Effectiveness and cost-effectiveness of double reading of mammograms in breast cancer screening: findings of a systematic review

Impact of Transfer Learning Using Local Data on Performance of a Deep Learning Model for Screening Mammography

Purpose To investigate the issues of generalizability and replication of deep learning models by assessing performance of a screening mammography deep learning system developed at New York University (NYU) on a local Australian dataset. Materials and Methods In this retrospective study, all individuals with biopsy or surgical pathology-proven lesions and age-matched controls were identified from a South Australian public mammography screening program (January 2010 to December 2016). The primary outcome was deep learning system performance-measured with area under the receiver operating characteristic curve (AUC)-in classifying invasive breast cancer or ductal carcinoma in situ (n = 425) versus no malignancy (n = 490) or benign lesions (n = 44). The NYU system, including models without (NYU1) and with (NYU2) heatmaps, was tested in its original form, after training from scratch (without transfer learning), and after retraining with transfer learning. Results The local test set comprised 959 individuals (mean age, 62.5 years ± 8.5 [SD]; all female). The original AUCs for the NYU1 and NYU2 models were 0.83 (95% CI: 0.82, 0.84) and 0.89 (95% CI: 0.88, 0.89), respectively. When NYU1 and NYU2 were applied in their original form to the local test set, the AUCs were 0.76 (95% CI: 0.73, 0.79) and 0.84 (95% CI: 0.82, 0.87), respectively. After local training without transfer learning, the AUCs were 0.66 (95% CI: 0.62, 0.69) and 0.86 (95% CI: 0.84, 0.88). After retraining with transfer learning, the AUCs were 0.82 (95% CI: 0.80, 0.85) and 0.86 (95% CI: 0.84, 0.88). Conclusion A deep learning system developed using a U.S. dataset showed reduced performance when applied "out of the box" to an Australian dataset. Local retraining with transfer learning using available model weights improved model performance. Keywords: Screening Mammography, Convolutional Neural Network (CNN), Deep Learning Algorithms, Breast Cancer Supplemental material is available for this article. © RSNA, 2024 See also commentary by Cadrin-Chênevert in this issue.

Effect of artificial intelligence-based computer-aided diagnosis on the screening outcomes of digital mammography: a matched cohort study

OBJECTIVE

To investigate whether artificial intelligence-based computer-aided diagnosis (AI-CAD) can improve radiologists' performance when used to support radiologists' interpretation of digital mammography (DM) in breast cancer screening.

METHODS

A retrospective database search identified 3158 asymptomatic Korean women who consecutively underwent screening DM between January and December 2019 without AI-CAD support, and screening DM between February and July 2020 with image interpretation aided by AI-CAD in a tertiary referral hospital using single reading. Propensity score matching was used to match the DM with AI-CAD group in a 1:1 ratio with the DM without AI-CAD group according to age, breast density, experience level of the interpreting radiologist, and screening round. Performance measures were compared with the McNemar test and generalized estimating equations.

RESULTS

A total of 1579 women who underwent DM with AI-CAD were matched with 1579 women who underwent DM without AI-CAD. Radiologists showed higher specificity (96% [1500 of 1563] vs 91.6% [1430 of 1561]; p < 0.001) and lower abnormal interpretation rates (AIR) (4.9% [77 of 1579] vs 9.2% [145 of 1579]; p < 0.001) with AI-CAD than without. There was no significant difference in the cancer detection rate (CDR) (AI-CAD vs no AI-CAD, 8.9 vs 8.9 per 1000 examinations; p = 0.999), sensitivity (87.5% vs 77.8%; p = 0.999), and positive predictive value for biopsy (PPV3) (35.0% vs 35.0%; p = 0.999) according to AI-CAD support.

CONCLUSIONS

AI-CAD increases the specificity for radiologists without decreasing sensitivity as a supportive tool in the single reading of DM for breast cancer screening.

CLINICAL RELEVANCE STATEMENT

This study shows that AI-CAD could improve the specificity of radiologists' DM interpretation in the single reading system without decreasing sensitivity, suggesting that it can benefit patients by reducing false positive and recall rates.

KEY POINTS

• In this retrospective-matched cohort study (DM without AI-CAD vs DM with AI-CAD), radiologists showed higher specificity and lower AIR when AI-CAD was used to support decision-making in DM screening. • CDR, sensitivity, and PPV for biopsy did not differ with and without AI-CAD support.

It's not about the errors, it's about the learning: How the Royal College of Radiologists has developed a Radiology Events and Learning process in the United Kingdom

The Royal College of Radiologists (RCR) is based in the United Kingdom but is a global organisation with members and fellows worldwide. In this invited article, the chair of the RCR Radiology Events and Learning (REAL) panel recounts his experience in looking at radiological errors. He starts with his personal work auditing his own mistakes as a junior consultant, describes what he learned in his departmental role in a large teaching hospital running a Radiology Events and Learning Meeting (REALM) and gives an overview of some of the work done over the last two decades by the RCR. This includes publishing national guidelines which set standards for running a REALM, setting up the REAL panel which produces a quarterly newsletter of cases from RCR members, and running an annual conference to share information with local radiology departments around the country. A review of the literature describing the drivers for this work and looking at the parallels with industry lies alongside the practical tips he found useful which he hopes would be helpful to anyone setting up their own departmental errors or discrepancy meeting.

Deep-Learning-Based Computer-Aided Systems for Breast Cancer Imaging: A Critical Review

This paper provides a critical review of the literature on deep learning applications in breast tumor diagnosis using ultrasound and mammography images. It also summarizes recent advances in computer-aided diagnosis/detection (CAD) systems, which make use of new deep learning methods to automatically recognize breast images and improve the accuracy of diagnoses made by radiologists. This review is based upon published literature in the past decade (January 2010–January 2020), where we obtained around 250 research articles, and after an eligibility process, 59 articles were presented in more detail. The main findings in the classification process revealed that new DL-CAD methods are useful and effective screening tools for breast cancer, thus reducing the need for manual feature extraction. The breast tumor research community can utilize this survey as a basis for their current and future studies.

Biennial screening mammography: How many women ask for more? Estimate of the interval mammogram rate in an organised population-based screening programme

OBJECTIVES

To estimate the interval mammogram rate, i.e. the undertaking of an additional mammography between scheduled screening rounds, and identify factors influencing this phenomenon.

METHODS

Data from our screening programme for the year 2014, excluding prevalent rounds, were analysed. Information about the number of women who underwent  interval mammograms was obtained reviewing the questionnaires and searching the department database. Data on age, breast density, family history of breast cancer, number of screening rounds, previous recalls, general practitioner, and city of residence (used as a proxy of local socio-economic differences) were evaluated using chi-square test.

RESULTS

Of 2780 screened women (incident rounds), 2566 had complete data (92%). The interval mammogram rate was 384/2566 (15%, 95% confidence interval 14-17%). Women classified with American College of Radiology c or d breast density categories showed a higher interval mammography probability than those with a and b density (p < 0.001); women in their second round showed a higher probability of interval mammogram compared to women in their fifth, sixth, or seventh round (p ≤ 0.004). No significant differences were found between women with and without an interval mammogram when considering previous recalls for a negative work-up (p = 0.241), positive breast cancer family history (p = 0.538), and city of residence (p = 0.177).

CONCLUSIONS

The interval mammogram rate was relatively low (15%). Higher breast density and first of years of adherence to the programme were associated with higher interval mammogram rate.

Methods Used in Computer-Aided Diagnosis for Breast Cancer Detection Using Mammograms: A Review

According to the American Cancer Society's forecasts for 2019, there will be about 268,600 new cases in the United States with invasive breast cancer in women, about 62,930 new noninvasive cases, and about 41,760 death cases from breast cancer. As a result, there is a high demand for breast imaging specialists as indicated in a recent report for the Institute of Medicine and National Research Council. One way to meet this demand is through developing Computer-Aided Diagnosis (CAD) systems for breast cancer detection and diagnosis using mammograms. This study aims to review recent advancements and developments in CAD systems for breast cancer detection and diagnosis using mammograms and to give an overview of the methods used in its steps starting from preprocessing and enhancement step and ending in classification step. The current level of performance for the CAD systems is encouraging but not enough to make CAD systems standalone detection and diagnose clinical systems. Unless the performance of CAD systems enhanced dramatically from its current level by enhancing the existing methods, exploiting new promising methods in pattern recognition like data augmentation in deep learning and exploiting the advances in computational power of computers, CAD systems will continue to be a second opinion clinical procedure.

Systematic reviews as a "lens of evidence": Determinants of cost-effectiveness of breast cancer screening

Systematic reviews with economic components are important decision tools for stakeholders seeking to evaluate technologies, such as breast cancer screening (BCS) programs. This overview of systematic reviews explores the determinants of the cost-effectiveness of BCS and assesses the quality of secondary evidence. The search identified 30 systematic reviews that reported on the determinants of the cost-effectiveness of BCS, including the costs of breast cancer and BCS. While the quality of the reviews varied widely, only four out of 30 papers were considered to be of a high quality. We did not identify publication bias in the original evidence on the cost-effectiveness of mammography screening; however, we highlight a need for improved clarity in both reporting and data verification. The reviews consisted mainly of studies from high-income countries. Breast cancer costs varied widely among the studies. Factors leading to higher costs included: time (diagnosis and last months before death), later stage or metastases, recurrence of the disease, age below 64 years and type of follow-up (more intensive or more specialized). Overall, screening with mammography was considered cost-effective in the age range 50-69 years in Western European and Northern American countries but not for older or younger women. Its cost-effectiveness was questionable for low-income settings and Asia. Mammography screening was more cost-effective with biennial screening compared to annual screening and single reading using computer-aided detection vs double reading. No information on the cost-effectiveness of ultrasonography was found, and there is much uncertainty on the cost-effectiveness of CBE because of methodological limitations.

Systematic reviews as a 'lens of evidence': Determinants of benefits and harms of breast cancer screening

This systematic review, stimulated by inconsistency in secondary evidence, reports the benefits and harms of breast cancer (BC) screening and their determinants according to systematic reviews. A systematic search, which identified 9,976 abstracts, led to the inclusion of 58 reviews. BC mortality reduction with screening mammography was 15-25% in trials and 28-56% in observational studies in all age groups, and the risk of stage III+ cancers was reduced for women older than 49 years. Overdiagnosis due to mammography was 1-60% in trials and 1-12% in studies with a low risk of bias, and cumulative false-positive rates were lower with biennial than annual screening (3-17% vs 0.01-41%). There is no consistency in the reviews' conclusions about the magnitude of BC mortality reduction among women younger than 50 years or older than 69 years, or determinants of benefits and harms of mammography, including the type of mammography (digital vs screen-film), the number of views and the screening interval. Similarly, there was no solid evidence on determinants of benefits and harms or BC mortality reduction with screening by ultrasonography or clinical breast examination (sensitivity ranges, 54-84% and 47-69%, respectively), and strong evidence of unfavourable benefit-to-harm ratio with breast self-examination. The reviews' conclusions were not dependent on the quality of the reviews or publication date. Systematic reviews on mammography screening, mainly from high-income countries, systematically disagree on the interpretation of the benefit-to-harm ratio. Future reviews are unlikely to clarify the discrepancies unless new original studies are published.

[Measurements errors in screening mammogram interpretation by radiologists]

The timely detection of breast cancer is achieved through mammography; however, the quality of the procedure should be addressed for proper performance and interpretation. Despite recent improvements in quality assurance in mammography, interpretation still depends on each reader; therefore, errors can be made when interpreting screening mammograms, leading to unnecessary biopsies and/or overdiagnosis, with sustained physical, economic and psychological consequences. Since interpretation is related to the perceptive and cognitive ability of the radiologist, it is necessary to have extensive knowledge about the possible errors that may occur during interpretation, as well as of the way how they can be reduced, prevented and/or corrected to provide the patient with the highest possible level of safety.

A blueprint for cancer screening and early detection: Advancing screening's contribution to cancer control

From the mid-20th century, accumulating evidence has supported the introduction of screening for cancers of the cervix, breast, colon and rectum, prostate (via shared decisions), and lung. The opportunity to detect and treat precursor lesions and invasive disease at a more favorable stage has contributed substantially to reduced incidence, morbidity, and mortality. However, as new discoveries portend advancements in technology and risk-based screening, we fail to fulfill the greatest potential of the existing technology, in terms of both full access among the target population and the delivery of state-of-the art care at each crucial step in the cascade of events that characterize successful cancer screening. There also is insufficient commitment to invest in the development of new technologies, incentivize the development of new ideas, and rapidly evaluate promising new technology. In this report, the authors summarize the status of cancer screening and propose a blueprint for the nation to further advance the contribution of screening to cancer control.

A review of computer aided detection in mammography

Breast screening with mammography is widely recognized as the most effective method of detecting early breast cancer and has consistently demonstrated a 20-40% decrease in mortality among screened women. Despite this, the sensitivity of mammography ranges between 70 and 90%. Computer aided detection (CAD) is an artificial intelligence (AI) technique that utilizes pattern recognition to highlight suspicious features on imaging and marks them for the radiologist to review and interpret. It aims to decrease oversights made by interpreting radiologists. Here we review the efficacy of CAD and potential future directions.

Can eye-tracking metrics be used to better pair radiologists in a mammogram reading task?

PURPOSE

To propose a framework for optimal pairing of radiologists when reading mammograms based on their search patterns.

MATERIALS AND METHODS

Four experienced and four less-experienced radiologists were asked to assess 120 cases (59 with cancers) while their eye positions were tracked. Fourteen eye-tracking metrics were extracted to quantify the differences among radiologists' visual search pattern. For each radiologist and metric, less-experienced radiologists and expert readers were ranked based on the level of similarities in gaze patterns (from the most different to the most similar). Less-experienced readers and experts were also ranked based on the values of area under the receiver operating characteristic curve (AUC) after pairing (the best possible way of ranking). Using the Kendall's tau distance, rankings based on different metrics were compared with the best possible ranking. Using paired Wilcoxon signed-rank test, the AUC values when pairing in the best way were compared with pairing based on different metrics. Finally, we investigated the robustness of pairing strategies against the small sample size.

RESULTS

For ranking the experienced radiologists, results from eight metrics were as good as the best possible ranking. For the less-experienced ones, only one metric resulted in a ranking comparable to the best possible way of ranking. The AUC values of pairings based on these metrics did not differ significantly from the best pairing scenario. Compared to the pairings based on the cognitive metrics, the ranking based on AUC values varied more greatly with the sample size, suggesting that it is less robust against the small sample size compared to the cognitive metrics.

CONCLUSION

Different pairings may have different effects on performance; some are detrimental while some improve the performance of the pair. Using the suggested cognitive metrics, we can optimize the pairings even with a small dataset.

Combining independent decisions increases diagnostic accuracy of reading lumbosacral radiographs and magnetic resonance imaging

Diagnosing the causes of low back pain is a challenging task, prone to errors. A novel approach to increase diagnostic accuracy in medical decision making is collective intelligence, which refers to the ability of groups to outperform individual decision makers in solving problems. We investigated whether combining the independent ratings of chiropractors, chiropractic radiologists and medical radiologists can improve diagnostic accuracy when interpreting diagnostic images of the lumbosacral spine. Evaluations were obtained from two previously published studies: study 1 consisted of 13 raters independently rating 300 lumbosacral radiographs; study 2 consisted of 14 raters independently rating 100 lumbosacral magnetic resonance images. In both studies, raters evaluated the presence of "abnormalities", which are indicators of a serious health risk and warrant immediate further examination. We combined independent decisions of raters using a majority rule which takes as final diagnosis the decision of the majority of the group. We compared the performance of the majority rule to the performance of single raters. Our results show that with increasing group size (i.e., increasing the number of independent decisions) both sensitivity and specificity increased in both data-sets, with groups consistently outperforming single raters. These results were found for radiographs and MR image reading alike. Our findings suggest that combining independent ratings can improve the accuracy of lumbosacral diagnostic image reading.

Strategies for Decreasing Screening Mammography Recall Rates While Maintaining Performance Metrics

RATIONALE AND OBJECTIVE

This study aimed to determine the impact of interventions designed to reduce screening mammography recall rates on screening performance metrics.

MATERIALS AND METHODS

We assessed baseline performance for full-field digital mammography (FFDM) and digital breast tomosynthesis mammography (DBT) for a 3-year period before intervention. The first intervention sought to increase awareness of recalls from screening mammography. Breast imagers discussed their perceptions regarding screening recalls and were required to review their own recalled cases, including outcomes of diagnostic evaluation and biopsy. The second intervention implemented consensus double reading of all recalls, requiring two radiologists to agree if recall was necessary. Recall rates, cancer detection rates, and positive predictive value 1 (PPV1) were compared before and after each intervention.

RESULTS

The baseline recall rate, cancer detection rate, and PPV1 were 11.1%, 3.8/1000, and 3.4%, respectively, for FFDM, and 7.6%, 4.8/1000, and 6.0%, respectively, for DBT. Recall rates decreased significantly to 9.2% for FFDM and to 6.6% for DBT after the first intervention promoting awareness, as well as to 9.9% for FFDM after the second intervention implementing group consensus. PPV1 increased significantly to 5.7% for FFDM and to 9.0% for DBT after the second intervention. Cancer detection rate did not significantly change with the implementation of these interventions. An average of 2.3 minutes was spent consulting for each recall.

CONCLUSION

Reduction in recall rates is desirable, provided performance metrics remain favorable. Our interventions improved performance and could be implemented in other breast imaging settings.

Effectiveness and cost-effectiveness of double reading in digital mammography screening: A systematic review and meta-analysis

PURPOSE

Double reading is the strategy of choice for mammogram interpretation in screening programmes. It remains, however, unknown whether double reading is still the strategy of choice in the context of digital mammography. Our aim was to determine the effectiveness and cost-effectiveness of double reading versus single reading of digital mammograms in screening programmes.

METHODS

We performed a systematic review by searching the PubMed, Embase, and Cochrane Library databases up to April 2017. We used the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies) tool and CHEERS (Consolidated Health Economic Evaluation Reporting Standards) checklist to assess the methodological quality of the diagnostic studies and economic evaluations, respectively. A proportion's meta-analysis approach, 95% Confidence Intervals (95% CI) and test of heterogeneity (P values) were used for pooled results. Costs are expressed US$ PPP (United States Dollar purchasing power parities). The PROSPERO ID of this Systematic Review's protocol is CRD42014013804.

RESULTS

Of 1473 potentially relevant hits, four high-quality studies were included. The pooled cancer detection rate of double reading was 6.01 per 1000 screens (CI: 4.47‰-7.77‰), and it was 5.65 per 1000 screens (CI: 3.95‰-7.65‰) for single reading (P=0.76). The pooled proportion of false-positives of double reading was 47.03 per 1000 screens (CI: 39.13‰-55.62‰) and it was 40.60 per 1000 screens (CI: 38.58‰-42.67‰) for single reading (P=0.12). One study reported, for double reading, an ICER (Incremental Cost-Effectiveness Ratio) of 16,684 Euros (24,717 US$ PPP; 2015 value) per detected cancer. Single reading+CAD (computer-aided-detection) was cost-effective in Japan.

CONCLUSION

The evidence of benefit for double reading compared to single reading for digital mammography interpretation is scarce. Double reading seems to increase operational costs, have a not significantly higher false-positive rate, and a similar cancer detection rate.

Foundation and methodologies in computer-aided diagnosis systems for breast cancer detection

Breast cancer is the most prevalent cancer that affects women all over the world. Early detection and treatment of breast cancer could decline the mortality rate. Some issues such as technical reasons, which related to imaging quality and human error, increase misdiagnosis of breast cancer by radiologists. Computer-aided detection systems (CADs) are developed to overcome these restrictions and have been studied in many imaging modalities for breast cancer detection in recent years. The CAD systems improve radiologists' performance in finding and discriminating between the normal and abnormal tissues. These procedures are performed only as a double reader but the absolute decisions are still made by the radiologist. In this study, the recent CAD systems for breast cancer detection on different modalities such as mammography, ultrasound, MRI, and biopsy histopathological images are introduced. The foundation of CAD systems generally consist of four stages: Pre-processing, Segmentation, Feature extraction, and Classification. The approaches which applied to design different stages of CAD system are summarised. Advantages and disadvantages of different segmentation, feature extraction and classification techniques are listed. In addition, the impact of imbalanced datasets in classification outcomes and appropriate methods to solve these issues are discussed. As well as, performance evaluation metrics for various stages of breast cancer detection CAD systems are reviewed.

Cost-Effectiveness of Double Reading versus Single Reading of Mammograms in a Breast Cancer Screening Programme

OBJECTIVES

The usual practice in breast cancer screening programmes for mammogram interpretation is to perform double reading. However, little is known about its cost-effectiveness in the context of digital mammography. Our purpose was to evaluate the cost-effectiveness of double reading versus single reading of digital mammograms in a population-based breast cancer screening programme.

METHODS

Data from 28,636 screened women was used to establish a decision-tree model and to compare three strategies: 1) double reading; 2) double reading for women in their first participation and single reading for women in their subsequent participations; and 3) single reading. We calculated the incremental cost-effectiveness ratio (ICER), which was defined as the expected cost per one additionally detected cancer. We performed a deterministic sensitivity analysis to test the robustness of the ICER.

RESULTS

The detection rate of double reading (5.17‰) was similar to that of single reading (4.78‰; P = .768). The mean cost of each detected cancer was €8,912 for double reading and €8,287 for single reading. The ICER of double reading versus single reading was €16,684. The sensitivity analysis showed variations in the ICER according to the sensitivity of reading strategies. The strategy that combines double reading in first participation with single reading in subsequent participations was ruled out due to extended dominance.

CONCLUSIONS

From our results, double reading appears not to be a cost-effective strategy in the context of digital mammography. Double reading would eventually be challenged in screening programmes, as single reading might entail important net savings without significantly changing the cancer detection rate. These results are not conclusive and should be confirmed in prospective studies that investigate long-term outcomes like quality adjusted life years (QALYs).

Evaluation of 12 strategies for obtaining second opinions to improve interpretation of breast histopathology: simulation study

OBJECTIVE

 To evaluate the potential effect of second opinions on improving the accuracy of diagnostic interpretation of breast histopathology.

DESIGN

 Simulation study.

SETTING

 12 different strategies for acquiring independent second opinions.

PARTICIPANTS

 Interpretations of 240 breast biopsy specimens by 115 pathologists, one slide for each case, compared with reference diagnoses derived by expert consensus.

MAIN OUTCOME MEASURES

 Misclassification rates for individual pathologists and for 12 simulated strategies for second opinions. Simulations compared accuracy of diagnoses from single pathologists with that of diagnoses based on pairing interpretations from first and second independent pathologists, where resolution of disagreements was by an independent third pathologist. 12 strategies were evaluated in which acquisition of second opinions depended on initial diagnoses, assessment of case difficulty or borderline characteristics, pathologists' clinical volumes, or whether a second opinion was required by policy or desired by the pathologists. The 240 cases included benign without atypia (10% non-proliferative, 20% proliferative without atypia), atypia (30%), ductal carcinoma in situ (DCIS, 30%), and invasive cancer (10%). Overall misclassification rates and agreement statistics depended on the composition of the test set, which included a higher prevalence of difficult cases than in typical practice.

RESULTS

 Misclassification rates significantly decreased (P<0.001) with all second opinion strategies except for the strategy limiting second opinions only to cases of invasive cancer. The overall misclassification rate decreased from 24.7% to 18.1% when all cases received second opinions (P<0.001). Obtaining both first and second opinions from pathologists with a high volume (≥10 breast biopsy specimens weekly) resulted in the lowest misclassification rate in this test set (14.3%, 95% confidence interval 10.9% to 18.0%). Obtaining second opinions only for cases with initial interpretations of atypia, DCIS, or invasive cancer decreased the over-interpretation of benign cases without atypia from 12.9% to 6.0%. Atypia cases had the highest misclassification rate after single interpretation (52.2%), remaining at more than 34% in all second opinion scenarios.

CONCLUSION

 Second opinions can statistically significantly improve diagnostic agreement for pathologists' interpretations of breast biopsy specimens; however, variability in diagnosis will not be completely eliminated, especially for breast specimens with atypia.

Inter-observer variability in mammographic density assessment using Royal Australian and New Zealand College of Radiologists (RANZCR) synoptic scales

INTRODUCTION

The aim of this study was to evaluate observer variability in mammographic density assessment as measured using the Royal Australian and New Zealand College of Radiologists (RANZCR) synoptic scales. Visual assessment of mammographic density by radiologists is commonly used in clinical practice; however, these assessments have been shown to be more subjective than quantitative methods.

METHODS

The study included 40 cases of left cranial-caudal (CC) and mediolateral oblique (MLO) radiographs from 20 women. RANZCR-registered radiologists (n = 26) estimated mammographic breast density of the same images according to RANZCR synoptic scales 1-4. All images had their volumetric density classified using Volpara Density Grade (VDG) and Average Volumetric Breast Density percentage (AvBD%).

RESULTS

The results showed that the radiologists sampled had specialized for 17.18 years (sd 12.03) and read 2072 (sd 2441) mammograms per year on average. Inter-observer agreement using RANZCR synoptic scales had an average Kappa of 0.360; (95% CI = 0.308-0.412) and a range of 0.078-0.499. Radiologists estimated percentage density was lower by 0.37 than VDG, with their mean being 2.18 and the mean VDG was 2.55 (Z = -3.873; P < 0.001). VDG and RANZCR showed a positive strong correlation (ρ = 0.898; P < 0.001). AvBD% and RANZCR also showed a positive strong correlation (ρ = 0.904; P < 0.001).

CONCLUSION

The inter-observer agreement with RANZCR synoptic scales was fair. Wide inter-observer variability was observed. Continued research on appropriate assessment methods for mammographic density assessment is required to avoid unnecessary variations.

Double versus single reading of mammograms in a breast cancer screening programme: a cost-consequence analysis

OBJECTIVES

To assess the costs and health-related outcomes of double versus single reading of digital mammograms in a breast cancer screening programme.

METHODS

Based on data from 57,157 digital screening mammograms from women aged 50-69 years, we compared costs, false-positive results, positive predictive value and cancer detection rate using four reading strategies: double reading with and without consensus and arbitration, and single reading with first reader only and second reader only. Four highly trained radiologists read the mammograms.

RESULTS

Double reading with consensus and arbitration was 15 % (Euro 334,341) more expensive than single reading with first reader only. False-positive results were more frequent at double reading with consensus and arbitration than at single reading with first reader only (4.5 % and 4.2 %, respectively; p < 0.001). The positive predictive value (9.3 % and 9.1 %; p = 0.812) and cancer detection rate were similar for both reading strategies (4.6 and 4.2 per 1000 screens; p = 0.283).

CONCLUSIONS

Our results suggest that changing to single reading of mammograms could produce savings in breast cancer screening. Single reading could reduce the frequency of false-positive results without changing the cancer detection rate. These results are not conclusive and cannot be generalized to other contexts with less trained radiologists.

KEY POINTS

• Double reading of digital mammograms is more expensive than single reading. • Compared to single reading, double reading yields a higher proportion of false-positive results. • The cancer detection rate was similar for double and single readings. • Single reading may be a cost-effective strategy in breast cancer screening programmes.

Mammography screening in less developed countries

Less developed countries (LDCs) are struggling with an increasing burden of breast cancer. It is important to identify what interventions might be most effective and feasible in reducing overall breast cancer mortality in a resource constrained settings. Mammography screening (MS) utilized in developed countries cannot be equally applied to LDCs. We provide a summary of the status of existing and past MS program attempts in LDCs, and try to determine the prerequisites under which any developing country is ready to benefit from a MS program. We make the case for a “mixed” portfolio of tools to reduce breast cancer mortality with MS reserved only for those sub-populations that meet the criteria. We hope our review will provide a background for policy makers to apply rigorous criteria before attempting to implement costly MS program and before judiciously evaluating additional competed programs in their countries.

Collective intelligence meets medical decision-making: the collective outperforms the best radiologist

While collective intelligence (CI) is a powerful approach to increase decision accuracy, few attempts have been made to unlock its potential in medical decision-making. Here we investigated the performance of three well-known collective intelligence rules (“majority”, “quorum”, and “weighted quorum”) when applied to mammography screening. For any particular mammogram, these rules aggregate the independent assessments of multiple radiologists into a single decision (recall the patient for additional workup or not). We found that, compared to single radiologists, any of these CI-rules both increases true positives (i.e., recalls of patients with cancer) and decreases false positives (i.e., recalls of patients without cancer), thereby overcoming one of the fundamental limitations to decision accuracy that individual radiologists face. Importantly, we find that all CI-rules systematically outperform even the best-performing individual radiologist in the respective group. Our findings demonstrate that CI can be employed to improve mammography screening; similarly, CI may have the potential to improve medical decision-making in a much wider range of contexts, including many areas of diagnostic imaging and, more generally, diagnostic decisions that are based on the subjective interpretation of evidence.

Breast cancer diagnosis in digitized mammograms using curvelet moments

BACKGROUND

Feature extraction is a key issue in designing a computer aided diagnosis system. Recent researches on breast cancer diagnosis have reported the effectiveness of multiscale transforms (wavelets and curvelets) for mammogram analysis and have shown the superiority of curvelet transform. However, the curse of dimensionality problem arises when using the curvelet coefficients and therefore a reduction method is required to extract a reduced set of discriminative features.

METHODS

This paper deals with this problem and proposes a feature extraction method based on curvelet transform and moment theory for mammogram description. First, we performed discrete curvelet transform and we computed the four first-order moments from curvelet coefficients distribution. Hence, two feature sets can be obtained: moments from each band and moments from each level. In this work, both sets are studied. Then, the t-test ranking technique was applied to select the best features from each set. Finally, a k-nearest neighbor classifier was used to distinguish between normal and abnormal breast tissues and to classify tumors as malignant or benign. Experiments were performed on 252 mammograms from the Mammographic Image Analysis Society (mini-MIAS) database using the leave-one-out cross validation as well as on 11553 mammograms from the Digital Database for Screening Mammography (DDSM) database using 2×5-fold cross validation.

RESULTS

Experimental results prove the effectiveness and the superiority of curvelet moments for mammogram analysis. Indeed, results on the mini-MIAS database show that curvelet moments yield an accuracy of 91.27% (resp. 81.35 %) with 10 (resp. 8) features for abnormality (resp. malignancy) detection. In addition, empirical comparisons of the proposed method against state-of-the-art curvelet-based methods on the DDSM database show that the suggested method does not only lead to a more reduced feature set, but it also statistically outperforms all the compared methods in terms of accuracy.

CONCLUSIONS

In summary, curvelet moments are an efficient and effective way to extract a reduced set of discriminative features for breast cancer diagnosis.

Double reading rates and quality assurance practices in Norwegian hospital radiology departments: two parallel national surveys

BACKGROUND

Double reading as a quality assurance (QA) tool is employed extensively in Norwegian hospital radiology departments. The practice is resource consuming and regularly debated.

PURPOSE

To investigate the rates of double reading in Norwegian hospital radiology departments, to identify department characteristics associated with double reading rates, and to investigate associations between double reading and other quality improvement.

MATERIAL AND METHODS

We issued two parallel national surveys to management and to consultant radiologists, respectively. Management was defined as the chief medical officer and/or the head of the radiology department. The management survey covered staffing, perceived resource situation, double reading, guidelines, and quality improvement. The radiologist survey served to validate management responses concerning double reading. Management survey items concerning practices of quality improvement were organized into three indices reflecting different quality approaches, namely: appropriateness of investigations; personal performance feedback; and system performance feedback.

RESULTS

The response rates of the surveys were 100% (45/45) for management and 55% (266/483) for radiologists. Of all exams read by consultants, 33% were double read. The double reading rate was highest in university hospital departments (59%), intermediate in other teaching departments (30%), and lowest in non-teaching departments (11%) (P = 0.01). Among the quality indices, mean scores were highest on appropriateness index (68%), intermediate on the person index (56%), and lowest on system index (37%). There were no correlations between double reading rates and scores on any of the quality indices.

CONCLUSION

The rate of double reading in Norwegian hospital radiology is significantly correlated to department teaching status, but not to other practices of quality work.

Second opinion in breast pathology: policy, practice and perception

AIMS

To assess the laboratory policies, pathologists' clinical practice and perceptions about the value of second opinions for breast pathology cases among pathologists practising in the USA.

METHODS

Cross-sectional data were collected from 252 pathologists who interpret breast specimens in eight states using a web-based survey. Descriptive statistics were used to characterise findings.

RESULTS

Most participants had >10 years of experience interpreting breast specimens (64%), were not affiliated with academic centres (73%) and were not considered experts by their peers (79%). Laboratory policies mandating second opinions varied by diagnosis: invasive cancer 65%; ductal carcinoma in situ (DCIS) 56%; atypical ductal hyperplasia 36% and other benign cases 33%. 81% obtained second opinions in the absence of policies. Participants believed they improve diagnostic accuracy (96%) and protect from malpractice suits (83%), and were easy to obtain, did not take too much time and did not make them look less adequate. The most common (60%) approach to resolving differences between the first and second opinion is to ask for a third opinion, followed by reaching a consensus.

CONCLUSIONS

Laboratory-based second opinion policies vary for breast pathology but are most common for invasive cancer and DCIS cases. Pathologists have favourable attitudes towards second opinions, adhere to policies and obtain them even when policies are absent. Those without a formal policy may benefit from supportive clinical practices and systems that help obtain second opinions.

MDCT in the assessment of laryngeal trauma: value of 2D multiplanar and 3D reconstructions

OBJECTIVE

The purpose of this study was to analyze fracture patterns and related effects of laryngeal trauma and to assess the value of 2D multiplanar reformation (MPR) and 3D reconstruction.

MATERIALS AND METHODS

Among 4222 consecutively registered trauma patients who underwent emergency MDCT, 38 patients had presented with laryngeal trauma. Axial, 2D MPR, 3D volume-rendered, and virtual endoscopic images were analyzed retrospectively by two blinded observers according to predefined criteria. Laryngeal fractures, soft-tissue injuries, and airway compromise were evaluated and correlated with clinical, endoscopic, surgical, and follow-up findings.

RESULTS

Fifty-nine fractures (37 thyroid, 13 cricoid, nine arytenoid) were present in 38 patients. They were isolated in 21 (55%) patients. The other 17 (45%) patients had additional injuries to the neck, face, brain, chest, or abdomen. Laryngeal fractures were bilateral in 31 (82%) patients and were associated with hyoid bone fractures in nine (24%) patients. Arytenoid luxation was present in eight cartilages. Axial imaging missed 7 of 59 (12%) laryngeal fractures, six of eight (75%) arytenoid luxations, and four of nine (44%) hyoid bone fractures. Additional 2D MPR imaging missed 5 of 59 (8%) laryngeal fractures, five of eight (62.5%) arytenoid luxations, and two of nine (22%) hyoid bone fractures, whereas 3D volume-rendered images depicted them all. Virtual endoscopy and 3D volume rendering added diagnostic accuracy with respect to the length, width, shape, and spatial orientation of fractures in 22 of 38 (58%) patients; arytenoid luxation in six of eight (75%) luxations; and the evaluation of airway narrowing in 19 of 38 (50%) patients. Three-dimensional volume rendering was not of additional value in evaluation of the cricoid cartilage.

CONCLUSION

The use of 2D MPR and 3D volume rendering with or without virtual endoscopy improved assessment of thyroid and hyoid bone fractures, arytenoid luxations, and laryngotracheal narrowing, providing helpful data for optimal management.

Performance of double reading mammography in an Iranian population and its effect on patient outcome

Background

Considering the importance and responsibility of reporting mammography and the necessity to notice details with a high degree of precision, double reading mammography has been introduced and recommended.

Objectives

This study aimed to assess the performance of double reading of mammograms and its effect on patient outcomes.

Patients and Methods

Throughout this cross sectional study, 1284 digitized mammographic views of 642 breasts which belonged to 339 women (of which 303 were bilateral and 36 were unilateral mammographies) were enrolled. Two independent radiologists interpreted these mammograms and BI-RADS categories of both reports were compared. Discordant results were determined and assumed significant if they were in the positive (BI-RADS 0, 4, 5) versus negative (BI-RADS 1, 2, 3) groups and then significant discordant cases were followed up to determine benign versus malignant final diagnosis. The recall rate was calculated for each reader. Inter-observer agreement in breast density was determined by Kappa test.

Results

Readers had consensus on BI-RADS categories in 459 breasts (71%), but diverse categories were used for 183 breasts (29%), including 132 significant and 51 non-significant discrepancies. According to weighted Kappa test, agreement between two readers in positive or negative reports was 0.78 (95% CI=0.73-0.83) and in parenchymal density, it was 0.73 (95% CI=0.7-0.77). Most of the discrepancies were between category zero versus categories 1 and 2 (63.4%). The recall rate was 36% for the first and 44% for the second reader. Among 132 significant discordant results, one case had the final diagnosis of malignancy and the others had benign or negative diagnosis. There was 0.2% increase in cancer detection rate by double reading.

Conclusion

This study shows no significant improvement in the cancer detection rate by double reading; however, a lower recall rate could be a more helpful consequence.

Prospective trial comparing full-field digital mammography (FFDM) versus combined FFDM and tomosynthesis in a population-based screening programme using independent double reading with arbitration

OBJECTIVES

To compare double readings when interpreting full field digital mammography (2D) and tomosynthesis (3D) during mammographic screening.

METHODS

A prospective, Ethical Committee approved screening study is underway. During the first year 12,621 consenting women underwent both 2D and 3D imaging. Each examination was independently interpreted by four radiologists under four reading modes: Arm A-2D; Arm B-2D + CAD; Arm C-2D + 3D; Arm D-synthesised 2D + 3D. Examinations with a positive score by at least one reader were discussed at an arbitration meeting before a final management decision. Paired double reading of 2D (Arm A + B) and 2D + 3D (Arm C + D) were analysed. Performance measures were compared using generalised linear mixed models, accounting for inter-reader performance heterogeneity (P < 0.05).

RESULTS

Pre-arbitration false-positive scores were 10.3 % (1,286/12,501) and 8.5 % (1,057/12,501) for 2D and 2D + 3D, respectively (P < 0.001). Recall rates were 2.9 % (365/12,621) and 3.7 % (463/12,621), respectively (P = 0.005). Cancer detection was 7.1 (90/12,621) and 9.4 (119/12,621) per 1,000 examinations, respectively (30 % increase, P < 0.001); positive predictive values (detected cancer patients per 100 recalls) were 24.7 % and 25.5 %, respectively (P = 0.97). Using 2D + 3D, double-reading radiologists detected 27 additional invasive cancers (P < 0.001).

CONCLUSION

Double reading of 2D + 3D significantly improves the cancer detection rate in mammography screening.

KEY POINTS

• Tomosynthesis-based screening was successfully implemented in a large prospective screening trial. • Double reading of tomosynthesis-based examinations significantly reduced false-positive interpretations. • Double reading of tomosynthesis significantly increased the detection of invasive cancers.

From randomized trials to the clinic: is it time to implement individual lung-cancer screening in clinical practice? A multidisciplinary statement from French experts on behalf of the French intergroup (IFCT) and the groupe d'Oncologie de langue francaise (GOLF)

Background

Despite advances in cancer therapy, mortality is still high except in early-stage tumors, and screening remains a challenge. The randomized National Lung Screening Trial (NLST), comparing annual low-dose computed tomography (LDCT) and chest X-rays, revealed a 20% decrease in lung-cancer-specific mortality. These results raised numerous questions. The French intergroup for thoracic oncology and the French-speaking oncology group convened an expert group to provide a coherent outlook on screening modalities in France.

Methods

A literature review was carried out and transmitted to the expert group, which was divided into three workshops to tackle specific questions, with responses presented in a plenary session. A writing committee drafted this article.

Results

The multidisciplinary group favored individual screening in France, when carried out as outlined in this article and after informing subjects of the benefits and risks. The target population involves subjects aged 55–74 years, who are smokers or have a 30 pack-year smoking history. Subjects should be informed about the benefits of quitting. Screening should involve LDCT scanning with specific modalities. Criteria for CT positivity and management algorithms for positive examinations are given.

Conclusions

Individual screening requires rigorous assessment and precise research in order to potentially develop a lung-cancer screening policy.

Inter- and intraradiologist variability in the BI-RADS assessment and breast density categories for screening mammograms

OBJECTIVE

The aim of this study was to evaluate reader variability in screening mammograms according to the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) assessment and breast density categories.

METHODS

A stratified random sample of 100 mammograms was selected from a population-based breast cancer screening programme in Barcelona, Spain: 13 histopathologically confirmed breast cancers and 51 with true-negative and 36 with false-positive results. 21 expert radiologists from radiological units of breast cancer screening programmes in Catalonia, Spain, reviewed the mammography images twice within a 6-month interval. The readers described each mammography using BI-RADS assessment and breast density categories. Inter- and intraradiologist agreement was assessed using percentage of concordance and the kappa (κ) statistic.

RESULTS

Fair interobserver agreement was observed for the BI-RADS assessment [κ=0.37, 95% confidence interval (CI) 0.36-0.38]. When the categories were collapsed in terms of whether additional evaluation was required (Categories III, 0, IV, V) or not (I and II), moderate agreement was found (κ=0.53, 95% CI 0.52-0.54). Intra-observer agreement for BI-RADS assessment was moderate using all categories (κ=0.53, 95% CI 0.50-0.55) and substantial on recall (κ=0.66, 95% CI 0.63-0.70). Regarding breast density, inter- and intraradiologist agreement was substantial (κ=0.73, 95% CI 0.72-0.74 and κ=0.69, 95% CI 0.68-0.70, respectively).

CONCLUSION

We observed a substantial intra-observer agreement in the BI-RADS assessment but only moderate interobserver agreement. Both inter- and intra-observer agreement in mammographic interpretation of breast density was substantial. Advances in knowledge Educational efforts should be made to decrease radiologists' variability in BI-RADS assessment interpretation in population-based breast screening programmes.

Mammography screening using independent double reading with consensus: is there a potential benefit for computer-aided detection?

BACKGROUND

Double reading improves the cancer detection rate in mammography screening. Single reading with computer-aided detection (CAD) has been considered to be an alternative to double reading. Little is known about the potential benefit of CAD in breast cancer screening with double reading.

PURPOSE

To compare prospective independent double reading of screen-film (SFM) and full-field digital (FFDM) mammography in population-based screening with retrospective standalone CAD performance on the baseline mammograms of the screen-detected cancers and subsequent cancers diagnosed during the follow-up period.

MATERIAL AND METHODS

The study had ethics committee approval. A 5-point rating scale for probability of cancer was used for 23,923 (SFM = 16,983; FFDM = 6940) screening mammograms. Of 208 evaluable cancers, 104 were screen-detected and 104 were subsequent (44 interval and 60 next screening round) cancers. Baseline mammograms of subsequent cancers were retrospectively classified in consensus without information about cancer location, histology, or CAD prompting as normal, non-specific minimal signs, significant minimal signs, and false-negatives. The baseline mammograms of the screen-detected cancers and subsequent cancers were evaluated by CAD. Significant minimal signs and false-negatives were considered 'actionable' and potentially diagnosable if correctly prompted by CAD.

RESULTS

CAD correctly marked 94% (98/104) of the baseline mammograms of the screen-detected cancers (SFM = 95% [61/64]; FFDM = 93% [37/40]), including 96% (23/24) of those with discordant interpretations. Considering only those baseline examinations of subsequent cancers prospectively interpreted as normal and retrospectively categorized as 'actionable', CAD input at baseline screening had the potential to increase the cancer detection rate from 0.43% to 0.51% (P = 0.13); and to increase cancer detection by 16% ([104 + 17]/104) and decrease interval cancers by 20% (from 44 to 35).

CONCLUSION

CAD may have the potential to increase cancer detection by up to 16%, and to reduce the number of interval cancers by up to 20% in SFM and FFDM screening programs using independent double reading with consensus review. The influence of true- and false-positive CAD marks on decision-making can, however, only be evaluated in a prospective clinical study.

Interpretation errors in CT angiography of the head and neck and the benefit of double reading

BACKGROUND AND PURPOSE

CTA provides high-resolution imaging of the head and neck vasculature but also of the soft tissues and bones. This results in a large volume of information to be interpreted. This study examines interpretation errors with head and neck CTAs and assesses whether double reading reduces miss rates.

MATERIALS AND METHODS

Consecutive CTAs of the neck and intracranial circulation were retrospectively identified and reviewed for vascular and nonvascular findings by a consensus of 2 neuroradiologists. The results were compared with the official report. Significant discrepancies were considered those that would have influenced follow-up or management.

RESULTS

We reviewed 503 studies; 144 were originally reported by a staff neuroradiologist alone, 209 by staff and diagnostic radiology resident, and 150 by staff and neuroradiology fellow. Twenty-six significant discrepancies were discovered in 20 studies, corresponding to 4.0% of studies with at least 1 miss, and an overall miss rate per study of 5.2%. There was at least 1 miss in 6.3% of studies interpreted by a staff neuroradiologist alone, 3.3% by staff and resident, and 2.7% by staff and fellow. The miss rate differences were not statistically significant. The most common misses were small aneurysms (50% of misses).

CONCLUSIONS

CTA neck and head datasets are now large, and there is a potential for missed findings. Significant discrepancies can occur with a low but not insignificant rate. Arterial pathology accounted for most discrepancies. This study emphasizes the need for careful systematic scrutiny for both vascular and nonvascular pathology regardless of indication. Double reading reduces error rates.

Double reading of radiological examinations in Norway

BACKGROUND

Double reading of images is a part of the quality assurance activities at many radiological centers.

PURPOSE

To investigate the extent of and routines for double reading in Norway and the institutional heads' attitudes toward double reading.

MATERIAL AND METHODS

A questionnaire was addressed to the heads of all radiological institutions in Norway. The questionnaire concerned staffing, examinations performed, extent of double reading per imaging modality (except mammography screening), guidelines for double reading, checks of completed radiology reports, frequency of regular quality assurance meetings to discuss missed findings, and the heads' attitudes toward double reading.

RESULTS

The response rate was 73% (53/73). The percentage across imaging modalities of examinations being double read was 41% overall: 56% at university hospitals, 37% at local hospitals, and 18% at private centers. Double reading was most common for positron emission tomography (PET)/PET-computed tomography (CT) examinations (100%), and clinical mammography (91%). Almost all examinations read by residents were double read. Only 15% of institutions had written guidelines for double reading, 15% performed random double readings of completed examinations, and 55% organized regular meetings to discuss missed findings. Forty-six percent of the institutional heads wanted an increased use of double reading.

CONCLUSION

Double reading is common in Norway, especially in residency training, mammography, and PET/PET-CT. It is less common at private centers. Established routines for double reading are scarce. Many institutional heads want more double reading. The potential of double reading to assure quality in radiology should be better exploited.

Computer-aided detection mammography for breast cancer screening: systematic review and meta-analysis

CONTEXT

Mammography is generally accepted as the best available breast cancer screening method; however, some cancers detectable on mammography images are missed. Computer-aided detection (CAD) systems for mammography are intended to reduce false negatives by marking suspicious areas of the mammograms for reviewers to consider. Although the prospect of improving the sensitivity of screening mammograms has led to the diffusion of CAD for mammography, little is known about its diagnostic accuracy.

OBJECTIVE

To assess the diagnostic performance of CAD for screening mammography in terms of sensitivity and specificity and incremental recall, biopsy, and cancer diagnosis rates.

DATA SOURCES

Published literature identified by systematic literature searches of 17 databases, including MEDLINE, EMBASE, and the Cochrane Library, searched through 25 September 2008.

STUDY SELECTION

A reviewer and an information specialist selected full-length English-language articles that enrolled asymptomatic women for routine breast cancer screening and provided data needed for our analyses using criteria established a priori. We identified 75 potentially relevant publications, of which 7 (9%) were included.

DATA EXTRACTION

Data were extracted and internal validity was assessed by a single review author, and forms were approved by the co-authors.

RESULTS

Three studies (n = 347,324) reported sensitivity and specificity, or data to calculate them, and five studies (n = 51,162) reported data to calculate incremental rates of cancer diagnoses and recall and biopsy of women who did not have breast cancer. The pooled sensitivity was 86.0% (95% CI 84.2-87.6%) and specificity was 88.2% (95% CI 88.1-88.3%). Of the 100,000 women screened, CAD yielded an additional 50 (95% CI 30-80) correct breast cancer diagnoses, 1,190 (95% CI 1,090-1,290) recalls of healthy women, and 80 (95% CI 60-100) biopsies of healthy women. A total of 96% (95% CI 93.9-97.3%) of women recalled based upon CAD and 65.1% (95% CI 52.3-76.0%) of women biopsied based upon CAD were healthy. No studies reported patient-oriented clinical outcomes.

Randomized trial of screen-film versus full-field digital mammography with soft-copy reading in population-based screening program: follow-up and final results of Oslo II study

PURPOSE

To prospectively compare performance indicators at screen-film mammography (SFM) and full-field digital mammography (FFDM) in a population-based screening program.

MATERIALS AND METHODS

The regional ethics committee approved the study; informed consent was obtained from patients. Women aged 45-69 years were assigned to undergo SFM (n=16 985) or FFDM (n=6944). Two-view mammograms were interpreted by using independent double reading and a five-point rating scale for probability of cancer. Positive scores were discussed at consensus meetings before decision for recall. The group was followed up for 1.5 years (women aged 45-49 years) and 2.0 years (women aged 50-69 years) to include subsequent cancers with positive scores at baseline interpretation and to estimate interval cancer rate. Recall rates, cancer detection, positive predictive values (PPVs), sensitivity, specificity, tumor characteristics, and discordant interpretations of cancers were compared.

RESULTS

Recall rate was 4.2% at FFDM and 2.5% at SFM (P<.001). Cancer detection rate was 0.59% at FFDM and 0.38% at SFM (P=.02). There was no significant difference in PPVs. Median size of screening-detected invasive cancers was 14 mm at FFDM and 13 mm at SFM. Including cancers dismissed at consensus meetings, overall true-positive rate at baseline reading was 0.63% at FFDM and 0.43% at SFM (P=.04). Sensitivity was 77.4% at FFDM and 61.5% at SFM (P=.07); specificity was 96.5% and 97.9%, respectively (P<.005). Interval cancer rate was 17.4 at FFDM and 23.6 at SFM. The proportion of cancers with discordant double readings was comparable at FFDM and SFM.

CONCLUSION

FFDM resulted in a significantly higher cancer detection rate than did SFM. The PPVs were comparable for the two imaging modalities.

Effect of Computer-Aided Detection on Independent Double Reading of Paired Screen-Film and Full-Field Digital Screening Mammograms

OBJECTIVE

The purpose of this study was to evaluate the performance and potential contribution of computer-aided detection (CAD) to independent double reading of paired screen-film and full-field digital screening mammograms.

MATERIALS AND METHODS

The cases of 3,683 women who underwent both screen-film mammography and full-field digital mammography (FFDM) with independent double reading for each technique were followed for 2 years to include cancers detected in the interval between screening rounds and cancers detected at the next screening round. Fifty-five biopsy-proven cancers were diagnosed. The baseline screening mammograms of the 55 cancers were defined as having positive findings if at least one of two independent readers scored it 2 or higher on a 5-point rating scale. The baseline mammograms of interval (n = 10) or secondround (n = 16) cancers were retrospectively classified as overlooked (n = 2), minimal sign actionable (n = 8), minimal sign nonactionable (n = 5), and normal (n = 11). The baseline mammograms of these cases of cancer were evaluated with a CAD system, and the CAD results were compared (McNemar's test for paired proportions) with the findings at prospective independent double reading of mammograms obtained with each technique.

RESULTS

For FFDM, CAD sensitivity was 95% (37/39) compared with 64% (25/39) for double reading (p = 0.006), and for screen-film mammography, CAD sensitivity was 85% (33/39) compared with 77% (30/39) for prospective double reading (p = 0.57) of radiographically visible lesions in baseline mammograms. CAD correctly marked five (13%) of 39 cancers on screen-film mammography and 14 (36%) of 39 cancers on FFDM not detected at prospective independent double reading.

CONCLUSION

CAD showed the potential to increase the cancer detection rate for FFDM and for screen-film mammography in breast cancer screening performed with independent double reading.