British Journal of Radiology最新文献

{"title":"Efficacy and safety of stereotactic radiosurgery in the management of brain metastases in patients with metastatic melanoma.","authors":"David J McMahon, Mohamed Aboulela, James De Boisanger, Juan Jose Soto-Castillo, Chloe Beland, Sowmya Cheruvu, Martin Brewer, Praveena Idaikkadar, Andrew Furness, Samra Turajlic, Kate Young, Lisa Pickering, Liam Welsh, Nicola Rosenfelder, James Larkin, Francesca Solda","doi":"10.1093/bjr/tqag022","DOIUrl":"10.1093/bjr/tqag022","url":null,"abstract":"<p><strong>Objectives: </strong>We sought to identify factors associated with prognosis and bleeding in patients with melanoma brain metastases (BM). Objectives were median overall survival (mOS [months]) and bleeding incidence.</p><p><strong>Methods: </strong>We conducted a retrospective analysis of patients receiving stereotactic radiosurgery (SRS) at our centre 11/15-11/23. Analysis was performed on Prism 10.1.1. Extraction from electronic medical records was undertaken by authors, with local R&D approval (NEU012).</p><p><strong>Results: </strong>A total of 81 patients were evaluable. One hundred and nineteen treatment courses were delivered. There were no bleeding events (BEs) in the first 7 days, 1 within 28 days (0.8%) and 19 within 90 days (16%). Overall, 7.6% (N = 9) died within 90 days, and 25.2% (N = 30) died within 6 m. There were no deaths within 30 days or related to treatment. mOS was 17.6 m (95% confidence interval [CI] 9.20-35.05). Significantly inferior survival outcomes were observed for patients with elevated lactate dehydrogenase (LDH), poor performance status (Karnofsky performance status [KPS]), higher total treated intracranial volume (gross target volume [GTV]), and total number of BM. mOS of patients with a normal LDH was 37.0 m vs 5.2 m for those with LDH > upper limit of normal (ULN) (HR 4.40, P<.0001). This was also true on multivariable analysis including KPS, BM number, and BM size (HR 3.75 95% CI, 1.94-7.30, P ≤.0001). For patients with KPS ≥90, mOS was 35.0 m vs 7.7 m for KPS ≤80 (HR 2.55, P<.0004). This was significant using the multivariable analysis described above (HR 2.12 95% CI, 1.123-3.948, P = .0181).</p><p><strong>Conclusions: </strong>Despite theoretically high risk of bleeding after SRS in MM BM, incidence of bleeding in our cohort was low. mOS was comparable to historical controls of 16-23 months.</p><p><strong>Advances in knowledge: </strong>We novelly performed univariate and multivariate analysis demonstrating poor survival outcomes in patients with high LDH, poor performance status, and larger BM (both by BM size and number).</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"906-912"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146046232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ultrasound accuracy in classifying full-thickness supraspinatus tear patterns: an arthroscopic correlation study.","authors":"Shengqiao Wang, Kuanli Tang, Juan Chen, Jun Jiang, Yi Dong","doi":"10.1093/bjr/tqag041","DOIUrl":"10.1093/bjr/tqag041","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate ultrasound's diagnostic value for full-thickness supraspinatus tendon (SST) tears and correlate it with arthroscopy findings.</p><p><strong>Methods: </strong>In this retrospective study, patients with a clinical diagnosis of rotator cuff tears were included. The inclusion criteria were patients with shoulder joint pain and limited shoulder mobility. The exclusion criteria comprised patients who had undergone any previous shoulder surgery or those with shoulder fractures or dislocations. Ultrasound imaging features were observed and recorded, including measurements of the length and width of tears to reveal tear patterns. The chi-square test was used to compare the clinical characteristics of categorical variables, and the Wilcoxon test was used to assess the correlation between ultrasound and arthroscopy.</p><p><strong>Results: </strong>From January 2023 to November 2023, 105 patients diagnosed with rotator cuff tears were included (50 males, 55 females, mean age 67.9 ± 14.2 years). Arthroscopy confirmed full-thickness SST tears in 82 patients (78.1%). Ultrasound demonstrated differential diagnostic performance across tear patterns: sensitivity/specificity 90.0%/95.8% (crescent), 93.1%/94.3% (U-shaped), 83.8%/98.0% (L-shaped), and 100%/98.6% (giant), with corresponding area under the ROC curves (AU-ROCs) of 0.789-0.882. Significant correlations were observed between ultrasound and arthroscopic measurements, particularly for crescent-type tear width (r = 0.949, P < .001).</p><p><strong>Conclusions: </strong>Ultrasound effectively distinguishes SST tear types, with high concordance to arthroscopy, supporting its use as a primary imaging tool for surgical planning in rotator cuff pathology.</p><p><strong>Advances in knowledge: </strong>Ultrasound has significant application value in the diagnosis of full-thickness of SST tears. Ultrasound shows high consistency with arthroscopic findings and can effectively differentiate various tear patterns.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"951-958"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147316477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Letter to the editor regarding \"how can MRI descriptors be optimally combined to predict idiopathic intracranial hypertension?\"","authors":"Erfan Naghavi, Hoda Asefi","doi":"10.1093/bjr/tqaf308","DOIUrl":"10.1093/bjr/tqaf308","url":null,"abstract":"","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"1006-1007"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146225683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantification of myocardial iron and fat-an experimental study with photon-counting detector CT.","authors":"Philipp N Maintz, Tristan T Demmert, Thomas Flohr, Konstantin Klambauer, Lukas J Moser, Victor Mergen, Matthias Eberhard, Johannes M Froehlich, Hatem Alkadhi","doi":"10.1093/bjr/tqag020","DOIUrl":"10.1093/bjr/tqag020","url":null,"abstract":"<p><strong>Objectives: </strong>To determine the feasibility and accuracy of photon-counting detector (PCD)-CT for iron and fat quantification in the myocardium.</p><p><strong>Methods: </strong>Cylindrical tubes were filled with porcine myocardium and iron citrate with iron concentrations of 0-20 mg Fe g-1. Dilution series were prepared with myocardium and iron (no-fat probes) and with 5% fat (fat probes). The tubes were positioned in a chest phantom and were scanned with a calcium-scoring protocol on a PCD-CT. A re-parameterized 3-material decomposition was used to separate iron and fat from myocardium.</p><p><strong>Results: </strong>On virtual monoenergetic images, attenuation increased linearly with iron concentrations in both fat and no-fat probes. In no-fat probes, linear regression yielded a slope of 1.2 HU (mg Fe g-1)-1 with an intercept of 35.8 HU (R2 = 0.964). In the fat probes, the slope was similar at 1.1 HU (mg Fe g-1)-1, while the regression line shifted downwards by 6.1 HU with an intercept of 29.6 HU (R2 = 0.985). Iron maps separated fat from iron with calculated median fat fractions of 4.85 in the fat and 0.90 in the no-fat probes. In iron images, attenuation increased linearly with increasing iron concentrations, with similar slopes between fat and no-fat probes and negligible differences in the intercept.</p><p><strong>Conclusions: </strong>Experimental evidence indicates the feasibility and accuracy of PCD-CT for iron and fat quantification in the myocardium. Iron-specific 3-material decomposition eliminates the confounding effect of fat on myocardial iron quantification.</p><p><strong>Advances in knowledge: </strong>This study highlights the value of dual-energy CT with 3-material decomposition for quantifying iron and fat in the myocardium. Thus, CT could serve as alternative for the current reference standard MRI.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"866-871"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13134854/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146046280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of a structured quantitative program on [18F]FP-CIT PET interpretation: a multicenter study of diagnostic accuracy, interrater reliability, and reader confidence.","authors":"Byungwook Choi, Seunggyun Ha, Sungmin Kang, Soo Jin Kwon, Junik Son, Shin Young Jeong, Sang-Woo Lee, Yeong Sim Choe, Donghyeon Kim, Chae Moon Hong","doi":"10.1093/bjr/tqag045","DOIUrl":"10.1093/bjr/tqag045","url":null,"abstract":"<p><strong>Objectives: </strong>This study assessed the impact of a structured quantitative analysis program on diagnostic performance, interrater reliability, and reader confidence in interpreting N-3-[18F]fluoropropyl-2β-carbomethoxy-3β-4-iodophenyl nortropane positron emission tomography ([18F]FP-CIT PET) scans to differentiate idiopathic Parkinson's disease from essential tremor-/drug-induced parkinsonism.</p><p><strong>Methods: </strong>This retrospective multicenter study included 637 patients who underwent [18F]FP-CIT PET and brain magnetic resonance imaging scans for evaluation of Parkinsonian syndromes at 4 institutions between January 2015 and October 2022. Readers evaluated the scans qualitatively in the pre-program session and with access to quantitative parameters in the post-program session, with a 1-month minimum interval to minimize recall bias. Diagnostic performance metrics (sensitivity, specificity, accuracy), interrater reliability (Cohen's κ), and reader confidence (low/moderate vs high) were compared.</p><p><strong>Results: </strong>Following program implementation, interrater reliability significantly improved from substantial to almost perfect (Cohen's κ: 0.768-0.970; P < .001), and a significant shift toward high-confidence ratings was observed (P < .001). Diagnostic accuracy also significantly improved from 92.8% to 94.3% (P = .036). Although the increase in the area under the curve (0.922-0.932) did not reach statistical significance (P = .163), likely due to a ceiling effect, reader performance significantly surpassed that of the quantitative parameter alone only after program implementation, whereas it was comparable at baseline.</p><p><strong>Conclusions: </strong>These findings suggest that integrating quantitative analysis tools may improve diagnostic consistency and reliability in clinical practice.</p><p><strong>Advances in knowledge: </strong>The structured quantitative analysis program significantly improved [18F]FP-CIT PET/CT interpretation by increasing diagnostic accuracy, interrater reliability, and reader confidence.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"980-988"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147289246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of tumour volume on outcomes in patients with locally advanced non-small cell lung cancer receiving chemoradiotherapy and consolidation durvalumab.","authors":"Vincent Legendre, Angela Botticella, Kristi Beshiri, Pernelle Lavaud, Mihaela Aldea, Jérôme Gambini, Cécile Le Péchoux, Antonin Levy","doi":"10.1093/bjr/tqag048","DOIUrl":"10.1093/bjr/tqag048","url":null,"abstract":"<p><strong>Objectives: </strong>The standard of care for unresectable locally advanced non-small-cell lung cancer (LA-NSCLC) is chemoradiotherapy (CRT) followed by consolidation immunotherapy. We analysed clinical, biological, and dosimetric data to identify survival predictors and inform personalized treatment strategies.</p><p><strong>Methods: </strong>We retrospectively reviewed LA-NSCLC patients treated between December 2015 and January 2023. Data were extracted from electronic medical records. Progression-free survival (PFS) and overall survival (OS) were calculated from diagnosis. Kaplan-Meier estimates and Cox proportional hazards models were used for survival analysis.</p><p><strong>Results: </strong>We included 92 patients (median follow-up: 28.9 months); 66% were male, 60% former smokers, and the median age was 64. Stage IIIB (50%) and adenocarcinoma (60%) were most common. PD-L1 positivity was observed in 75%, and 27% had oncogenic drivers (KRAS: 24%, EGFR: 3%). Most (86%) received concurrent CRT, typically cisplatin-based (60%). Median durvalumab duration was 9.5 months. Median PFS and OS were 31.9 and 51.2 months, respectively. Overall, 47% of patients experienced recurrence, including 15% locoregional recurrence and 85% metastatic recurrence (60% oligometastatic and 25% polymetastatic). PFS was longer in patients aged <66 years (45.4 vs 16.0 months; P = .01), with pre-RT lymphocyte counts >1.80 G/L (45.4 vs 27.7 months; P = .047), or PTV <450 cc (44.3 vs 16.1 months; P = .01). On multivariate analysis, only radiotherapy planning target volume (PTV) <450 cc remained significant (HR: 0.45; P = .04). No factor was associated with OS in multivariable analysis.</p><p><strong>Conclusions: </strong>Smaller PTV was independently associated with longer PFS, supporting its value in personalized care for LA-NSCLC.</p><p><strong>Advances in knowledge: </strong>PTV emerged as an independent prognostic factor for PFS, suggesting its utility in risk stratification and individualized treatment planning.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":"999-1005"},"PeriodicalIF":3.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147462568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cardiac Devices Audit Evaluating the use of 6 MV FFF beams following Updated Radiotherapy Board Guidance.","authors":"E Russell, D Paynter, P Bownes, V Cosgrove","doi":"10.1093/bjr/tqag097","DOIUrl":"https://doi.org/10.1093/bjr/tqag097","url":null,"abstract":"<p><strong>Objectives: </strong>Following guidance from The Radiotherapy Board for radiotherapy of patients with cardiac implantable electronic devices (CIED), a local audit of retrospective patient treatments was carried out. The guidance includes recommendations to avoid direct irradiation of the devices and avoidance of Flattening Filter Free (FFF) beams due to the high dose rates associated with FFF beams, leading to higher risk of device malfunction.</p><p><strong>Methods: </strong>Linac dose-rates were measured by noting the dose rate of the linac whilst delivering 6 MV FFF treatment plans, the local standard energy used for Volumetric Modulated Arc Therapy (VMAT). Plans were also assessed to determine whether treatment beams travel through the CIED.</p><p><strong>Results: </strong>It was found that 1 out of 540 patients with CIEDs was treated with the CIED receiving a dose rate above 8 Gy/minute for a 6 MV FFF beam. The plan for this treatment did not have the 'protect function' for CIEDs, which is now standard planning practice.</p><p><strong>Conclusions: </strong>The dose rate when using 6 MV FFF beams is sufficiently reduced, and therefore we have confidence in our current practice for the protection of patients with CIEDs.</p><p><strong>Advances in knowledge: </strong>This technical note responds to recent guidance published on the safety of radiotherapy exposures for patients with cardiac devices. It audits the standard practices in a large NHS Trust, assessing whether these standard practices align with the guidance. This will be of interest to many other trusts aspiring to adhere to this guidance.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147763249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Noninvasive prediction of coronary artery disease progression using pericoronary adipose tissue radiomics from coronary CTA.","authors":"Mengyuan Jing, Haoxiang Lu, Qing Liu, Huaze Xi, Hao Zhu, Qiu Sun, Yuting Zhang, Jialiang Ren, Wei Ren, Liangna Deng, Tao Han, Bin Zhang, Junlin Zhou","doi":"10.1093/bjr/tqag089","DOIUrl":"https://doi.org/10.1093/bjr/tqag089","url":null,"abstract":"<p><strong>Objective: </strong>Coronary artery disease (CAD) progression is directly associated with major adverse cardiovascular events and death. This study aimed to construct a pericoronary adipose tissue (PCAT) radiomics model to predict subsequent progression in patients with CAD.</p><p><strong>Methods: </strong>Data from 116 patients who had at least two coronary computed tomography angiography (CCTA) exams between March 1, 2020, and August 30, 2022, were collected at our institution. Obstructive stenosis, CAD-RADS classification, segment involvement score (SIS), and segment stenosis score (SSS) were noted. The radiomics features of the proximal to the left anterior descending artery, left circumflex artery, and right coronary artery were extracted on CCTA images using fully automated software. According to CAD-RADS, SIS, and SSS, non-progression was identified in 96, 80, and 72 patients and progression was identified in 20, 36, and 44 patients, respectively. All patients were randomly divided into the training and testing cohorts in a 7:3 ratio. Cox regression models were constructed based on PCAT radiomics signatures, and their predictive abilities were measured using receiver operating characteristic curves.</p><p><strong>Results: </strong>We included 116 patients (age 58.00 [53.25, 64.00] years; 78 [67.20%] were male). After screening, 16 PCAT radiomics features were identified as being significantly related to CAD progression. The Cox regression models had area under the curve values of 0.841, 0.838, and 0.725 in the training cohort and 0.818, 0.817, and 0.851 in the testing cohort, respectively, to predict 2-year CAD-RADS, SIS, and SSS progression.</p><p><strong>Conclusions: </strong>PCAT-based radiomics models demonstrated promising performance in predicting subsequent CAD progression.</p><p><strong>Advances in knowledge: </strong>PCAT-based radiomics signatures derived from coronary CT angiography provided incremental predictive value for CAD progression beyond conventional imaging markers (CAD-RADS, SIS, SSS), and may serve as noninvasive imaging biomarkers for individualized risk stratification.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147763197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coronary artery disease imaging with total-body PET.","authors":"Laura E Clark, Craig Balmforth, Matthew Gil, Jacek Kwiecinski, Piotr Slomka, David E Newby, Marc R Dweck, Michelle C Williams","doi":"10.1093/bjr/tqag093","DOIUrl":"https://doi.org/10.1093/bjr/tqag093","url":null,"abstract":"<p><p>Coronary artery positron emission tomography (PET) is a promising imaging tool that provides mechanistic insights into coronary artery disease, as well as clinical risk stratification. However, conventional PET systems remain limited by poor sensitivity and low signal to noise ratios (SNRs), hampering the robust assessment of coronary artery disease. The recent introduction of total-body PET scanners has enabled the simultaneous acquisition of data across multi-organ systems, with major enhancements in sensitivity and resultant effective resolution. These technological advancements facilitate the detection of lesions with lower radiotracer uptake signals and allow the complex systemic processes contributing to coronary artery disease to be unveiled, with important implications for clinical trial design and novel drug development. The aim of this review is to discuss the current literature on coronary artery PET and to explore how total-body PET systems can be employed to expand our current understanding of coronary artery disease.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147763190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning models to predict mammographic density jointly on standard dose and low dose images.","authors":"Steven Squires, Alistair Mackenzie, D Gareth Evans, Sacha J Howell, Susan M Astley","doi":"10.1093/bjr/tqag090","DOIUrl":"https://doi.org/10.1093/bjr/tqag090","url":null,"abstract":"<p><strong>Objectives: </strong>Mammographic density is associated with increased risk of developing breast cancer. Automated estimation of density in women below normal screening age would enable earlier risk stratification. We are piloting the use of low dose mammograms at 10% of standard dose combined with models that make accurate mammographic density estimates.</p><p><strong>Methods: </strong>Three models were trained on a joint set (107,619) of standard dose mammograms with associated density scores and their simulated low dose counterparts such that the models made predictions on standard and low dose mammograms. A second set of models was trained separately on the standard and simulated low dose mammograms. All models were tested on a held-out set from the training data and an independent dataset with 294 pairs of standard and real low dose mammograms.</p><p><strong>Results: </strong>The root mean squared errors (RMSE) between model predictions and density scores on standard and simulated low dose images were 8.26 (8.16-8.36) and 8.27 (8.17-8.38) respectively. The RMSE between predictions on standard and simulated low dose images for the jointly trained models was 1.91 (1.88-1.96). The RMSE of the predictions on real low dose images compared to standard dose images was 3.79 (2.75-4.99).</p><p><strong>Conclusions: </strong>Deep learning models make density predictions on low dose images with similar quality as on standard dose images. Automated analysis of low dose mammograms could contribute to accurate breast cancer risk estimation in younger women enabling stratification for further monitoring and preventative therapy.</p><p><strong>Advances in knowledge: </strong>Mammographic density can be estimated in low dose mammograms with similar quality to standard dose mammograms.</p>","PeriodicalId":9306,"journal":{"name":"British Journal of Radiology","volume":" ","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147763199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}