outcomes and optimize surgical planning for MRB in children with EHPVO? FindingsAssociated factors include bypass-to-SMV angle, number of MRB, RV branches and diameter, left-right portal vein connection diameter, and type of bypass vessel. Clinical RelevanceAssociated factors are useful for early postoperative risk stratification. Moreover, the internal jugular vein as the bridge vessel and the bypass-to-SMV angle > 110° could be used to optimize the surgical plan for a better outcome.
PMID:42593498 | DOI:10.1007/s00330-026-12712-3
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12784-1. Online ahead of print.
NO ABSTRACT
PMID:42593497 | DOI:10.1007/s00330-026-12784-1
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12789-w. Online ahead of print.
NO ABSTRACT
PMID:42593496 | DOI:10.1007/s00330-026-12789-w
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12785-0. Online ahead of print.
NO ABSTRACT
PMID:42593495 | DOI:10.1007/s00330-026-12785-0
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12788-x. Online ahead of print.
NO ABSTRACT
PMID:42593494 | DOI:10.1007/s00330-026-12788-x
Eur Radiol. 2026 Aug 13. doi: 10.1007/s00330-026-12782-3. Online ahead of print.
ABSTRACT
OBJECTIVES: To evaluate a DINOv2-based medical slice transformer (MST) for triaging abbreviated breast MRI by ruling out examinations with suspicious findings that are immediately actionable (Breast Imaging Reporting and Data System [BI-RADS] ≥ 4) across contrast-enhanced and non-contrast-enhanced protocols.
MATERIALS AND METHODS: This institutional review board-approved retrospective study included 1847 single-breast MRI examinations (377 BI-RADS ≥ 4) from an in-house dataset and 924 from an external dataset (Duke). Four abbreviated protocols were tested: T1-weighted early subtraction (T1sub), diffusion-weighted imaging with b = 1500 s/mm² (DWI1500), DWI1500 + T2-weighted (T2w), and T1sub + T2w. Performance was assessed at 90%, 95%, and 97.5% sensitivity using five-fold cross-validation and area under the receiver operating characteristic curve (AUC). AUC differences were compared with the DeLong test. False negatives were characterized, and attention maps were rated in the external dataset.
RESULTS: A total of 1448 female patients (mean age, 49 ± 12 years) were included. T1sub + T2w achieved an AUC of 0.77 ± 0.04; DWI1500 + T2w, 0.74 ± 0.04, with no significant differences across protocols. At 97.5% sensitivity, T1sub + T2w had the highest specificity (19% ± 7%), followed by DWI1500 + T2w (17% ± 11%). At 95% and 97.5% sensitivity, missed lesions were predominantly < 10 mm, mainly non-mass enhancements. External validation of the T1sub protocol yielded an AUC of 0.77, with 88% of attention maps rated good or moderate.
CONCLUSION: At 97.5% sensitivity, the MST framework triaged cases without BI-RADS ≥ 4, achieving 19% specificity for contrast-enhanced and 17% for non-contrast-enhanced MRI. These findings highlight the potential of foundation-model-based AI to support efficient triaging of abbreviated breast MRI protocols.
KEY POINTS: Question Can an adapted DINOv2-based MST accurately stratify abbreviated breast MRI examinations to support triaging across contrast-enhanced and non-contrast-enhanced imaging protocols? Findings The model achieved 19% specificity for contrast-enhanced and 17% for non-contrast-enhanced MRI at 97.5% sensitivity; external validation of the T1sub protocol yielded an AUC of 0.77. Clinical relevance A foundation-model-based triage approach may reduce radiologist workload in breast MRI screening by identifying examinations unlikely to contain suspicious findings, including on non-contrast-enhanced protocols, thereby enabling prioritization of higher-risk studies.
PMID:42593493 | DOI:10.1007/s00330-026-12782-3