Search

Background Deep learning (DL) algorithms have shown promising results in mammographic screening either compared to a single reader or, when deployed in conjunction with a human reader, compared with double reading. Purpose To externally validate the performance of three DL algorithms as mammographic screen readers in an independent UK data set. Materials and Methods Three commercial DL algorithms (DL-1, DL-2, and DL-3) were retrospectively investigated from January 2022 to June 2022 using consecutive full-field digital mammograms collected at two UK sites during 1 year (2017). Normal cases with 3-year follow-up and histopathologically proven cancer cases detected either at screening (that round or next) or within the 3-year interval were included. A preset specificity threshold equivalent to a single reader was applied. Performance was evaluated for stand-alone DL reading compared with single human reading, and for DL reading combined with human reading compared with double reading, using sensitivity and specificity as the primary metrics. P < .025 was considered to indicate statistical significance for noninferiority testing. Results A total of 26 722 cases (median patient age, 59.0 years [IQR, 54.0-63.0 years]) with mammograms acquired using machines from two vendors were included. Cases included 332 screen-detected, 174 interval, and 254 next-round cancers. Two of three stand-alone DL algorithms achieved noninferior sensitivity (DL-1: 64.8%, P < .001; DL-2: 56.7%, P = .03; DL-3: 58.9%, P < .001) compared with the single first reader (62.8%), and specificity was noninferior for DL-1 (92.8%; P < .001) and DL-2 (96.8%; P < .001) and superior for DL-3 (97.9%; P < .001) compared with the single first reader (96.5%). Combining the DL algorithms with human readers achieved noninferior sensitivity (67.0%, 65.6%, and 65.4% for DL-1, DL-2, and DL-3, respectively; P < .001 for all) compared with double reading (67.4%), and superior specificity (97.4%, 97.6%, and 97.6%; P < .001 for all) compared with double reading (97.1%). Conclusion Use of stand-alone DL algorithms in combination with a human reader could maintain screening accuracy while reducing workload. Published under a CC BY 4.0 license. Supplemental material is available for this article.

Background: Integrating artificial intelligence (AI) into mammography screening can support radiologists and improve programme metrics, yet the potential of different strategies for integrating the technology remains understudied. We compared programme-level performance metrics of seven AI integration strategies., Methods: We performed a retrospective comparative evaluation of seven strategies for integrating AI into mammography screening using datasets generated from screening programmes in Germany (n=1 657 068), the UK (n=223 603) and Sweden (n=22 779). The commercially available AI model used was Vara version 2.10, trained from scratch on German data. We simulated the performance of each strategy in terms of cancer detection rate (CDR), recall rate, and workload reduction, and compared the metrics with those of the screening programmes. We also assessed the distribution of the stages and grades of the cancers detected by each strategy and the AI model's ability to correctly localise those cancers., Findings: Compared with the German screening programme (CDR 6·32 per 1000 examinations, recall rate 4·11 per 100 examinations), replacement of both readers (standalone AI strategy) achieved a non-inferior CDR of 6·37 (95% CI 6·10-6·64) at a recall rate of 3·80 (95% CI 3·67-3·93), whereas single reader replacement achieved a CDR of 6·49 (6·31-6·67), a recall rate of 4·01 (3·92-4·10), and a 49% workload reduction. Programme-level decision referral achieved a CDR of 6·85 (6·61-7·11), a recall rate of 3·55 (3·43-3·68), and an 84% workload reduction. Compared with the UK programme CDR of 8·19, the reader-level, programme-level, and deferral to single reader strategies achieved CDRs of 8·24 (7·82-8·71), 8·59 (8·12-9·06), and 8·28 (7·86-8·71), without increasing recall and while reducing workload by 37%, 81%, and 95%, respectively. On the Swedish dataset, programme-level decision referral increased the CDR by 17·7% without increasing recall and while reducing reading workload by 92%., Interpretation: The decision referral strategies offered the largest improvements in cancer detection rates and reduction in recall rates, and all strategies except normal triaging showed potential to improve screening metrics., Funding: Vara., Competing Interests: Declaration of interests ZVF, MB, TM, VŠ, SB, and CL are current or former employees of Vara, with stock options as part of the standard compensation package. FJG, NRP, and SEH have research collaborations with Vara, ScreenPoint, Lunit, Google, Volpara, iCAD, Therapixel, CureMetrix, Sunnybrook Research Institute, and Massachusetts Institute of Technology., (Copyright © 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.)

Aims: To establish the safety and feasibility of delivering neoadjuvant radiotherapy and endocrine therapy for oestrogen receptor-positive breast cancers with palpable size 20mm or greater, for which radiotherapy might facilitate more conservative surgery., Materials and Methods: A single-arm feasibility study was conducted. Patients received whole breast radiotherapy with or without radiotherapy to nodal areas. Dose/fractionation was 40Gy in 15 fractions over 3 weeks, with or without either a simultaneous integrated boost to 48Gy or sequential boost to the tumour bed. This was followed by endocrine treatment for 20 weeks, then surgery. The primary endpoint of the study was the proportion of patients successfully completing neoadjuvant radiotherapy and endocrine treatment followed by breast surgery. Response and toxicity endpoints including mastectomy rate, peri/postoperative complications, and pathological response were also evaluated. The primary analysis is descriptive. The study regimen would be considered feasible if more than 70% of patients completed treatment, while it might not be considered feasible if less than 50% did so. With a one-sided 5% significance level and 80% power, a maximum of 43 patients would be required to detect a rate of ≤50% vs ≥70%., Results: 14 patients were recruited out of the planned 43. Due to slow recruitment, particularly during the COVID-19 pandemic, the decision was made to stop the trial in October 2021. One registered patient was found to be ineligible before starting treatment. 13/13 patients (100%, 90% CI: 75.3%, 100%) who received any trial treatment successfully completed all trial treatments. The lower bound of the Clopper-Pearson (exact) 90% confidence interval was 79%, indicating that the primary endpoint would have been met if the planned recruitment had been achieved. 3/13 patients underwent mastectomy. 7/13 had more conservative surgery than had been planned at baseline. 4/13 patients experienced any peri/postoperative complication. The only acute radiotherapy toxicities reported were grade 1/2 dermatitis and grade 1 fatigue. Long-term breast outcomes were clinician assessed as none/mild at all timepoints in 12/13 patients. All tumours showed evidence of some pathological response to treatment, but none had a pathological complete response., Conclusion: This treatment schedule is likely feasible. It is difficult to draw strong conclusions on safety/toxicity given the small numbers, but these seem in keeping with other recent reports of neoadjuvant breast radiotherapy., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Population screening for breast cancer (BC) is currently offered in the UK for women aged 50 to 71 with the aim of reducing mortality. There is additional screening within the national programme for women identified as having a very high risk of BC. There is growing interest in further risk stratification in breast screening, which would require a whole population risk assessment and the subsequent offer of screening tailored to the individual's risk. Some women would be offered more intensive screening than others or no screening. This might provide a better balance of screening benefits and harms for each individual than the current population age-based programme alone. The UK National Screening Committee (UK NSC) is considering using decision-analytic and other models to evaluate different risk stratification screening strategies and identify remaining gaps in evidence. This paper reports the proceedings of a UK NSC workshop where experts in the field discussed both risk prediction models, as well as decision-analytic models providing a benefit-harm analysis/economic evaluation of risk-stratified screening programmes (see Table 1). The aim of the meeting was to present and discuss the current work of experts, including some data which had not been published at the time of the meeting, to inform the UK NSC. The workshop was not intended to present a balanced evaluation of how to deliver screening in future. Areas for further work identified included methods for comparing models to assess accuracy, the optimum risk assessment tools, the digital screening infrastructure, acceptability of stratification, choice of screening test and reducing inequalities. A move to risk stratification of the whole programme would require a careful phased introduction with continuing assessment of real-world evidence during deployment., (© 2024. The Author(s).)

Digital breast tomosynthesis (DBT) with full-field digital mammography (FFDM) exposes women to a higher radiation dose. A synthetic 2D mammogram (S2D) is a two-dimensional image constructed from DBT. We aim to evaluate the S2D performance when used alone or combined with DBT compared to FFDM alone or with DBT. Studies were included if they recruited screening participants and reported on S2D performance. Studies were excluded if they included symptomatic patients, imaging was for diagnostic purposes, or if participants had a breast cancer history. Meta-analyses for cancer detection rates (CDR) and Specificities were conducted where available. Differences in the performance of imaging modalities were calculated within individual studies, and these were pooled by meta-analysis. Out of 3241 records identified, 17 studies were included in the review and 13 in the meta-analysis. The estimated combined difference in CDRs per thousand among individual studies that reported on DBT plus S2D vs. FFDM and those reporting on DBT plus S2D versus DBT plus FFDM was 2.03 (95% CI 0.81-3.25) and - 0.15 (95% CI -1.17 to 0.86), respectively. The estimated difference in percent specificities was 1.13 (95% CI -0.06 to 2.31) in studies comparing DBT plus S2D and FFDM. In studies comparing DBT plus S2D and DBT plus FFDM, the estimated difference in specificities was 1.08 (95% CI 0.59-1.56). DBT plus S2D showed comparable accuracy to FFDM plus DPT and improved cancer detection to FFDM alone. Integrating S2D with DBT in breast cancer screening is safe and preserves performance., (© 2024 The Author(s). International Journal of Cancer published by John Wiley & Sons Ltd on behalf of UICC.)

Background Artificial intelligence (AI) systems can be used to identify interval breast cancers, although the localizations are not always accurate. Purpose To evaluate AI localizations of interval cancers (ICs) on screening mammograms by IC category and histopathologic characteristics. Materials and Methods A screening mammography data set (median patient age, 57 years [IQR, 52-64 years]) that had been assessed by two human readers from January 2011 to December 2018 was retrospectively analyzed using a commercial AI system. The AI outputs were lesion locations (heatmaps) and the highest per-lesion risk score (range, 0-100) assigned to each case. AI heatmaps were considered false positive (FP) if they occurred on normal screening mammograms or on IC screening mammograms (ie, in patients subsequently diagnosed with IC) but outside the cancer boundary. A panel of consultant radiology experts classified ICs as normal or benign (true negative [TN]), uncertain (minimal signs of malignancy [MS]), or suspicious (false negative [FN]). Several specificity and sensitivity thresholds were applied. Mann-Whitney U tests, Kruskal-Wallis tests, and χ 2 tests were used to compare groups. Results A total of 2052 screening mammograms (514 ICs and 1548 normal mammograms) were included. The median AI risk score was 50 (IQR, 32-82) for TN ICs, 76 (IQR, 41-90) for ICs with MS, and 89 (IQR, 81-95) for FN ICs ( P = .005). Higher median AI scores were observed for invasive tumors (62 [IQR, 39-88]) than for noninvasive tumors (33 [IQR, 20-55]; P < .01) and for high-grade (grade 2-3) tumors (62 [IQR, 40-87]) than for low-grade (grade 0-1) tumors (45 [IQR, 26-81]; P = .02). At the 96% specificity threshold, the AI algorithm flagged 121 of 514 (23.5%) ICs and correctly localized the IC in 93 of 121 (76.9%) cases, with 48 FP heatmaps on the mammograms for ICs (rate, 0.093 per case) and 74 FP heatmaps on normal mammograms (rate, 0.048 per case). The AI algorithm correctly localized a lower proportion of TN ICs (54 of 427; 12.6%) than ICs with MS (35 of 76; 46%) and FN ICs (four of eight; 50% [95% CI: 13, 88]; P < .001). The AI algorithm localized a higher proportion of node-positive than node-negative cancers ( P = .03). However, no evidence of a difference by cancer type ( P = .09), grade ( P = .27), or hormone receptor status ( P = .12) was found. At 89.8% specificity and 79% sensitivity thresholds, AI detection increased to 181 (35.2%) and 256 (49.8%) of the 514 ICs, respectively, with FP heatmaps on 158 (10.2%) and 307 (19.8%) of the 1548 normal mammograms. Conclusion Use of a standalone AI system improved early cancer detection by correctly identifying some cancers missed by two human readers, with no differences based on histopathologic features except for node-positive cancers. © RSNA, 2024 Supplemental material is available for this article .

Breast complaints are frequent reasons for consultations in primary care or breast clinics. Breast pain, breast lumps, and nipple discharge are the most common complaints. Less common symptoms such as skin changes and axillary abnormalities also require specific diagnostic approaches. Imaging the symptomatic breast should be performed by appropriately trained breast radiologists following the best practice guidelines and quality standards. Full-field digital mammography (FFDM), digital breast tomosynthesis (DBT), and breast ultrasound (US) are the main modalities used in this primary setting. The choice depends on the patient's age and symptoms. Women younger than 30-years-old are first imaged by US, whereas women over 40-years-old usually require both FFDM or DBT and US. For women between 30-years-old and 40-years-old, the US is the modality of choice, whereas FFDM or DBT might also be performed if needed. Pregnant or lactating women with palpable lesions or nipple discharge are imaged with US as the first method; FFDM or DBT can also be performed depending on the degree of suspicion as the dose to the fetus is minimal, and shielding may even further reduce the dose. More advanced techniques such as breast magnetic resonance imaging or contrast-enhanced mammography are not indicated in this first diagnostic setting and are reserved for cases of established malignancy (local staging) or rare cases of equivocal findings not otherwise resolved or inflammatory breast cancer. Last, but not least, male breast symptoms should also be addressed with US and/or FFDM. CLINICAL RELEVANCE STATEMENT: It is equally important to correctly diagnose an underlying malignancy and to avoid false positives that would lead to unnecessary biopsies, increased costs, and anxiety for the patient. Proper use of imaging modalities ensures optimal diagnostic approach and minimizes false negatives. KEY POINTS: Ultrasound, full-field digital mammography, or digital breast tomosynthesis are the main imaging modalities in the diagnostic setting, while MRI or contrast-enhanced mammography should be reserved to selected cases. Initial imaging modality includes ultrasound combined with mammography or digital breast tomosynthesis depending on women's age and the presence (or not) of inconclusive findings. A negative imaging evaluation should not deter biopsy when a highly suspicious finding is found on physical examination., (© 2024. The Author(s), under exclusive licence to European Society of Radiology.)

Objectives: To assess the performance of breast cancer screening by category of breast density and age in a UK screening cohort., Methods: Raw full-field digital mammography data from a single site in the UK, forming a consecutive 3-year cohort of women aged 50 to 70 years from 2016 to 2018, were obtained retrospectively. Breast density was assessed using Volpara software. Examinations were grouped by density category and age group (50-60 and 61-70 years) to analyse screening performance. Statistical analysis was performed to determine the association between density categories and age groups. Volumetric breast density was assessed as a binary classifier of interval cancers (ICs) to find an optimal density threshold., Results: Forty-nine thousand nine-hundred forty-eight screening examinations (409 screen-detected cancers (SDCs) and 205 ICs) were included in the analysis. Mammographic sensitivity, SDC/(SDC + IC), decreased with increasing breast density from 75.0% for density a (p = 0.839, comparisons made to category b), to 73.5%, 59.8% (p = 0.001), and 51.3% (p < 0.001) in categories b, c, and d, respectively. IC rates were highest in the densest categories with rates of 1.8 (p = 0.039), 3.2, 5.7 (p < 0.001), and 7.9 (p < 0.001) per thousand for categories a, b, c, and d, respectively. The recall rate increased with breast density, leading to more false positive recalls, especially in the younger age group. There was no significant difference between the optimal density threshold found, 6.85, and that Volpara defined as the b/c boundary, 7.5., Conclusions: The performance of screening is significantly reduced with increasing density with IC rates in the densest category four times higher than in women with fatty breasts. False positives are a particular issue for the younger subgroup without prior examinations., Clinical Relevance Statement: In women attending screening there is significant underdiagnosis of breast cancer in those with dense breasts, most marked in the highest density category but still three times higher than in women with fatty breasts in the second highest category., Key Points: Breast density can mask cancers leading to underdiagnosis on mammography. Interval cancer rate increased with breast density categories 'a' to 'd'; 1.8 to 7.9 per thousand. Recall rates increased with increasing breast density, leading to more false positive recalls., (© 2024. The Author(s).)

Background: To study the reproducibility of 23 Na magnetic resonance imaging (MRI) measurements from breast tissue in healthy volunteers., Methods: Using a dual-tuned bilateral 23 Na/ 1 H breast coil at 3-T MRI, high-resolution 23 Na MRI three-dimensional cones sequences were used to quantify total sodium concentration (TSC) and fluid-attenuated sodium concentration (FASC). B 1 -corrected TSC and FASC maps were created. Two readers manually measured mean, minimum and maximum TSC and mean FASC values using two sampling methods: large regions of interest (LROIs) and small regions of interest (SROIs) encompassing fibroglandular tissue (FGT) and the highest signal area at the level of the nipple, respectively. The reproducibility of the measurements and correlations between density, age and FGT apparent diffusion coefficient (ADC) values were evaluatedss., Results: Nine healthy volunteers were included. The inter-reader reproducibility of TSC and FASC using SROIs and LROIs was excellent (intraclass coefficient range 0.945-0.979, p < 0.001), except for the minimum TSC LROI measurements (p = 0.369). The mean/minimum LROI TSC and mean LROI FASC values were lower than the respective SROI values (p < 0.001); the maximum LROI TSC values were higher than the SROI TSC values (p = 0.009). TSC correlated inversely with age but not with FGT ADCs. The mean and maximum FGT TSC and FASC values were higher in dense breasts in comparison to non-dense breasts (p < 0.020)., Conclusions: The chosen sampling method and the selected descriptive value affect the measured TSC and FASC values, although the inter-reader reproducibility of the measurements is in general excellent., Relevance Statement: 23 Na MRI at 3 T allows the quantification of TSC and FASC sodium concentrations. The sodium measurements should be obtained consistently in a uniform manner., Key Points: • 23 Na MRI allows the quantification of total and fluid-attenuated sodium concentrations (TSC/FASC). • Sampling method (large/small region of interest) affects the TSC and FASC values. • Dense breasts have higher TSC and FASC values than non-dense breasts. • The inter-reader reproducibility of TSC and FASC measurements was, in general, excellent. • The results suggest the importance of stratifying the sodium measurements protocol., (© 2024. The Author(s).)

Objectives: To investigate the influence of preoperative breast MRI on mastectomy and reoperation rates in patients with pure ductal carcinoma in situ (DCIS)., Methods: The MIPA observational study database (7245 patients) was searched for patients aged 18-80 years with pure unilateral DCIS diagnosed at core needle or vacuum-assisted biopsy (CNB/VAB) and planned for primary surgery. Patients who underwent preoperative MRI (MRI group) were matched (1:1) to those who did not receive MRI (noMRI group) according to 8 confounding covariates that drive referral to MRI (age; hormonal status; familial risk; posterior-to-nipple diameter; BI-RADS category; lesion diameter; lesion presentation; surgical planning at conventional imaging). Surgical outcomes were compared between the matched groups with nonparametric statistics after calculating odds ratios (ORs)., Results: Of 1005 women with pure unilateral DCIS at CNB/VAB (507 MRI group, 498 noMRI group), 309 remained in each group after matching. First-line mastectomy rate in the MRI group was 20.1% (62/309 patients, OR 2.03) compared to 11.0% in the noMRI group (34/309 patients, p = 0.003). The reoperation rate was 10.0% in the MRI group (31/309, OR for reoperation 0.40) and 22.0% in the noMRI group (68/309, p < 0.001), with a 2.53 OR of avoiding reoperation in the MRI group. The overall mastectomy rate was 23.3% in the MRI group (72/309, OR 1.40) and 17.8% in the noMRI group (55/309, p = 0.111)., Conclusions: Compared to those going directly to surgery, patients with pure DCIS at CNB/VAB who underwent preoperative MRI had a higher OR for first-line mastectomy but a substantially lower OR for reoperation., Clinical Relevance Statement: When confounding factors behind MRI referral are accounted for in the comparison of patients with CNB/VAB-diagnosed pure unilateral DCIS, preoperative MRI yields a reduction of reoperations that is more than twice as high as the increase in overall mastectomies., Key Points: • Confounding factors cause imbalance when investigating the influence of preoperative MRI on surgical outcomes of pure DCIS. • When patient matching is applied to women with pure unilateral DCIS, reoperation rates are significantly reduced in women who underwent preoperative MRI. • The reduction of reoperations brought about by preoperative MRI is more than double the increase in overall mastectomies., (© 2023. The Author(s).)

In the mid-1990s, the identification of BRCA1/2 genes for breast cancer susceptibility led to testing breast MRI accuracy in screening women at increased risk. From 2000 onwards, ten intraindividual comparative studies showed the marked superiority of MRI: the sensitivity ranged 25-58% for mammography, 33-52% for ultrasound, 48-67% for mammography plus ultrasound, and 71-100% for MRI; specificity 93-100%, 91-98%, 89-98%, and 81-98%, respectively. Based on the available evidence, in 2006-2007, the UK National Institute for Clinical Excellence and the American Cancer Society recommended MRI screening of high-risk women, followed by other international guidelines. Despite evidence-based medicine ideally requiring randomised controlled trials (RCTs) for policy changes regarding screening procedures, breast MRI for high-risk screening was adopted in many countries worldwide. In 2019, the results of the "DENSE" RCT were published in favour of breast MRI screening of women with extremely dense breasts compared to mammography alone, showing a reduction of more than 80% of the interval cancer rate in women who attended MRI screening. Even though international recommendations in favour of this practice were issued, substantial obstacles still prevent health systems from adopting breast MRI for screening women with extremely dense breasts. A paradox is evident: we adopted a screening procedure without evidence from RCTs, and now that we have this level-1 evidence for the same procedure, we fail to do so. This critical review tries to explain the differences between the two cases, as examples of the complex pathways of translating radiological research into everyday practice.Critical relevance statement The high-level evidence in favour of breast MRI screening of women with extremely dense breasts is failing to persuade policy makers to translate this into clinical practice.Key points• Breast MRI screening of high-risk women was adopted on basis of the evidence provided by test accuracy comparative studies showing an MRI performance greatly superior to that of mammography.• Breast MRI screening of women with extremely dense breasts has not been adopted although the evidence of a large reduction in interval cancer rate from a RCT.• We illustrate the differences between the two cases, as an example of the complex ways of translation of radiological research in clinical practice according to the EBM theory., (© 2024. The Author(s).)

Objectives: Quality assurance (QA) of image interpretation plays a key role in screening and diagnostic mammography, maintaining minimum standards and supporting continuous improvement in interpreting images. However, the QA structure across Europe shows considerable variation. The European Society of Breast Imaging (EUSOBI) conducted a survey among the members to collect information on radiologists' preferences regarding QA measures in mammography., Materials and Methods: An anonymous online survey consisting of 25 questions was distributed to all EUSOBI members and national breast radiology bodies in Europe. The questions were designed to collect demographic characteristics, information on responders' mammography workload and data about QA measures currently used in their country. Data was analysed using descriptive statistical analysis, the χ 2 test, linear regression, and Durbin-Watson statistic test., Results: In total, 251 breast radiologists from 34 countries completed the survey. Most respondents were providing both screening and symptomatic services (137/251, 54.6%), working in an academic hospital (85/251, 33.9%) and reading 1000-4999 cases per year (109/251, 43.4%). More than half of them (133/251, 53%) had established QA measures in their workplace. Although less than one-third (71/251, 28.3%) had to participate in regular performance testing, the vast majority (190/251, 75.7%) agreed that a mandatory test would be helpful to improve their skills., Conclusion: QA measures were in place for more than half of the respondents working in screening and diagnostic mammography to evaluate their breast imaging performance. Although there were substantial differences between countries, the importance of having QA in the workplace and implemented was widely acknowledged by radiologists., Clinical Relevance Statement: Although several quality assurance (QA) measures of image interpretation are recommended by European bodies or national organisations, the QA in mammography is quite heterogenous between countries and reporting settings, and not always actively implemented across Europe., Key Points: The first survey that presents radiologists' preferences regarding QA measures of image interpretation in mammography. Quality assurance measures in the workplace are better-established for breast screening compared to diagnostic mammography. Radiologists consider that performance tests would help to improve their mammography interpretation skills., (© 2023. The Author(s).)

Background Breast screening enables early detection of cancers; however, most women have normal mammograms, resulting in repetitive and resource-intensive reading tasks. Purpose To investigate if deep learning (DL) algorithms can be used to triage mammograms by identifying normal results to reduce workload or flag cancers that may be overlooked. Materials and Methods In this retrospective study, three commercial DL algorithms were investigated using consecutive mammograms from two UK Breast Screening Program sites from January 2015 to December 2017 and January 2017 to December 2018 on devices from two mammography vendors. Normal mammograms with a 3-year follow-up and histopathologically proven cancer detected at screening, the subsequent round, or in the 3-year interval were included. Two algorithm thresholds were set: in scenario A, 99.0% sensitivity for rule-out triage to a lone reader, and in scenario B, approximately 1.0% additional recall providing a rule-in triage for further assessment. Both thresholds were then applied to the screening workflow in scenario C. The sensitivity and specificity were used to assess the overall predictive performance of each DL algorithm. Results The data set comprised 78 849 patients (median age, 59 years [IQR, 53-63 years]) and 887 screening-detected, 439 interval, and 688 subsequent screening round-detected cancers. In scenario A (rule-out triage), models DL-1, DL-2, and DL-3 triaged 35.0% (27 565 of 78 849), 53.2% (41 937 of 78 849), and 55.6% (43 869 of 78 849) of mammograms, respectively, with 0.0% (0 of 887) to 0.1% (one of 887) of screening-detected cancers undetected. In scenario B, DL algorithms triaged in 4.6% (20 of 439) to 8.2% (36 of 439) of interval and 5.2% (36 of 688) to 6.1% (42 of 688) of subsequent-round cancers when applied after the routine double-reading workflow. Combining both approaches in scenario C resulted in an overall noninferior specificity (difference, -0.9%; P < .001) and superior sensitivity (difference, 2.7%; P < .001) for the adaptive workflow compared with routine double reading for all three algorithms. Conclusion Rule-out and rule-in DL-adapted triage workflows can improve the efficiency and efficacy of mammography breast cancer screening. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Nishikawa and Lu in this issue.

Cross-institution collaborations are constrained by data-sharing challenges. These challenges hamper innovation, particularly in artificial intelligence, where models require diverse data to ensure strong performance. Federated learning (FL) solves data-sharing challenges. In typical collaborations, data is sent to a central repository where models are trained. With FL, models are sent to participating sites, trained locally, and model weights aggregated to create a master model with improved performance. At the 2021 Radiology Society of North America's (RSNA) conference, a panel was conducted titled "Accelerating AI: How Federated Learning Can Protect Privacy, Facilitate Collaboration and Improve Outcomes." Two groups shared insights: researchers from the EXAM study (EMC CXR AI Model) and members of the National Cancer Institute's Early Detection Research Network's (EDRN) pancreatic cancer working group. EXAM brought together 20 institutions to create a model to predict oxygen requirements of patients seen in the emergency department with COVID-19 symptoms. The EDRN collaboration is focused on improving outcomes for pancreatic cancer patients through earlier detection. This paper describes major insights from the panel, including direct quotes. The panelists described the impetus for FL, the long-term potential vision of FL, challenges faced in FL, and the immediate path forward for FL., Competing Interests: Declaration of conflicting interestsThe author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Accumulating evidence from retrospective studies demonstrate at least non-inferior performance when using AI algorithms with different strategies versus double-reading in mammography screening. In addition, AI algorithms for mammography screening can reduce work load by moving to single human reading. Prospective trials are essential to avoid unintended adverse consequences before incorporation of AI algorithms into UK's National Health Service (NHS) Breast Screening Programme (BSP). A stakeholders' meeting was organized in Newnham College, Cambridge, UK to undertake a review of the current evidence to enable consensus discussion on next steps required before implementation into a screening programme. It was concluded that a multicentre multivendor testing platform study with opt-out consent is preferred. AI thresholds from different vendors should be determined while maintaining non-inferior screening performance results, particularly ensuring recall rates are not increased. Automatic recall of cases using an agreed high sensitivity AI score versus automatic rule out with a low AI score set at a high sensitivity could be used. A human reader should still be involved in decision making with AI-only recalls requiring human arbitration. Standalone AI algorithms used without prompting maintain unbiased screening reading performance, but reading with prompts should be tested prospectively and ideally provided for arbitration., Competing Interests: Declaration of Competing Interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: [Support for this meeting was from Newnham College, University of Cambridge, Cambridge, UK. F.J.G. is supported by a Cancer Research UK early detection programme grant. H.A. is Chief Scientific Officer, Preemptive Medicine and Health, Flagship Pioneering.]., (Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.)

Objectives: To report mastectomy and reoperation rates in women who had breast MRI for screening (S-MRI subgroup) or diagnostic (D-MRI subgroup) purposes, using multivariable analysis for investigating the role of MRI referral/nonreferral and other covariates in driving surgical outcomes., Methods: The MIPA observational study enrolled women aged 18-80 years with newly diagnosed breast cancer destined to have surgery as the primary treatment, in 27 centres worldwide. Mastectomy and reoperation rates were compared using non-parametric tests and multivariable analysis., Results: A total of 5828 patients entered analysis, 2763 (47.4%) did not undergo MRI (noMRI subgroup) and 3065 underwent MRI (52.6%); of the latter, 2441/3065 (79.7%) underwent MRI with preoperative intent (P-MRI subgroup), 510/3065 (16.6%) D-MRI, and 114/3065 S-MRI (3.7%). The reoperation rate was 10.5% for S-MRI, 8.2% for D-MRI, and 8.5% for P-MRI, while it was 11.7% for noMRI (p ≤ 0.023 for comparisons with D-MRI and P-MRI). The overall mastectomy rate (first-line mastectomy plus conversions from conserving surgery to mastectomy) was 39.5% for S-MRI, 36.2% for P-MRI, 24.1% for D-MRI, and 18.0% for noMRI. At multivariable analysis, using noMRI as reference, the odds ratios for overall mastectomy were 2.4 (p < 0.001) for S-MRI, 1.0 (p = 0.957) for D-MRI, and 1.9 (p < 0.001) for P-MRI., Conclusions: Patients from the D-MRI subgroup had the lowest overall mastectomy rate (24.1%) among MRI subgroups and the lowest reoperation rate (8.2%) together with P-MRI (8.5%). This analysis offers an insight into how the initial indication for MRI affects the subsequent surgical treatment of breast cancer., Key Points: • Of 3065 breast MRI examinations, 79.7% were performed with preoperative intent (P-MRI), 16.6% were diagnostic (D-MRI), and 3.7% were screening (S-MRI) examinations. • The D-MRI subgroup had the lowest mastectomy rate (24.1%) among MRI subgroups and the lowest reoperation rate (8.2%) together with P-MRI (8.5%). • The S-MRI subgroup had the highest mastectomy rate (39.5%) which aligns with higher-than-average risk in this subgroup, with a reoperation rate (10.5%) not significantly different to that of all other subgroups., (© 2023. The Author(s).)

Objectives: To explore the relationship between indices of hypoxia and vascular function from 18 F-fluoromisonidazole ([ 18 F]-FMISO)-PET/MRI with immunohistochemical markers of hypoxia and vascularity in oestrogen receptor-positive (ER +) breast cancer., Methods: Women aged  > 18 years with biopsy-confirmed, treatment-naïve primary ER + breast cancer underwent [ 18 F]-FMISO-PET/MRI prior to surgery. Parameters of vascular function were derived from DCE-MRI using the extended Tofts model, whilst hypoxia was assessed using the [ 18 F]-FMISO influx rate constant, K i . Histological tumour sections were stained with CD31, hypoxia-inducible factor (HIF)-1α, and carbonic anhydrase IX (CAIX). The number of tumour microvessels, median vessel diameter, and microvessel density (MVD) were obtained from CD31 immunohistochemistry. HIF-1α and CAIX expression were assessed using histoscores obtained by multiplying the percentage of positive cells stained by the staining intensity. Regression analysis was used to study associations between imaging and immunohistochemistry variables., Results: Of the lesions examined, 14/22 (64%) were ductal cancers, grade 2 or 3 (19/22; 86%), with 17/22 (77%) HER2-negative. [ 18 F]-FMISO K i associated negatively with vessel diameter (p = 0.03), MVD (p = 0.02), and CAIX expression (p = 0.002), whilst no significant relationships were found between DCE-MRI pharmacokinetic parameters and immunohistochemical variables. HIF-1α did not significantly associate with any PET/MR imaging indices., Conclusion: Hypoxia measured by [ 18 F]-FMISO-PET was associated with increased CAIX expression, low MVD, and smaller vessel diameters in ER + breast cancer, further corroborating the link between inadequate vascularity and hypoxia in ER + breast cancer., Key Points: • Hypoxia, measured by [ 18 F]-FMISO-PET, was associated with low microvessel density and small vessel diameters, corroborating the link between inadequate vascularity and hypoxia in ER + breast cancer. • Increased CAIX expression was associated with higher levels of hypoxia measured by [ 18 F]-FMISO-PET. • Morphologic and functional abnormalities of the tumour microvasculature are the major determinants of hypoxia in cancers and support the previously reported perfusion-driven character of hypoxia in breast carcinomas., (© 2023. The Author(s).)

Photoacoustic imaging (PAI), also referred to as optoacoustic imaging, has shown promise in early-stage clinical trials in a range of applications from inflammatory diseases to cancer. While the first PAI systems have recently received regulatory approvals, successful adoption of PAI technology into healthcare systems for clinical decision making must still overcome a range of barriers, from education and training to data acquisition and interpretation. The International Photoacoustic Standardisation Consortium (IPASC) undertook an community exercise in 2022 to identify and understand these barriers, then develop a roadmap of strategic plans to address them. Here, we outline the nature and scope of the barriers that were identified, along with short-, medium- and long-term community efforts required to overcome them, both within and beyond the IPASC group., Competing Interests: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: The promotions and salary of Kurinchi Gurusamy are dependent upon journal publications. Ferdinand Knieling is co-inventor with iThera Medical GmbH, Germany on an EU patent (EP 19 163 304.9) relating to a device and a method for analysing optoacoustic data, an optoacoustic system and a computer program. He received travel support by iThera Medical GmbH, Germany, lecture fees from Sanofi Genzyme and Siemens Healthcare GmbH outside the submitted work. Given his role as Section Editor of this journal, Ferdinand Knieling had no involvement in the peer-review of articles for which he was an author and had no access to information regarding their peer-review. AAO is an employee and stock holder of TomoWave and Seno Medical, which develop and sell optoacoustic tomography systems. Given his role as the Editor in Chief of this Journal, AAO had no involvement in the peer-review process or the decisions regarding acceptance of manuscripts for which he is an author. Given her role as an Editorial Board member of this journal, Sarah Bohndiek had no involvement in the peer-review of articles for which she was an author and had no access to information regarding their peer-review. Full responsibility for the peer-review process for this article was delegated to another Editor. Martin Leahy is the inventor of WIPO patent (WO2015110349A1) relating to photoacoustic tomography method and system which uses known intrinsic absorbers for fluence correction. Lihong Wang has a financial interest in Microphotoacoustics, Inc., CalPACT, LLC, and Union Photoacoustic Technologies, Ltd., which, however, did not support this work. Ledia Lilaj is an employee of iThera Medical GmbH, a vendor of optoacoustic imaging instruments. Mithun Kuniyil Ajith Singh is an employee of CYBERDYNE Inc, a vendor of photoacoustic imaging instruments. The other authors do not declare any conflicts of interest., (© 2023 Published by Elsevier GmbH.)

Axillary lymphadenopathy is a common side effect of COVID-19 vaccination, leading to increased imaging-detected asymptomatic and symptomatic unilateral axillary lymphadenopathy. This has threatened to negatively impact the workflow of breast imaging services, leading to the release of ten recommendations by the European Society of Breast Imaging (EUSOBI) in August 2021. Considering the rapidly changing scenario and data scarcity, these initial recommendations kept a highly conservative approach. As of 2023, according to newly acquired evidence, EUSOBI proposes the following updates, in order to reduce unnecessary examinations and avoid delaying necessary examinations. First, recommendation n. 3 has been revised to state that breast examinations should not be delayed or rescheduled because of COVID-19 vaccination, as evidence from the first pandemic waves highlights how delayed or missed screening tests have a negative effect on breast cancer morbidity and mortality, and that there is a near-zero risk of subsequent malignant findings in asymptomatic patients who have unilateral lymphadenopathy and no suspicious breast findings. Second, recommendation n. 7 has been revised to simplify follow-up strategies: in patients without breast cancer history and no imaging findings suspicious for cancer, symptomatic and asymptomatic imaging-detected unilateral lymphadenopathy on the same side of recent COVID-19 vaccination (within 12 weeks) should be classified as a benign finding (BI-RADS 2) and no further work-up should be pursued. All other recommendations issued by EUSOBI in 2021 remain valid., (© 2023. The Author(s).)

Background There is considerable interest in the potential use of artificial intelligence (AI) systems in mammographic screening. However, it is essential to critically evaluate the performance of AI before it can become a modality used for independent mammographic interpretation. Purpose To evaluate the reported standalone performances of AI for interpretation of digital mammography and digital breast tomosynthesis (DBT). Materials and Methods A systematic search was conducted in PubMed, Google Scholar, Embase (Ovid), and Web of Science databases for studies published from January 2017 to June 2022. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) values were reviewed. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and Comparative (QUADAS-2 and QUADAS-C, respectively). A random effects meta-analysis and meta-regression analysis were performed for overall studies and for different study types (reader studies vs historic cohort studies) and imaging techniques (digital mammography vs DBT). Results In total, 16 studies that include 1 108 328 examinations in 497 091 women were analyzed (six reader studies, seven historic cohort studies on digital mammography, and four studies on DBT). Pooled AUCs were significantly higher for standalone AI than radiologists in the six reader studies on digital mammography (0.87 vs 0.81, P = .002), but not for historic cohort studies (0.89 vs 0.96, P = .152). Four studies on DBT showed significantly higher AUCs in AI compared with radiologists (0.90 vs 0.79, P < .001). Higher sensitivity and lower specificity were seen for standalone AI compared with radiologists. Conclusion Standalone AI for screening digital mammography performed as well as or better than radiologists. Compared with digital mammography, there is an insufficient number of studies to assess the performance of AI systems in the interpretation of DBT screening examinations. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Scaranelo in this issue.

Background: Solitary pulmonary nodules (SPNs) measuring 8 to 30 mm in diameter require further workup to determine the likelihood of malignancy., Research Question: What is the diagnostic performance of a lung cancer prediction convolutional neural network (LCP-CNN) in SPNs using unenhanced and contrast-enhanced CT imaging compared with the current clinical workup?, Study Design and Methods: This was a post hoc analysis of the Single Pulmonary Nodule Investigation: Accuracy and Cost-Effectiveness of Dynamic Contrast Enhanced Computed Tomography in the Characterisation of Solitary Pulmonary Nodules trial, a prospective multicenter study comparing the diagnostic accuracy of dynamic contrast-enhanced (DCE) CT imaging with PET imaging in SPNs. The LCP-CNN was designed and validated in an external cohort. LCP-CNN-generated risk scores were created from the noncontrast and contrast-enhanced CT scan images from the DCE CT imaging. The gold standard was histologic analysis or 2 years of follow-up. The area under the receiver operating characteristic curves (AUC) were calculated using LCP-CNN score, maximum standardized uptake value, and DCE CT scan maximum enhancement and were compared using the DeLong test., Results: Two hundred seventy participants (mean ± SD age, 68.3 ± 8.8 years; 49% women) underwent PET with CT scan imaging and DCE CT imaging with CT scan data available centrally for LCP-CNN analysis. The accuracy of the LCP-CNN on the noncontrast images (AUC, 0.83; 95% CI, 0.79-0.88) was superior to that of DCE CT imaging (AUC, 0.76; 95% CI, 0.69-0.82; P = .03) and equal to that of PET with CT scan imaging (AUC, 0.86; 95% CI, 0.81-0.90; P = .35). The presence of contrast resulted in a small reduction in diagnostic accuracy, with the AUC falling from 0.83 (95% CI, 0.79-0.88) on the noncontrast images to 0.80 to 0.83 after contrast (P < .05 for 240 s after contrast only)., Interpretation: An LCP-CNN algorithm provides an AUC equivalent to PET with CT scan imaging in the diagnosis of solitary pulmonary nodules., Trial Registration: ClinicalTrials.gov Identifier; No.: NCT02013063., (Copyright © 2022 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved.)

Background: Assessing inflammatory disease activity in large vessel vasculitis (LVV) can be challenging by conventional measures., Objectives: We aimed to investigate somatostatin receptor 2 (SST 2 ) as a novel inflammation-specific molecular imaging target in LVV., Methods: In a prospective, observational cohort study, in vivo arterial SST 2 expression was assessed by positron emission tomography/magnetic resonance imaging (PET/MRI) using 68 Ga-DOTATATE and 18 F-FET-βAG-TOCA. Ex vivo mapping of the imaging target was performed using immunofluorescence microscopy; imaging mass cytometry; and bulk, single-cell, and single-nucleus RNA sequencing., Results: Sixty-one participants (LVV: n = 27; recent atherosclerotic myocardial infarction of ≤2 weeks: n = 25; control subjects with an oncologic indication for imaging: n = 9) were included. Index vessel SST 2 maximum tissue-to-blood ratio was 61.8% (P < 0.0001) higher in active/grumbling LVV than inactive LVV and 34.6% (P = 0.0002) higher than myocardial infarction, with good diagnostic accuracy (area under the curve: ≥0.86; P < 0.001 for both). Arterial SST 2 signal was not elevated in any of the control subjects. SST 2 PET/MRI was generally consistent with 18 F-fluorodeoxyglucose PET/computed tomography imaging in LVV patients with contemporaneous clinical scans but with very low background signal in the brain and heart, allowing for unimpeded assessment of nearby coronary, myocardial, and intracranial artery involvement. Clinically effective treatment for LVV was associated with a 0.49 ± 0.24 (standard error of the mean [SEM]) (P = 0.04; 22.3%) reduction in the SST 2 maximum tissue-to-blood ratio after 9.3 ± 3.2 months. SST 2 expression was localized to macrophages, pericytes, and perivascular adipocytes in vasculitis specimens, with specific receptor binding confirmed by autoradiography. SSTR2-expressing macrophages coexpressed proinflammatory markers., Conclusions: SST 2 PET/MRI holds major promise for diagnosis and therapeutic monitoring in LVV. (PET Imaging of Giant Cell and Takayasu Arteritis [PITA], NCT04071691; Residual Inflammation and Plaque Progression Long-Term Evaluation [RIPPLE], NCT04073810)., Competing Interests: Funding Support and Author Disclosures This work was funded by grants to Dr Tarkin from the Wellcome Trust (Clinical Research Career Development Fellowship 211100/Z/18/Z), the National Institute for Health Research (NIHR) Imperial Biomedical Research Centre (BRC); and the British Heart Foundation (BHF) (Clinical Research Training Fellowship for Dr Ćorović [FS/CRTF/20/24035]). This work was additionally supported by the Cambridge BHF Centre of Research Excellence (18/1/34212) and the Cancer Research UK Cambridge Centre (A25177). For the purpose of open access, the lead author has applied a CC BY public copyright license to any Author Accepted Manuscript. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health and Social Care. Dr Nus; authors Imaz and Lambert; Dr Frontini (FS/18/53/33863); Dr Davenport (TG/18/4/33770); and Drs Huang, Mallat, Dweck, Newby, and Bennett are supported by the BHF. Author Zulcinski is supported by the European Union’s Horizon 2020 Research and Innovation Programme (Marie Skłodowska-Curie grant agreement no. 813545). Drs Jayne, Rassl, and Graves are supported by the NIHR Cambridge BRC. Dr Fayad is supported by the National Institutes of Health/National Heart, Lung, and Blood Institute (R01HL135878). Dr Reynolds is supported by the Wellcome Trust. Dr Morgan is supported by the Medical Research Council (MRC) (MR/N011775/1), the NIHR Leeds BRC, the NIHR Leeds Medtech, and In Vitro Diagnostics Co-operative as well as an NIHR Senior Investigator award. Dr Aboagye acknowledges support from Imperial Experimental Cancer Research Centre and MRC (MR/J007986/1, MR/N020782/1); and is an inventor on the patent that developed the (18)F-FET-βAG-TOCA radiotracer. Dr Peters is supported by a UK Research and Innovation Fellowship at Health Data Research UK (MR/S004068/2). Dr Rudd is partly supported by the NIHR Cambridge BRC, the BHF, the Higher Education Funding Council for England, the Engineering and Physical Sciences Research Council, and the Wellcome Trust. Drs Gopalan, Maughan, Pericleous, Barwick, Aboagye, Peters, and Mason acknowledge support from the NIHR Imperial BRC. The authors have reported that they have no relationships relevant to the contents of this paper to disclose., (Copyright © 2023 The Authors. Published by Elsevier Inc. All rights reserved.)

Background: Magnetic resonance imaging (MRI) can be used to diagnose breast cancer. Diffusion weighted imaging (DWI) and the apparent diffusion coefficient (ADC) can reflect tumor microstructure in a non-invasive manner. The correct prediction of response of neoadjuvant chemotherapy (NAC) is crucial for clinical routine. Our aim was to compare ADC values between patients with pathological complete response (pCR) and non-responders based upon a multi-center design to improve the correct patient selection, which patient would more benefit from NAC and which patient would not., Methods: For this study, data from 4 centers (from Japan, Brazil, Spain and United Kingdom) were retrospectively acquired. The time period was overall 2003-2019. The patient sample comprises 250 patients (all female; median age, 50.5). In every case, pretreatment breast MRI with DWI was performed. pCR was assessed by experienced pathologists in every center using the surgical specimen in the clinical routine work up. pCR was defined as no residual invasive disease in either breast or axillary lymph nodes after NAC. ADC values between the group with pCR and those with no pCR were compared using the Mann-Whitney U test (two-group comparisons). Univariable and multivariabe logistic regression analysis was performed to predict pCR status., Results: Overall, 83 patients (33.2%) achieved pCR. The ADC values of the patient group with pCR were lower compared with patients without pCR (0.98 ± 0.23 × 10 - 3  mm 2 /s versus 1.07 ± 0.24 × 10 - 3  mm 2 /s, p = 0.02). The ADC value achieved an odds ratio of 4.65 (95% CI 1.40-15.49) in univariable analysis and of 3.0 (95% CI 0.85-10.63) in multivariable analysis (overall sample) to be associated with pCR status. The odds ratios differed in the subgroup analyses in accordance with the molecular subtype., Conclusions: The pretreatment ADC-value is associated with pathological complete response after NAC in breast cancer patients. This could aid in clinical routine to reduce treatment toxicity for patients, who would not benefit from NAC. However, this must be tested in further studies, as the overlap of the ADC values in both groups is too high for clinical prediction., (© 2022. The Author(s).)

The task of classifying mammograms is very challenging because the lesion is usually small in the high resolution image. The current state-of-the-art approaches for medical image classification rely on using the de-facto method for convolutional neural networks-fine-tuning. However, there are fundamental differences between natural images and medical images, which based on existing evidence from the literature, limits the overall performance gain when designed with algorithmic approaches. In this paper, we propose to go beyond fine-tuning by introducing a novel framework called MorphHR, in which we highlight a new transfer learning scheme. The idea behind the proposed framework is to integrate function-preserving transformations, for any continuous non-linear activation neurons, to internally regularise the network for improving mammograms classification. The proposed solution offers two major advantages over the existing techniques. Firstly and unlike fine-tuning, the proposed approach allows for modifying not only the last few layers but also several of the first ones on a deep ConvNet. By doing this, we can design the network front to be suitable for learning domain specific features. Secondly, the proposed scheme is scalable to hardware. Therefore, one can fit high resolution images on standard GPU memory. We show that by using high resolution images, one prevents losing relevant information. We demonstrate, through numerical and visual experiments, that the proposed approach yields to a significant improvement in the classification performance over state-of-the-art techniques, and is indeed on a par with radiology experts. Moreover and for generalisation purposes, we show the effectiveness of the proposed learning scheme on another large dataset, the ChestX-ray14, surpassing current state-of-the-art techniques., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 The Author(s). Published by Elsevier B.V. All rights reserved.)

Introduction: Dynamic contrast-enhanced CT (DCE-CT) and positron emission tomography/CT (PET/CT) have a high reported accuracy for the diagnosis of malignancy in solitary pulmonary nodules (SPNs). The aim of this study was to compare the accuracy and cost-effectiveness of these., Methods: In this prospective multicentre trial, 380 participants with an SPN (8-30 mm) and no recent history of malignancy underwent DCE-CT and PET/CT. All patients underwent either biopsy with histological diagnosis or completed CT follow-up. Primary outcome measures were sensitivity, specificity and overall diagnostic accuracy for PET/CT and DCE-CT. Costs and cost-effectiveness were estimated from a healthcare provider perspective using a decision-model., Results: 312 participants (47% female, 68.1±9.0 years) completed the study, with 61% rate of malignancy at 2 years. The sensitivity, specificity, positive predictive value and negative predictive values for DCE-CT were 95.3% (95% CI 91.3 to 97.5), 29.8% (95% CI 22.3 to 38.4), 68.2% (95% CI 62.4% to 73.5%) and 80.0% (95% CI 66.2 to 89.1), respectively, and for PET/CT were 79.1% (95% CI 72.7 to 84.2), 81.8% (95% CI 74.0 to 87.7), 87.3% (95% CI 81.5 to 91.5) and 71.2% (95% CI 63.2 to 78.1). The area under the receiver operator characteristic curve (AUROC) for DCE-CT and PET/CT was 0.62 (95% CI 0.58 to 0.67) and 0.80 (95% CI 0.76 to 0.85), respectively (p<0.001). Combined results significantly increased diagnostic accuracy over PET/CT alone (AUROC=0.90 (95% CI 0.86 to 0.93), p<0.001). DCE-CT was preferred when the willingness to pay per incremental cost per correctly treated malignancy was below £9000. Above £15 500 a combined approach was preferred., Conclusions: PET/CT has a superior diagnostic accuracy to DCE-CT for the diagnosis of SPNs. Combining both techniques improves the diagnostic accuracy over either test alone and could be cost-effective., Trial Registration Number: NCT02013063., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2022. No commercial re-use. See rights and permissions. Published by BMJ.)

Objectives: To perform a survey among all European Society of Breast Imaging (EUSOBI) radiologist members to gather representative data regarding the clinical use of breast DWI., Methods: An online questionnaire was developed by two board-certified radiologists, reviewed by the EUSOBI board and committees, and finally distributed among EUSOBI active and associated (not based in Europe) radiologist members. The questionnaire included 20 questions pertaining to technical preferences (acquisition time, magnet strength, breast coils, number of b values), clinical indications, imaging evaluation, and reporting. Data were analyzed using descriptive statistics, the Chi-square test of independence, and Fisher's exact test., Results: Of 1411 EUSOBI radiologist members, 275/1411 (19.5%) responded. Most (222/275, 81%) reported using DWI as part of their routine protocol. Common indications for DWI include lesion characterization (using an ADC threshold of 1.2-1.3 × 10 -3 mm 2 /s) and prediction of response to chemotherapy. Members most commonly acquire two separate b values (114/217, 53%), with b value = 800 s/mm 2 being the preferred value for appraisal among those acquiring more than two b values (71/171, 42%). Most did not use synthetic b values (169/217, 78%). While most mention hindered diffusion in the MRI report (161/213, 76%), only 142/217 (57%) report ADC values., Conclusion: The utilization of DWI in clinical practice among EUSOBI radiologists who responded to the survey is generally in line with international recommendations, with the main application being the differentiation of benign and malignant enhancing lesions, treatment response assessment, and prediction of response to chemotherapy. Report integration of qualitative and quantitative DWI data is not uniform., Key Points: • Clinical performance of breast DWI is in good agreement with the current recommendations of the EUSOBI International Breast DWI working group. • Breast DWI applications in clinical practice include the differentiation of benign and malignant enhancing, treatment response assessment, and prediction of response to chemotherapy. • Report integration of DWI results is not uniform., (© 2022. The Author(s), under exclusive licence to European Society of Radiology.)

Objectives: High-grade DCIS (HG DCIS) is associated with upgrade to invasive disease but few studies evaluate the role of MRI in this subset of DCIS. This study compared surgical outcomes of females with HG DCIS on biopsy who had pre-operative MRI with those that proceeded directly to surgery., Methods: This single-centre retrospective, observational study identified patients with pure HG DCIS on pre-operative biopsy from the pathology database. Surgical outcomes, clinicopathological and radiological features were obtained for all patients., Results: From August 2015 to February 2020, 217 patients had HG DCIS on biopsy. Pre-operative MRI was performed in 40 (MRI group) and not in 88 (No MRI group) patients. Initial mastectomy was performed in 25/40 (63%) women in the MRI group and 20/88 (23%) women in the no MRI group ( p < 0.0001). No difference was observed in re-operation rate between the two groups, 15% in MRI group vs 22% in No MRI group ( p = 0.4749). Mean tumour size on histology was larger in mastectomy cases in the MRI group (73.4 mm, range 6-140 mm), than the total MRI group, (58.3 mm, range 0-140 mm) or no MRI group (30.7 mm, range 0-130 mm) ( p < 0.0001)., Conclusions: Pre-operative MRI in HG DCIS is associated with higher mastectomy rates, possibly due to patient selection for MRI, as tumours on final histology were significantly larger. Fewer re-operations were observed in the MRI group although this was not significant., Advances in Knowledge: Breast MRI performed pre-operatively in HG DCIS is associated with higher mastectomy rates and fewer re-operation rates.

Combining optoacoustic (OA) imaging with ultrasound (US) enables visualisation of functional blood vasculature in breast lesions by OA to be overlaid with the morphological information of US. Here, we develop a simple OA feature set to differentiate benign and malignant breast lesions. 94 female patients with benign, indeterminate or suspicious lesions were recruited and underwent OA-US. An OA-US imaging feature set was developed using images from the first 38 patients, which contained 14 malignant and 8 benign solid lesions. Two independent radiologists blindly scored the OA-US images of a further 56 patients, which included 31 malignant and 13 benign solid lesions, with a sensitivity of 96.8% and specificity of 84.6%. Our findings indicate that OA-US can reveal vascular patterns of breast lesions that indicate malignancy using a simple feature set based on single wavelength OA data, which is therefore amenable to application in low resource settings for breast cancer management., Competing Interests: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests, Sarah Bohndiek reports a relationship with EPFL Center for Biomedical Imaging that includes: speaking and lecture fees. Sarah Bohndiek reports a relationship with PreXion Inc that includes: funding grants. Sarah Bohndiek reports a relationship with iThera Medical GmbH that includes: non-financial support. Stefan Morscher, Nina Dalhaus, Steven Ford and Neal Burton report a relationship with iThera Medical GmbH that includes: employment. Fiona Gilbert reports a relationship with Google and Kheiron that includes: consultancy. Fiona Gilbert reports a relationship with PreXion Inc, GE Healthcare, Bayer and Hologic that includes: funding grants. Fiona Gilbert reports a relationship with Lunit, Screenpoint, Volpara, iCAD, Therapixel and Vara that includes: research collaboration., (© 2022 The Authors.)

Background: Breast cancer remains a leading cause of death in women and novel imaging biomarkers are urgently required. Here, we demonstrate the diagnostic and treatment-monitoring potential of non-invasive sodium ( 23 Na) MRI in preclinical models of breast cancer., Methods: Female Rag2 -/- Il2rg -/- and Balb/c mice bearing orthotopic breast tumours (MDA-MB-231, EMT6 and 4T1) underwent MRI as part of a randomised, controlled, interventional study. Tumour biology was probed using ex vivo fluorescence microscopy and electrophysiology., Results: 23 Na MRI revealed elevated sodium concentration ([Na + ]) in tumours vs non-tumour regions. Complementary proton-based diffusion-weighted imaging (DWI) linked elevated tumour [Na + ] to increased cellularity. Combining 23 Na MRI and DWI measurements enabled superior classification accuracy of tumour vs non-tumour regions compared with either parameter alone. Ex vivo assessment of isolated tumour slices confirmed elevated intracellular [Na + ] ([Na + ] i ); extracellular [Na + ] ([Na + ] e ) remained unchanged. Treatment with specific inward Na + conductance inhibitors (cariporide, eslicarbazepine acetate) did not affect tumour [Na + ]. Nonetheless, effective treatment with docetaxel reduced tumour [Na + ], whereas DWI measures were unchanged., Conclusions: Orthotopic breast cancer models exhibit elevated tumour [Na + ] that is driven by aberrantly elevated [Na + ] i . Moreover, 23 Na MRI enhances the diagnostic capability of DWI and represents a novel, non-invasive biomarker of treatment response with superior sensitivity compared to DWI alone., (© 2022. The Author(s).)

Breast density is an independent risk factor for the development of breast cancer and also decreases the sensitivity of mammography for screening. Consequently, women with extremely dense breasts face an increased risk of late diagnosis of breast cancer. These women are, therefore, underserved with current mammographic screening programs. The results of recent studies reporting on contrast-enhanced breast MRI as a screening method in women with extremely dense breasts provide compelling evidence that this approach can enable an important reduction in breast cancer mortality for these women and is cost-effective. Because there is now a valid option to improve breast cancer screening, the European Society of Breast Imaging (EUSOBI) recommends that women should be informed about their breast density. EUSOBI thus calls on all providers of mammography screening to share density information with the women being screened. In light of the available evidence, in women aged 50 to 70 years with extremely dense breasts, the EUSOBI now recommends offering screening breast MRI every 2 to 4 years. The EUSOBI acknowledges that it may currently not be possible to offer breast MRI immediately and everywhere and underscores that quality assurance procedures need to be established, but urges radiological societies and policymakers to act on this now. Since the wishes and values of individual women differ, in screening the principles of shared decision-making should be embraced. In particular, women should be counselled on the benefits and risks of mammography and MRI-based screening, so that they are capable of making an informed choice about their preferred screening method. KEY POINTS: • The recommendations in Figure 1 summarize the key points of the manuscript., (© 2022. The Author(s).)

Background: The MyPeBS study is an ongoing randomised controlled trial testing whether a risk-stratified breast cancer screening strategy is non-inferior, or eventually superior, to standard age-based screening at reducing incidence of stage 2 or more cancers. This large European Commission-funded initiative aims to include 85,000 women aged 40 to 70 years, without prior breast cancer and not previously identified at high risk in six countries (Belgium, France, Italy, Israel, Spain, UK). A specific work package within MyPeBS examines psychological, socio-economic and ethical aspects of this new screening strategy. It compares women's reported data and outcomes in both trial arms on the following issues: general anxiety, cancer-related worry, understanding of breast cancer screening strategy and information-seeking behaviour, socio-demographic and economic characteristics, quality of life, risk perception, intention to change health-related behaviours, satisfaction with the trial., Methods: At inclusion, 3-months, 1-year and 4-years, each woman participating in MyPeBS is asked to fill online questionnaires. Descriptive statistics, bivariate analyses, subgroup comparisons and analysis of variations over time will be performed with appropriate tests to assess differences between arms. Multivariate regression models will allow modelling of different patient reported data and outcomes such as comprehension of the information provided, general anxiety or cancer worry, and information seeking behaviour. In addition, a qualitative study (48 semi-structured interviews conducted in France and in the UK with women randomised in the risk-stratified arm), will help further understand participants' acceptability and comprehension of the trial, and their experience of risk assessment., Discussion: Beyond the scientific and medical objectives of this clinical study, it is critical to acknowledge the consequences of such a paradigm shift for women. Indeed, introducing a risk-based screening relying on individual biological differences also implies addressing non-biological differences (e.g. social status or health literacy) from an ethical perspective, to ensure equal access to healthcare. The results of the present study will facilitate making recommendations on implementation at the end of the trial to accompany any potential change in screening strategy., Trial Registration: Study sponsor: UNICANCER. My personalised breast screening (MyPeBS)., Clinicaltrials: gov (2018) available at: https://clinicaltrials.gov/ct2/show/NCT03672331 Contact: Cécile VISSAC SABATIER, PhD, + 33 (0)1 73 79 77 58 ext + 330,142,114,293, contact@mypebs.eu., (© 2022. The Author(s).)

Objectives: Preoperative breast magnetic resonance imaging (MRI) can inform surgical planning but might cause overtreatment by increasing the mastectomy rate. The Multicenter International Prospective Analysis (MIPA) study investigated this controversial issue., Methods: This observational study enrolled women aged 18-80 years with biopsy-proven breast cancer, who underwent MRI in addition to conventional imaging (mammography and/or breast ultrasonography) or conventional imaging alone before surgery as routine practice at 27 centers. Exclusion criteria included planned neoadjuvant therapy, pregnancy, personal history of any cancer, and distant metastases., Results: Of 5896 analyzed patients, 2763 (46.9%) had conventional imaging only (noMRI group), and 3133 (53.1%) underwent MRI that was performed for diagnosis, screening, or unknown purposes in 692/3133 women (22.1%), with preoperative intent in 2441/3133 women (77.9%, MRI group). Patients in the MRI group were younger, had denser breasts, more cancers ≥ 20 mm, and a higher rate of invasive lobular histology than patients who underwent conventional imaging alone (p < 0.001 for all comparisons). Mastectomy was planned based on conventional imaging in 22.4% (MRI group) versus 14.4% (noMRI group) (p < 0.001). The additional planned mastectomy rate in the MRI group was 11.3%. The overall performed first- plus second-line mastectomy rate was 36.3% (MRI group) versus 18.0% (noMRI group) (p < 0.001). In women receiving conserving surgery, MRI group had a significantly lower reoperation rate (8.5% versus 11.7%, p < 0.001)., Conclusions: Clinicians requested breast MRI for women with a higher a priori probability of receiving mastectomy. MRI was associated with 11.3% more mastectomies, and with 3.2% fewer reoperations in the breast conservation subgroup., Key Points: • In 19% of patients of the MIPA study, breast MRI was performed for screening or diagnostic purposes. • The current patient selection to preoperative breast MRI implies an 11% increase in mastectomies, counterbalanced by a 3% reduction of the reoperation rate. • Data from the MIPA study can support discussion in tumor boards when preoperative MRI is under consideration and should be shared with patients to achieve informed decision-making., (© 2021. The Author(s).)

Background: Current pathways recommend positron emission tomography-computerised tomography for the characterisation of solitary pulmonary nodules. Dynamic contrast-enhanced computerised tomography may be a more cost-effective approach., Objectives: To determine the diagnostic performances of dynamic contrast-enhanced computerised tomography and positron emission tomography-computerised tomography in the NHS for solitary pulmonary nodules. Systematic reviews and a health economic evaluation contributed to the decision-analytic modelling to assess the likely costs and health outcomes resulting from incorporation of dynamic contrast-enhanced computerised tomography into management strategies., Design: Multicentre comparative accuracy trial., Setting: Secondary or tertiary outpatient settings at 16 hospitals in the UK., Participants: Participants with solitary pulmonary nodules of ≥ 8 mm and of ≤ 30 mm in size with no malignancy in the previous 2 years were included., Interventions: Baseline positron emission tomography-computerised tomography and dynamic contrast-enhanced computer tomography with 2 years' follow-up., Main Outcome Measures: Primary outcome measures were sensitivity, specificity and diagnostic accuracy for positron emission tomography-computerised tomography and dynamic contrast-enhanced computerised tomography. Incremental cost-effectiveness ratios compared management strategies that used dynamic contrast-enhanced computerised tomography with management strategies that did not use dynamic contrast-enhanced computerised tomography., Results: A total of 380 patients were recruited (median age 69 years). Of 312 patients with matched dynamic contrast-enhanced computer tomography and positron emission tomography-computerised tomography examinations, 191 (61%) were cancer patients. The sensitivity, specificity and diagnostic accuracy for positron emission tomography-computerised tomography and dynamic contrast-enhanced computer tomography were 72.8% (95% confidence interval 66.1% to 78.6%), 81.8% (95% confidence interval 74.0% to 87.7%), 76.3% (95% confidence interval 71.3% to 80.7%) and 95.3% (95% confidence interval 91.3% to 97.5%), 29.8% (95% confidence interval 22.3% to 38.4%) and 69.9% (95% confidence interval 64.6% to 74.7%), respectively. Exploratory modelling showed that maximum standardised uptake values had the best diagnostic accuracy, with an area under the curve of 0.87, which increased to 0.90 if combined with dynamic contrast-enhanced computerised tomography peak enhancement. The economic analysis showed that, over 24 months, dynamic contrast-enhanced computerised tomography was less costly (£3305, 95% confidence interval £2952 to £3746) than positron emission tomography-computerised tomography (£4013, 95% confidence interval £3673 to £4498) or a strategy combining the two tests (£4058, 95% confidence interval £3702 to £4547). Positron emission tomography-computerised tomography led to more patients with malignant nodules being correctly managed, 0.44 on average (95% confidence interval 0.39 to 0.49), compared with 0.40 (95% confidence interval 0.35 to 0.45); using both tests further increased this (0.47, 95% confidence interval 0.42 to 0.51)., Limitations: The high prevalence of malignancy in nodules observed in this trial, compared with that observed in nodules identified within screening programmes, limits the generalisation of the current results to nodules identified by screening., Conclusions: Findings from this research indicate that positron emission tomography-computerised tomography is more accurate than dynamic contrast-enhanced computerised tomography for the characterisation of solitary pulmonary nodules. A combination of maximum standardised uptake value and peak enhancement had the highest accuracy with a small increase in costs. Findings from this research also indicate that a combined positron emission tomography-dynamic contrast-enhanced computerised tomography approach with a slightly higher willingness to pay to avoid missing small cancers or to avoid a 'watch and wait' policy may be an approach to consider., Future Work: Integration of the dynamic contrast-enhanced component into the positron emission tomography-computerised tomography examination and the feasibility of dynamic contrast-enhanced computerised tomography at lung screening for the characterisation of solitary pulmonary nodules should be explored, together with a lower radiation dose protocol., Study Registration: This study is registered as PROSPERO CRD42018112215 and CRD42019124299, and the trial is registered as ISRCTN30784948 and ClinicalTrials.gov NCT02013063., Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment ; Vol. 26, No. 17. See the NIHR Journals Library website for further project information.

Background Advances in computer processing and improvements in data availability have led to the development of machine learning (ML) techniques for mammographic imaging. Purpose To evaluate the reported performance of stand-alone ML applications for screening mammography workflow. Materials and Methods Ovid Embase, Ovid Medline, Cochrane Central Register of Controlled Trials, Scopus, and Web of Science literature databases were searched for relevant studies published from January 2012 to September 2020. The study was registered with the PROSPERO International Prospective Register of Systematic Reviews (protocol no. CRD42019156016). Stand-alone technology was defined as a ML algorithm that can be used independently of a human reader. Studies were quality assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and the Prediction Model Risk of Bias Assessment Tool, and reporting was evaluated using the Checklist for Artificial Intelligence in Medical Imaging. A primary meta-analysis included the top-performing algorithm and corresponding reader performance from which pooled summary estimates for the area under the receiver operating characteristic curve (AUC) were calculated using a bivariate model. Results Fourteen articles were included, which detailed 15 studies for stand-alone detection ( n = 8) and triage ( n = 7). Triage studies reported that 17%-91% of normal mammograms identified could be read by adapted screening, while "missing" an estimated 0%-7% of cancers. In total, an estimated 185 252 cases from three countries with more than 39 readers were included in the primary meta-analysis. The pooled sensitivity, specificity, and AUC was 75.4% (95% CI: 65.6, 83.2; P = .11), 90.6% (95% CI: 82.9, 95.0; P = .40), and 0.89 (95% CI: 0.84, 0.98), respectively, for algorithms, and 73.0% (95% CI: 60.7, 82.6), 88.6% (95% CI: 72.4, 95.8), and 0.85 (95% CI: 0.78, 0.97), respectively, for readers. Conclusion Machine learning (ML) algorithms that demonstrate a stand-alone application in mammographic screening workflows achieve or even exceed human reader detection performance and improve efficiency. However, this evidence is from a small number of retrospective studies. Therefore, further rigorous independent external prospective testing of ML algorithms to assess performance at preassigned thresholds is required to support these claims. ©RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Whitman and Moseley in this issue.

Population differences in dental development between Black and White ethnic groups have been debated but not previously studied in the UK. Using inappropriate data for dental age estimation (DAE) could lead to erroneous results and injustice. Data were collected from dental panoramic radiographs of 5590 subjects aged 6-24 years in a teaching hospital archive. Demirjian stages were determined for left-sided teeth and third molars and data collected regarding hypodontia and third molar agenesis. Third molar development in self-assigned Black British, including other self-assigned Black ethnicity, was compared with that of self-assigned White British subjects. Data were compared for males and females in the two ethnic groups using T-tests for Demirjian Stages A-G of third molar development and Mann-Whitney tests for Stage H once a cut-off age at the maximum age for Stage G had been imposed. Third molar development occurred earlier in subjects of Black ancestry compared to those of White ancestry. While both ethnic groups showed large age ranges for every third molar stage, in female subjects these generally occurred at least 1.5 years earlier, and in males at least one year earlier. Hypodontia and third molar agenesis were more prevalent in White British, but the ethnic difference in third molar development persisted in subjects with complete dentitions. This is a large study that confirms ethnic differences in a London population, emphasises the difficulties of establishing the 18-year-old threshold using DAE, and confirms the risk of overestimating the age of individuals of Black ethnicity using White ethnic reference data., (© 2021 The Authors. Journal of Forensic Sciences published by Wiley Periodicals LLC on behalf of American Academy of Forensic Sciences.)

Hyperpolarized 13 C-MRI is an emerging tool for probing tissue metabolism by measuring 13 C-label exchange between intravenously injected hyperpolarized [1- 13 C]pyruvate and endogenous tissue lactate. Here, we demonstrate that hyperpolarized 13 C-MRI can be used to detect early response to neoadjuvant therapy in breast cancer. Seven patients underwent multiparametric 1 H-MRI and hyperpolarized 13 C-MRI before and 7-11 days after commencing treatment. An increase in the lactate-to-pyruvate ratio of approximately 20% identified three patients who, following 5-6 cycles of treatment, showed pathological complete response. This ratio correlated with gene expression of the pyruvate transporter MCT1 and lactate dehydrogenase A ( LDHA ), the enzyme catalyzing label exchange between pyruvate and lactate. Analysis of approximately 2,000 breast tumors showed that overexpression of LDHA and the hypoxia marker CAIX was associated with reduced relapse-free and overall survival. Hyperpolarized 13 C-MRI represents a promising method for monitoring very early treatment response in breast cancer and has demonstrated prognostic potential. SIGNIFICANCE: Hyperpolarized carbon-13 MRI allows response assessment in patients with breast cancer after 7-11 days of neoadjuvant chemotherapy and outperformed state-of-the-art and research quantitative proton MRI techniques., (©2021 The Authors; Published by the American Association for Cancer Research.)