Missed Breast Cancer: Effects of Subconscious Bias and Lesion Characteristics

Evaluating algorithmic bias on biomarker classification of breast cancer pathology reports

Objectives

This work evaluated algorithmic bias in biomarkers classification using electronic pathology reports from female breast cancer cases. Bias was assessed across 5 subgroups: cancer registry, race, Hispanic ethnicity, age at diagnosis, and socioeconomic status.

Materials and Methods

We utilized 594 875 electronic pathology reports from 178 121 tumors diagnosed in Kentucky, Louisiana, New Jersey, New Mexico, Seattle, and Utah to train 2 deep-learning algorithms to classify breast cancer patients using their biomarkers test results. We used balanced error rate (BER), demographic parity (DP), equalized odds (EOD), and equal opportunity (EOP) to assess bias.

Results

We found differences in predictive accuracy between registries, with the highest accuracy in the registry that contributed the most data (Seattle Registry, BER ratios for all registries >1.25). BER showed no significant algorithmic bias in extracting biomarkers (estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2) for race, Hispanic ethnicity, age at diagnosis, or socioeconomic subgroups (BER ratio <1.25). DP, EOD, and EOP all showed insignificant results.

Discussion

We observed significant differences in BER by registry, but no significant bias using the DP, EOD, and EOP metrics for socio-demographic or racial categories. This highlights the importance of employing a diverse set of metrics for a comprehensive evaluation of model fairness.

Conclusion

A thorough evaluation of algorithmic biases that may affect equality in clinical care is a critical step before deploying algorithms in the real world. We found little evidence of algorithmic bias in our biomarker classification tool. Artificial intelligence tools to expedite information extraction from clinical records could accelerate clinical trial matching and improve care.

Reducing Errors in Breast Imaging: Insights From Missed and Near-Missed Cases

Errors and misdiagnosis in breast imaging are significant concerns for breast imaging radiologists due to the negative impacts on patients and the high legal risks. Using missed and nearly missed diagnoses of breast cancer cases, this article introduces radiologists to common factors contributing to errors and misdiagnosis in breast imaging, including radiologist errors, improper imaging techniques, lesion characteristics, and work environment challenges. The article also provides practical recommendations and potential strategies to reduce these errors focusing on actions applicable to individual radiologists. Understanding the common causes of diagnostic errors in breast imaging and implementing targeted mitigating strategies can help radiologists improve diagnostic precision, reduce malpractice risk, and enhance patient care.

Incidental breast cancer on CT: factors associated with detection and relationship to prognostics and treatment options

OBJECTIVES

With the increasing use of CT, it may help detecting incidental breast cancers. Our study analysed the relationship between breast cancer detection on CT and features of the cancer, factors related to the scan and report, the treatment offered, and cancer prognostics, in NHS settings.

METHODOLOGY

56 scans in 42 patients were retrospectively included.

RESULTS

38 reports (67.9%) missed the breast cancers. Missed cancers were found to be smaller (P = .0042), progressed more by the time they were diagnosed (P = .0011), and their initial treatment was delayed by a median of 3.4 years (P < .0001). Cancers were more likely to be missed out of hours (P = .0485), in an outpatient reporting session (P = .0397), when the cancer presented as a circumscribed mass (P = .0196), and when the breasts were dense (P = .0250).

CONCLUSION

A significant percentage of breast cancer is missed on CT, with subsequent delay in starting treatment. Systematic approach when reporting, awareness of atypical cancer presentations, and minimizing distractions while reporting, may improve the detection of breast cancer on CT.

ADVANCES IN KNOWLEDGE

This study identified opportunities to detect, and the factors associated with missing and delayed treatment of, incidental breast cancer on CT, specifically in NHS settings. By increasing radiologists' awareness of those factors, it is hoped to prevent delay in treatment of this cohort of cancer patients.

Misdiagnosis in breast imaging: a statement paper from European Society Breast Imaging (EUSOBI)-Part 1: The role of common errors in radiology in missed breast cancer and implications of misdiagnosis

IMPORTANCE

Misdiagnosis in breast imaging can have significant implications for patients, healthcare providers, and the healthcare system as a whole.

OBSERVATIONS

Some of the potential implications of misdiagnosis in breast imaging include delayed diagnosis or false reassurance, which can result in a delay in treatment and potentially a worse prognosis. Misdiagnosis can also lead to unnecessary procedures, which can cause physical discomfort, anxiety, and emotional distress for patients, as well as increased healthcare costs. All these events can erode patient trust in the healthcare system and in individual healthcare providers. This can have negative implications for patient compliance with screening and treatment recommendations, as well as overall health outcomes. Moreover, misdiagnosis can also result in legal consequences for healthcare providers, including medical malpractice lawsuits and disciplinary action by licensing boards.

CONCLUSION AND RELEVANCE

To minimize the risk of misdiagnosis in breast imaging, it is important for healthcare providers to use appropriate imaging techniques and interpret images accurately and consistently. This requires ongoing training and education for radiologists and other healthcare providers, as well as collaboration and communication among healthcare providers to ensure that patients receive appropriate and timely care. If a misdiagnosis does occur, it is important for healthcare providers to communicate with patients and provide appropriate follow-up care to minimize the potential implications of the misdiagnosis. This may include repeat imaging, additional biopsies or other procedures, and referral to specialists for further evaluation and management.

KEY POINTS

Question What factors most contribute to and what implications stem from misdiagnosis in breast imaging? Findings Ongoing training and education for radiologists and other healthcare providers, as well as interdisciplinary collaboration and communication, is paramount. Clinical relevance Misdiagnosis in breast imaging can have significant implications for patients, healthcare providers, and the entire healthcare system.

Misdiagnosis in breast imaging: a statement paper from European Society Breast Imaging (EUSOBI)-Part 2: Main causes of errors in breast imaging and recommendations from European Society of Breast Imaging to limit misdiagnosis

IMPORTANCE

Breast cancer is one of the leading causes of negligence claims in radiology. The objective of this document is to describe the specific main causes of errors in breast imaging and provide European Society of Breast Imaging (EUSOBI) recommendations to try to minimize these.

OBSERVATIONS

Technical failures represent 17% of all mammographic diagnostic negligence claims. Mammography quality control protocol and dedicated training for technologists and radiologists are essential. Lack of consideration of the clinical context is a second critical issue, as a clinical abnormality is found in 80% of malpractice claims. EUSOBI emphasizes the importance of communication and clinical examination before the diagnostic investigation. Detection errors or misapplications of the lexicon or Breast Imaging Reporting Data System (BI-RADS) score account for 5% of malpractice claims and should be reduced by limiting radiologists' distraction or fatigue, and being aware of satisfaction of search errors and the importance of a personal systematic review. Errors related to pathological concordance and MDT review can be limited by the use of markers after biopsy and the use of standardized reports, which can aid communication with other specialities. Finally, errors related to tumor or patient factors should be discussed, considering the use of contrast-enhanced mammography and magnetic resonance imaging.

CONCLUSION

Several factors are responsible for misdiagnosis in breast cancer, including errors in the practice of the technician and/or radiologist (technical failures, lack of consideration of the clinical context, incorrect application of the BI-RADS score, false reassurances), lack of communication with other specialists or with the patient, and the type of tumor and breast parenchyma.

KEY POINTS

Question What factors most contribute to and what implications stem from misdiagnosis in breast imaging? Findings Ongoing training and education for radiologists and other healthcare providers, as well as interdisciplinary collaboration and communication is paramount. Clinical relevance Misdiagnosis in breast imaging can have significant implications for patients, healthcare providers, and the entire healthcare system.

Malpractice litigation in diagnostic radiology with special focus on cases in the abdomen and pelvis: A comprehensive analysis from a national legal database

OBJECTIVE

Diagnostic radiology is regarded as a "high-risk" specialty in the medical malpractice literature. This study examines the causes and patterns and types of medical malpractice litigation and outcomes in radiology in the United States, with a particular focus on diagnostic radiology errors involving the abdomen and pelvis.

METHODS

Malpractice suits in which the defendant was a radiologist in the United States from 2008 to 2018 were identified using LexisAdvance, a national legal database. 2775 cases were initially identified, and 1165 cases fit the inclusion criteria.

RESULTS

Diagnostic error was the most prevalent error type, (n = 925, 82.9 %), followed by procedural errors (n = 106, 9.5 %), communication errors (66 cases, 5.9 %), and mixed/other errors (n = 19, 1.7 %). Breast was the most common imaging modality implicated in medical error (n = 211, 26.4 % of total cases), followed by CT (n = 186, 23.3 %), and XR (n = 146, 18.3 %). Out-of-court settlement was the most common outcome (n = 402, 44.5 %), followed by a verdict ruled in favor of the defendant (n = 246, 27.2 %) and case dismissal (n = 131, 14.5 %). The average award in a settlement was $1,500,690 USD (range: $25,000- $10,200,000 USD). The average award in a jury verdict for the plaintiff was $2,857,203 USD (range: $60,000- $31,490,000 USD), and the average award in arbitration for the plaintiff was $1,354,497 USD (range: $200,000- $2,800,000 USD). The gastrointestinal (GI) system and the genitourinary (GU) system accounted for 51.9 % and 25.9 % of errors in the abdomen and pelvis, respectively.

DISCUSSION

Diagnostic error was the most prevalent source of error leading to malpractice litigation. Breast imaging was the most frequently implicated imaging modality in litigations, followed closely by CT and XR. A majority of cases were resolved through out-of-court settlement or with judgments in favor of the defendant radiologists. However, in cases with trial judgments in favor of the plaintiff, average financial awards were higher than out-of-court settlements. Abdomen and pelvic involvement accounted for frequent sources of error.

Breast Cancer Detection with Standalone AI versus Radiologist Interpretation of Unilateral Surveillance Mammography after Mastectomy

Background Limited data are available regarding the accuracy of artificial intelligence (AI) algorithms trained on bilateral mammograms for second breast cancer surveillance in patients with a personal history of breast cancer treated with unilateral mastectomy. Purpose To compare the performance of standalone AI for second breast cancer surveillance on unilateral mammograms with that of radiologists reading mammograms without AI assistance. Materials and Methods In this retrospective institutional database study, patients who were diagnosed with breast cancer between January 2001 and December 2018 and underwent postmastectomy surveillance mammography from January 2011 to March 2023 were included. Radiologists' mammogram interpretations without AI assistance were collected from these records and compared with AI interpretations of the same mammograms. The reference standards were histologic examination and 1-year follow-up data. The cancer detection rate per 1000 screening examinations, sensitivity, and specificity of standalone AI and the radiologists' interpretations without AI were compared using the McNemar test. Results Among the 4184 asymptomatic female patients (mean age, 52 years), 111 (2.7%) had contralateral second breast cancer. The cancer detection rate (17.4 per 1000 examinations [73 of 4184]; 95% CI: 13.7, 21.9) and sensitivity (65.8% [73 of 111]; 95% CI: 56.2, 74.5) were greater for standalone AI than for radiologists (14.6 per 1000 examinations [61 of 4184]; 95% CI: 11.2, 18.7; P = .01; 55.0% [61 of 111]; 95% CI: 45.2, 64.4; P = .01). The specificity was lower for standalone AI than for radiologists (91.5% [3725 of 4073]; 95% CI: 90.6, 92.3 vs 98.1% [3996 of 4073]; 95% CI: 97.6, 98.5; P < .001). AI detected 16 of 50 (32%) cancers missed by radiologists; however, 34 of 111 (30.6%) breast cancers were missed by both radiologists and AI. Conclusion Standalone AI for surveillance mammography showed higher sensitivity with lower specificity for contralateral breast cancer detection in patients treated with unilateral mastectomy than radiologists without AI assistance. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Philpotts in this issue.

Artificial intelligence for breast cancer screening in mammography (AI-STREAM): preliminary analysis of a prospective multicenter cohort study

Artificial intelligence (AI) improves the accuracy of mammography screening, but prospective evidence, particularly in a single-read setting, remains limited. This study compares the diagnostic accuracy of breast radiologists with and without AI-based computer-aided detection (AI-CAD) for screening mammograms in a real-world, single-read setting. A prospective multicenter cohort study is conducted within South Korea's national breast cancer screening program for women. The primary outcomes are screen-detected breast cancer within one year, with a focus on cancer detection rates (CDRs) and recall rates (RRs) of radiologists. A total of 24,543 women are included in the final cohort, with 140 (0.57%) screen-detected breast cancers. The CDR is significantly higher by 13.8% for breast radiologists using AI-CAD (n = 140 [5.70‰]) compared to those without AI (n = 123 [5.01‰]; p < 0.001), with no significant difference in RRs (p = 0.564). These preliminary results show a significant improvement in CDRs without affecting RRs in a radiologist's standard single-reading setting (ClinicalTrials.gov: NCT05024591).

Simulation training in mammography with AI-generated images: a multireader study

OBJECTIVES

The interpretation of mammograms requires many years of training and experience. Currently, training in mammography, like the rest of diagnostic radiology, is through institutional libraries, books, and experience accumulated over time. We explore whether artificial Intelligence (AI)-generated images can help in simulation education and result in measurable improvement in performance of residents in training.

METHODS

We developed a generative adversarial network (GAN) that was capable of generating mammography images with varying characteristics, such as size and density, and created a tool with which a user could control these characteristics. The tool allowed the user (a radiology resident) to realistically insert cancers within different regions of the mammogram. We then provided this tool to residents in training. Residents were randomized into a practice group and a non-practice group, and the difference in performance before and after practice with such a tool (in comparison to no intervention in the non-practice group) was assessed.

RESULTS

Fifty residents participated in the study, 27 underwent simulation training, and 23 did not. There was a significant improvement in the sensitivity (7.43 percent, significant at p-value = 0.03), negative predictive value (5.05 percent, significant at p-value = 0.008) and accuracy (6.49 percent, significant at p-value = 0.01) among residents in the detection of cancer on mammograms after simulation training.

CONCLUSION

Our study shows the value of simulation training in diagnostic radiology and explores the potential of generative AI to enable such simulation training.

CLINICAL RELEVANCE STATEMENT

Using generative artificial intelligence, simulation training modules can be developed that can help residents in training by providing them with a visual impression of a variety of different cases.

KEY POINTS

Generative networks can produce diagnostic imaging with specific characteristics, potentially useful for training residents. Training with generating images improved residents' mammographic diagnostic abilities. Development of a game-like interface that exploits these networks can result in improvement in performance over a short training period.

Machine Learning-Based Approaches for Breast Density Estimation from Mammograms: A Comprehensive Review

Breast cancer, as of 2022, is the most prevalent type of cancer in women. Breast density-a measure of the non-fatty tissue in the breast-is a strong risk factor for breast cancer that can be estimated from mammograms. The importance of studying breast density is twofold. First, high breast density can be a factor in lowering mammogram sensitivity, as dense tissue can mask tumors. Second, higher breast density is associated with an increased risk of breast cancer, making accurate assessments vital. This paper presents a comprehensive review of the mammographic density estimation literature, with an emphasis on machine-learning-based approaches. The approaches reviewed can be classified as visual, software-, machine learning-, and segmentation-based. Machine learning methods can be further broken down into two categories: traditional machine learning and deep learning approaches. The most commonly utilized models are support vector machines (SVMs) and convolutional neural networks (CNNs), with classification accuracies ranging from 76.70% to 98.75%. Major limitations of the current works include subjectivity and cost-inefficiency. Future work can focus on addressing these limitations, potentially through the use of unsupervised segmentation and state-of-the-art deep learning models such as transformers. By addressing the current limitations, future research can pave the way for more reliable breast density estimation methods, ultimately improving early detection and diagnosis.

Cancers not detected in one-view breast tomosynthesis screening-characteristics and reasons for non-detection

OBJECTIVES

Limited understanding exists regarding non-detected cancers in digital breast tomosynthesis (DBT) screening. This study aims to classify non-detected cancers into true or false negatives, compare them with true positives, and analyze reasons for non-detection.

MATERIALS AND METHODS

Conducted between 2010 and 2015, the prospective single-center Malmö Breast Tomosynthesis Screening Trial (MBTST) compared one-view DBT and two-view digital mammography (DM). Cancers not detected by DBT, i.e., interval cancers, those detected in the next screening round, and those only identified by DM, underwent a retrospective informed review by in total four breast radiologists. Reviewers classified cancers into true negative, false negative, or non-visible based on both DBT and DM findings and assessed radiographic appearances at screening and diagnosis, breast density, and reasons for non-detection. Statistics included the Pearson X2 test.

RESULTS

In total, 89 cancers were not detected with DBT in the MBTST; eight cancers were solely in the DM reading mode, 59 during subsequent DM screening rounds, and 22 interval cancers. The proportion of cancers classified as false negative was 25% (22/89) based on DBT, compared with 18% (14/81) based on DM screening. The primary reason for false negatives was normal-appearing density, 50% (11/22). False negatives exhibited lower rates of high breast density, 36% (8/22), compared with true positives, 61% (78/129), p = 0.04, and spiculated densities were less frequent in false negatives, 41% (9/22) compared with true positives, 68% (88/129), p = 0.01.

CONCLUSION

False negatives in one-view DBT screening commonly presented with spiculated features, but less frequently than true positives, and were missed or misinterpreted due to benign appearances.

KEY POINTS

Question Cancers not detected in digital breast tomosynthesis screening, including false negatives, remain partly unexplored. Findings The most common reason behind false-negative cancers in a large screening trial was a normal-appearing density. Clinical relevance Recognizing the factors contributing to false negative findings in digital breast tomosynthesis screening is essential to further improve cancer detection.

A Novel Hybrid Model for Automatic Non-Small Cell Lung Cancer Classification Using Histopathological Images

Background/Objectives: Despite recent advances in research, cancer remains a significant public health concern and a leading cause of death. Among all cancer types, lung cancer is the most common cause of cancer-related deaths, with most cases linked to non-small cell lung cancer (NSCLC). Accurate classification of NSCLC subtypes is essential for developing treatment strategies. Medical professionals regard tissue biopsy as the gold standard for the identification of lung cancer subtypes. However, since biopsy images have very high resolutions, manual examination is time-consuming and depends on the pathologist's expertise. Methods: In this study, we propose a hybrid model to assist pathologists in the classification of NSCLC subtypes from histopathological images. This model processes deep, textural and contextual features obtained by using EfficientNet-B0, local binary pattern (LBP) and vision transformer (ViT) encoder as feature extractors, respectively. In the proposed method, each feature matrix is flattened separately and then combined to form a comprehensive feature vector. The feature vector is given as input to machine learning classifiers to identify the NSCLC subtype. Results: We set up 13 different training scenarios to test 4 different classifiers: support vector machine (SVM), logistic regression (LR), light gradient boosting machine (LightGBM) and extreme gradient boosting (XGBoost). Among these scenarios, we obtained the highest classification accuracy (99.87%) with the combination of EfficientNet-B0 + LBP + ViT Encoder + SVM. The proposed hybrid model significantly enhanced the classification accuracy of NSCLC subtypes. Conclusions: The integration of deep, textural, and contextual features assisted the model in capturing subtle information from the images, thereby reducing the risk of misdiagnosis and facilitating more effective treatment planning.

Interval carcinomas in a breast cancer screening program (2007-2018): Characteristics and prognosis

BACKGROUND AND AIMS

To analyze the radiologic and histologic characteristics of screening and interval cancers diagnosed in the period comprising 2007 through 2018 in a total of six rounds of a population-based breast cancer screening program.

MATERIAL AND METHODS

We analyzed 1395 carcinomas detected at screening and 300 interval carcinomas diagnosed in women aged 50-69 years old who underwent digital mammography every two years during the study period. Screening mammograms were read once. To classify the interval carcinomas, we retrospectively reviewed (blind reading followed by unblinded reading) at the end of each round, recording the radiologic findings, breast density, histologic characteristics, phenotype, and surgical treatment.

RESULTS

The interval carcinomas were classified as true interval cancers in 156 (52%) cases, false-negatives in 62 (20.5%), minimal signs in 39 (13%), occult lesions in 29 (9.5%), and impossible to classify in 14 (5%). Retrospectively, the most common radiologic findings in the false-negative cases were mass/asymmetry (64%), calcifications (16%), and distortion (13%); the most common radiologic findings in the cases with minimal signs were mass/asymmetry (58%) and calcifications (31%). There were significant differences in the histologic characteristics between cancers detected at screening and interval cancers: T1a-b [9% of the interval cancers vs. 34% of those detected at screening, P < .001]; T1c [30% of the interval cancers vs. 44% of those detected at screening P < .001], T2 or greater [61% of the interval cancers vs. 22% of those detected at screening P < .001], and the degree of axillary involvement [45% of the interval cancers vs. 27% of those detected at screening, P < .001]. There were also significant differences between cancers detected at screening and interval cancers in the proportion of cases with more aggressive subtypes (HER2+ and triple-negative): [38.5% of the interval cancers vs. 23% of those detected at screening, P < .001]. A significantly higher proportion of interval cancers were treated with mastectomies [80% vs. 67% of those detected at screening, P < .001].

CONCLUSIONS

About 20% of interval cancers were evident on screening mammograms. The most common radiologic finding in interval cancers was asymmetry/mass. Interval cancers are diagnosed at a more advanced stage than cancers identified at screening, so they sre more often treated by mastectomy. Reviewing interval cancers is essential for quality control in screening programs.

Positive predictive value of malignancy for additional calcifications found during evaluation of a synchronous breast cancer

PURPOSE

To evaluate the positive predictive value and factors predictive of malignancy of additional calcifications in the pre-therapeutic work-up of a synchronous breast cancer.

MATERIALS AND METHODS

Institutional review board approval was obtained for this retrospective study and informed consent was waved. Consecutive patients referred to our center between January 1st 2018 and December 31st 2022 for a breast cancer and who presented additional calcifications detected during the pretreatment work-up were eligible for inclusion in this study. Morphology, distribution and BI-RADS category of the calcifications were assessed in consensus by 3 radiologists specialized in breast imaging. Side and distance from the cancer were collected. The predictive value of malignancy of the calcifications was calculated for each BI-RADS category. Factors associated with malignancy were evaluated by logistic non-conditional regression on univariate and multivariate analysis.

RESULTS

One hundred and thirteen clusters of calcifications in 103 patients were included. Among the groups of calcifications 41 % were malignant, 31 % benign and 28 % were atypia on biopsy. After exclusion of the non-operated atypia, 50.5 % of additional calcifications were ultimately malignant and 49.5 % were benign. The predictive value of malignancy was 20.7 %; 40.7 %; 63 %; 85.7 % and 100 % for category BI-RADS 3, 4a, 4B, 4c and 5 respectively. On multivariate analysis, multifocality or centricity of the index tumour (P = 0.01), BI-RADS classifications (P = 0.0001) and location ipsilateral less than 35 mm to the index cancer (P = 0.008) of the additional calcifications were found to be independent predictors of malignancy. Sixty percent of calcifications were not described on the initial out-center diagnostic work-up.

CONCLUSION

Additional calcifications detected during the pretreatment work-up of a breast cancer are associated with a higher probability of malignancy than in a screening population and require biopsy even when demonstrating probably benign (BI-RADS 3) features.

Assessing the Factors Leading to Missed Breast Cancer Diagnoses in Mammography Among Pakistani Women: A Prospective Cross-Sectional Study

Objective To determine the frequency of false-negative mammograms, and identify the factors contributing to missed breast cancer diagnoses in Pakistani women. Materials and methods This descriptive, prospective cross-sectional study was conducted at a tertiary care hospital from December 15, 2020, to December 10, 2023, including 150 women aged 30 to 60 who underwent bilateral mammography and concurrent breast ultrasound. The study analyzed the frequency and causes of false negatives, categorizing them into patient-related, tumor-related, technical-related, and provider-related factors. Stratification was performed based on age groups and Breast Imaging Reporting and Data System (BI-RADS) scores, and statistical significance was assessed using Chi-square tests. Results The study found a 5.1% frequency of false-negative mammograms. Lesion-related factors were seen in 59 (39.7%) patients; patient-related factors were seen in 40 (26.7%) patients; provider-related factors were seen in 29 (19.3%) patients; and technical-related factors were seen in 22 (26.7%) patients. Conclusion Dense breast tissue significantly contributes to missed breast cancer diagnoses in Pakistani women. While lesion-related, provider-related, and technical-related factors uniformly affect mammography outcomes, addressing patient-specific challenges - particularly in younger women with dense breasts - is crucial. The study suggests incorporating supplementary imaging modalities, like ultrasound, in routine screening for better detection, potentially informing national breast cancer screening guidelines in Pakistan.

Screening Outcomes of Mammography with AI in Dense Breasts: A Comparative Study with Supplemental Screening US

Background Comparative performance between artificial intelligence (AI) and breast US for women with dense breasts undergoing screening mammography remains unclear. Purpose To compare the performance of mammography alone, mammography with AI, and mammography plus supplemental US for screening women with dense breasts, and to investigate the characteristics of the detected cancers. Materials and Methods A retrospective database search identified consecutive asymptomatic women (≥40 years of age) with dense breasts who underwent mammography plus supplemental whole-breast handheld US from January 2017 to December 2018 at a primary health care center. Sequential reading for mammography alone and mammography with the aid of an AI system was conducted by five breast radiologists, and their recall decisions were recorded. Results of the combined mammography and US examinations were collected from the database. A dedicated breast radiologist reviewed marks for mammography alone or with AI to confirm lesion identification. The reference standard was histologic examination and 1-year follow-up data. The cancer detection rate (CDR) per 1000 screening examinations, sensitivity, specificity, and abnormal interpretation rate (AIR) of mammography alone, mammography with AI, and mammography plus US were compared. Results Among 5707 asymptomatic women (mean age, 52.4 years ± 7.9 [SD]), 33 (0.6%) had cancer (median lesion size, 0.7 cm). Mammography with AI had a higher specificity (95.3% [95% CI: 94.7, 95.8], P = .003) and lower AIR (5.0% [95% CI: 4.5, 5.6], P = .004) than mammography alone (94.3% [95% CI: 93.6, 94.8] and 6.0% [95% CI: 5.4, 6.7], respectively). Mammography plus US had a higher CDR (5.6 vs 3.5 per 1000 examinations, P = .002) and sensitivity (97.0% vs 60.6%, P = .002) but lower specificity (77.6% vs 95.3%, P < .001) and higher AIR (22.9% vs 5.0%, P < .001) than mammography with AI. Supplemental US alone helped detect 12 cancers, mostly stage 0 and I (92%, 11 of 12). Conclusion Although AI improved the specificity of mammography interpretation, mammography plus supplemental US helped detect more node-negative early breast cancers that were undetected using mammography with AI. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Whitman and Destounis in this issue.

Artificial Intelligence for Breast Cancer Detection on Mammography: Factors Related to Cancer Detection

RATIONALE AND OBJECTIVES

Little is known about the factors affecting the Artificial Intelligence (AI) software performance on mammography for breast cancer detection. This study was to identify factors associated with abnormality scores assigned by the AI software.

MATERIALS AND METHODS

A retrospective database search was conducted to identify consecutive asymptomatic women who underwent breast cancer surgery between April 2016 and December 2019. A commercially available AI software (Lunit INSIGHT, MMG, Ver. 1.1.4.0) was used for preoperative mammography to assign individual abnormality scores to the lesions and score of 10 or higher was considered as positive detection by AI software. Radiologists without knowledge of the AI results retrospectively assessed the mammographic density and classified mammographic findings into positive and negative finding. General linear model (GLM) analysis was used to identify the clinical, pathological, and mammographic findings related to the abnormality scores, obtaining coefficient β values that represent the mean difference per unit or comparison with the reference value. Additionally, the reasons for non-detection by the AI software were investigated.

RESULTS

Among the 1001 index cancers (830 invasive cancers and 171 ductal carcinoma in situs) in 1001 patients, 717 (72%) were correctly detected by AI, while the remaining 284 (28%) were not detected. Multivariable GLM analysis showed that abnormal mammography findings (β = 77.0 for mass, β = 73.1 for calcification only, β = 49.4 for architectural distortion, and β = 47.6 for asymmetry compared to negative; all Ps < 0.001), invasive tumor size (β = 4.3 per 1 cm, P < 0.001), and human epidermal growth receptor type 2 (HER2) positivity (β = 9.2 compared to hormone receptor positive, HER2 negative, P = 0.004) were associated with higher mean abnormality score. AI failed to detect small asymmetries in extremely dense breasts, subcentimeter-sized or isodense lesions, and faint amorphous calcifications.

CONCLUSION

Cancers with positive abnormal mammographic findings on retrospective review, large invasive size, HER2 positivity had high AI abnormality scores. Understanding the patterns of AI software performance is crucial for effectively integrating AI into clinical practice.

Discordant and false-negative interpretations at digital breast tomosynthesis in the prospective Oslo Tomosynthesis Screening Trial (OTST) using independent double reading

OBJECTIVES

To analyze discordant and false-negatives of double reading digital breast tomosynthesis (DBT) versus digital mammography (DM) including reading times in the Oslo Tomosynthesis Screening Trial (OTST), and reclassify these in a retrospective reader study as missed, minimal sign, or true-negatives.

METHODS

The prospective OTST comparing double reading DBT vs. DM had paired design with four parallel arms: DM, DM + computer aided detection, DBT + DM, and DBT + synthetic mammography. Eight radiologists interpreted images in batches using a 5-point scale. Reading time was automatically recorded. A retrospective reader study including four radiologists classified screen-detected cancers with at least one false-negative score and screening examinations of interval cancers as negative, non-specific minimal sign, significant minimal sign, and missed; the two latter groups are defined "actionable." Statistics included chi-square, Fisher's exact, McNemar's, and Mann-Whitney U tests.

RESULTS

Discordant rate (cancer missed by one reader) for screen-detected cancers was overall comparable (DBT (31% [71/227]) and DM (30% [52/175]), p = .81), significantly lower at DBT for spiculated cancers (DBT, 19% [20/106] vs. DM, 36% [38/106], p = .003), but high (28/49 = 57%, p = 0.001) for DBT-only detected spiculated cancers. Reading time and sensitivity varied among readers. False-negative DBT-only detected spiculated cancers had shorter reading time than true-negatives in 46% (13/28). Retrospective evaluation classified the following DBT exams "actionable": three missed by both readers, 95% (39/41) of discordant cancers detected by both modes, all 30 discordant DBT-only cancers, 25% (13/51) of interval cancers.

CONCLUSIONS

Discordant rate was overall comparable for DBT and DM, significantly lower at DBT for spiculated cancers, but high for DBT-only detected spiculated lesions. Most false-negative screen-detected DBT were classified as "actionable."

CLINICAL RELEVANCE STATEMENT

Retrospective evaluation of false-negative interpretations from the Oslo Tomosynthesis Screening Trial shows that most discordant and several interval cancers could have been detected at screening. This underlines the potential for modern AI-based reading aids and triage, as high-volume screening is a demanding task.

KEY POINTS

• Digital breast tomosynthesis (DBT) screening is more sensitive and has higher specificity compared to digital mammography screening, but high-volume DBT screening is a demanding task which can result in high discordance rate among readers. • Independent double reading DBT screening had overall comparable discordance rate as digital mammography, lower for spiculated masses seen on both modalities, and higher for small spiculated cancer seen only on DBT. • Almost all discordant digital breast tomosynthesis-detected cancers (72 of 74) and 25% (13 of 51) of the interval cancers in the Oslo Tomosynthesis Screening Trial were retrospectively classified as actionable and could have been detected by the readers.

Breast imaging with an ultra-low field MRI scanner: a pilot study

Breast cancer screening is necessary to reduce mortality due to undetected breast cancer. Current methods have limitations, and as a result many women forego regular screening. Magnetic resonance imaging (MRI) can overcome most of these limitations, but access to conventional MRI is not widely available for routine annual screening. Here, we used an MRI scanner operating at ultra-low field (ULF) to image the left breasts of 11 women (mean age, 35 years ±13 years) in the prone position. Three breast radiologists reviewed the imaging and were able to discern the breast outline and distinguish fibroglandular tissue (FGT) from intramammary adipose tissue. Additionally, the expert readers agreed on their assessment of the breast tissue pattern including fatty, scattered FGT, heterogeneous FGT, and extreme FGT. This preliminary work demonstrates that ULF breast MRI is feasible and may be a potential option for comfortable, widely deployable, and low-cost breast cancer diagnosis and screening.

A case report of challenges in distinguishing gastroesophageal junction hepatoid adenocarcinoma from testicular germ cell tumor: Insights for improved diagnosis with gene expression profiling

Gastroesophageal junction hepatoid adenocarcinoma is a rare form of gastroesophageal cancer. We present a case of a 38-year-old man with no significant medical history who was diagnosed with gastroesophageal junction hepatoid adenocarcinoma but initially misdiagnosed with a testicular germ cell tumor, given the elevated alpha-feto protein and poorly differentiated pathology. We will elaborate on the importance of gene expression profiling in modern oncology to better define the tumor of origin in patients with cancer of unknown primary origin, how it helped us to diagnose gastroesophageal junction hepatoid adenocarcinoma and how it can help identify potential additional therapeutic targets in some cases. Due to the rarity of this subtype of gastroesophageal junction cancer there is a lack of standard therapeutic options, and we will discuss the most commonly used treatment regimens. The patient underwent three lines of antineoplastic therapy and unfortunately passed after 51 weeks of follow-up.

The effect of a pre-scheduled appointment on attendance in a population-based mammography screening programme

BACKGROUND

Pre-scheduled appointments can increase attendance in breast cancer screening programmes compared to 'open invitations' but relatively few randomized controlled trials exist. We investigated the effect of a pre-scheduled appointment on uptake in the Flemish population-based mammography screening programme.

METHODS

Between September and December 2022, a total of 4798 women were randomly assigned to receive either a pre-scheduled appointment or open invitation. The difference in attendance was compared with Poisson regression analysis for the primary endpoint (attendance ≤92 days after date of invitation), yielding relative risks (RRs). This was done separately for three groups: women invited to a mobile unit and a history of nonattendance (group M-NA); women invited to a hospital-based unit and a history of nonattendance (group HB-NA); women invited to a hospital-based unit and a history of irregular attendance (group HB-IA). There were no women invited to a mobile unit and a history of irregular attendance.

RESULTS

The RRs in favour of the pre-scheduled appointment were 2.3 [95% confidence interval (CI) 1.80-2.88], 1.8 (95% CI 1.07-2.97) and 1.8 (95% CI 1.43-2.39), for groups M-NA, HB-NA and HB-IA, respectively. We found no statistically significant difference between the various RRs. The respective absolute gains in attendance between pre-scheduled appointment and open invitation were 8.3%, 4.4% and 15.8%.

CONCLUSIONS

Sending an invitation with a pre-scheduled appointment is an effective tool to increase screening attendance in both mobile and hospital-based screening units. The pre-scheduled appointment is associated with a considerable absolute gain in attendance which varies depending on the screening history.

False-Positive and False-Negative Contrast-enhanced Mammograms: Pitfalls and Strategies to Improve Cancer Detection

Contrast-enhanced mammography (CEM) is a relatively new breast imaging modality that uses intravenous contrast material to increase detection of breast cancer. CEM combines the structural information of conventional mammography with the functional information of tumor neovascularity. Initial studies have demonstrated that CEM and MRI perform with similar accuracies, with CEM having a slightly higher specificity (fewer false positives), although larger studies are needed. There are various reasons for false positives and false negatives at CEM. False positives at CEM can be caused by benign lesions with vascularity, including benign tumors, infection or inflammation, benign lesions in the skin, and imaging artifacts. False negatives at CEM can be attributed to incomplete or inadequate visualization of lesions, marked background parenchymal enhancement (BPE) obscuring cancer, lack of lesion contrast enhancement due to technical issues or less-vascular cancers, artifacts, and errors of lesion perception or characterization. When possible, real-time interpretation of CEM studies is ideal. If additional views are necessary, they may be obtained while contrast material is still in the breast parenchyma. Until recently, a limitation of CEM was the lack of CEM-guided biopsy capability. However, in 2020, the U.S. Food and Drug Administration cleared two devices to support CEM-guided biopsy using a stereotactic biopsy technique. The authors review various causes of false-positive and false-negative contrast-enhanced mammograms and discuss strategies to reduce these diagnostic errors to improve cancer detection while mitigating unnecessary additional imaging and procedures. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

Investigating the impact of cognitive biases in radiologists' image interpretation: A scoping review

RATIONALE AND OBJECTIVE

Image interpretation is a fundamental aspect of radiology. The treatment and management of patients relies on accurate and timely imaging diagnosis. However, errors in radiological reports can negatively impact on patient health outcomes. These misdiagnoses can be caused by several different errors, but cognitive biases account for 74 % of all image interpretation errors. There are many biases that can impact on a radiologist's perception and cognitive processes. Several recent narrative reviews have discussed these cognitive biases and have offered possible strategies to mitigate their effects. However, these strategies remain untested. Therefore, the purpose of this scoping review is to evaluate the current knowledge on the extent that cognitive biases impact on medical image interpretation.

MATERIAL AND METHODS

Scopus and Medline Databases were searched using relevant keywords to identify papers published between 2012 and 2022. A subsequent hand search of the narrative reviews was also performed. All studies collected were screened and assessed against the inclusion and exclusion criteria.

RESULTS

Twenty-four publications were included and categorised into five main themes: satisfaction of search, availability bias, hindsight bias, framing bias and other biases. From these studies, there were mixed results regarding the impact of cognitive biases, highlighting the need for further investigation in this area. Moreover, the limited and untested debiasing methods offered by a minority of the publications and narrative reviews also suggests the need for further research. The potential of role of artificial intelligence is also highlighted to further assist radiologists in identifying and mitigating these cognitive biases.

CONCLUSION

Cognitive biases can impact radiologists' image interpretation, however the effectiveness of debiasing strategies remain largely untested.

Standalone AI for Breast Cancer Detection at Screening Digital Mammography and Digital Breast Tomosynthesis: A Systematic Review and Meta-Analysis

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.

Critical Review of Oncologic Medical Malpractice Claims Against Orthopaedic Surgeons

INTRODUCTION

The purpose of this study was to determine the most common allegations for malpractice litigation brought against orthopaedic surgeons for oncologic matters and the resulting verdicts.

METHODS

The Westlaw Legal research database was queried for malpractice cases filed against orthopaedic surgeons for oncologic matters in the United States after 1980. Plaintiff demographics, state of filing, allegations, and outcomes of lawsuits were recorded and reported accordingly.

RESULTS

A total of 36 cases met the inclusion and exclusion criteria and were subsequently included in the final analysis. The overall rate of cases filed remained consistent through the past four decades and was primarily related to a primary sarcoma diagnosis in adult women. The primary reason for litigation was failure to diagnose a primary malignant sarcoma (42%) followed by failure to diagnose unrelated carcinoma (19%). The most common states of filing were primarily located in the Northeast (47%), where a plaintiff verdict was also more commonly encountered as compared with other regions. Damages awarded averaged $1,672,500 with a range of $134, 231 to $6,250,000 and a median of $918,750.

CONCLUSION

Failure to diagnose primary malignant sarcoma and unrelated carcinoma was the most common reason for oncologic litigation brought against orthopaedic surgeons. Although most of the cases ruled in favor of the defendant surgeon, it is important for orthopaedic surgeons to be aware of the potential errors that not only prevent litigation but also improve patient care.

Audit of Prior Screening Mammograms of Screen-Detected Cancers: Implications for the Delay in Breast Cancer Detection

When cancer is detected in a screening mammogram, on occasion retrospective review of prior screening (pre-index) mammograms indicates a likely presence of cancer. These missed cancers during pre-index screens constitute a delay in detection and diagnosis. This study was undertaken to quantify the missed cancer rate by auditing pre-index screens to improve the quality of mammography screening practice. From a cohort of 135 screen-detected cancers, 120 pre-index screening mammograms could be retrieved and served as the study sample. A consensus read by 2 radiologists who interpreted the pre-index screens in an unblinded manner with full knowledge of cancer location, cancer type, lesion type, and pathology served as the truth or reference standard. Five radiologists interpreted the pre-index screens in a blinded manner. Established performance metrics such as sensitivity and specificity were quantified for each reader in interpreting these pre-index screens in a blinded manner. All five radiologists detected lesions in 8/120 (6.7%) screens. Excluding the 2 readers whose performance was close to random, all the 3 remaining readers detected lesions in 13 pre-index screens. This indicates that there is a delay in diagnosis by at least one cycle from 8/120 (6.7%) to 13/120 (10.8%). There were no observable trends in terms of either the cancer type or the lesion type. Auditing prior screening mammograms in screen-detected cancers can help in identifying the proportion of cases that were missed during interpretation and help in quantifying the delay in breast cancer detection.

Peer learning in breast imaging

With the increasing focus on quality and safety in medicine, radiology practices are increasingly transitioning from traditional score-based peer review to peer learning. Participation in a peer learning program can increase learning, practice improvement, and cultivation of interpersonal relationships in a non-punitive environment. As breast imaging errors are the most cited in medical malpractice cases, learning and attention to and reduction of these errors in breast imaging are especially important. We describe the strengths of a peer learning program, implementation process in a breast imaging program, challenges to overcome, and strategies to support success.

True and Missed Interval Cancer in Organized Mammographic Screening: A Retrospective Review Study of Diagnostic and Prior Screening Mammograms

RATIONALE AND OBJECTIVES

To explore radiological aspects of interval breast cancer in a population-based screening program.

MATERIALS AND METHODS

We performed a consensus-based informed review of mammograms from diagnosis and prior screening from women diagnosed with interval cancer 2004-2016 in BreastScreen Norway. Cases were classified as true (no findings on prior screening mammograms), occult (no findings at screening or diagnosis), minimal signs (minor/non-specific findings) and missed (obvious findings). We analyzed mammographic findings, density, time since prior screening, and histopathological characteristics between the classification groups.

RESULTS

The study included 1010 interval cancer cases. Mean age at diagnosis was 61 years (SD = 6), mean time between screening and diagnosis 14 months (SD = 7). A total of 48% (479/1010) were classified as true or occult, 28% (285/1010) as minimal signs and 24% (246/1010) as missed. We observed no differences in mammographic density between the groups, except from a higher percentage of dense breasts in women with occult cancer. Among cancers classified as missed, about 1/3 were masses and 1/3 asymmetries at prior screening. True interval cancers were diagnosed later in the screening interval than the other classification categories. No differences in histopathological characteristics were observed between true, minimal signs and missed cases.

CONCLUSION

In an informed review, 24% of the interval cancers were classified as missed based on visibility and mammographic findings on prior screening mammograms. Three out of four true interval cancers were diagnosed in the second year of the screening interval. We observed no statistical differences in histopathological characteristics between true and missed interval cancers.

What's Hot in Breast MRI

Several articles in the literature have demonstrated a promising role for breast MRI techniques that are more economic in total exam time than others when used as supplement to mammography for detection and diagnosis of breast cancer. There are many technical factors that must be considered in the shortened breast MRI protocols to cut down time of standard ones, including using optimal fat suppression, gadolinium-chelates intravascular contrast administrations for dynamic imaging with post processing subtractions and maximum intensity projections (MIP) high spatial and temporal resolution among others. Multiparametric breast MRI that includes both gadolinium-dependent, i.e., dynamic contrast-enhanced (DCE-MRI) and gadolinium-free techniques, i.e., diffusion-weighted/diffusion-tensor magnetic resonance imaging (DWI/DTI) are shown by several investigators that can provide extremely high sensitivity and specificity for detection of breast cancer. This article provides an overview of the proven indications for breast MRI including breast cancer screening for higher than average risk, determining chemotherapy induced tumor response, detecting residual tumor after incomplete surgical excision, detecting occult cancer in patients presenting with axillary node metastasis, detecting residual tumor after incomplete breast cancer surgical excision, detecting cancer when results of conventional imaging are equivocal, as well patients suspicious of having breast implant rupture. Despite having the highest sensitivity for breast cancer detection, there are pitfalls, however, secondary to false positive and false negative contrast enhancement and contrast-free MRI techniques. Awareness of the strengths and limitations of different approaches to obtain state of the art MR images of the breast will facilitate the work-up of patients with suspicious breast lesions.

Errors in Breast Imaging: How to Reduce Errors and Promote a Safety Environment

Medical errors have a significant impact on patient care, worker safety, and health care finances. Breast imaging has the most common cause of organ-related misdiagnosis subject to malpractice suits. In order to effectively develop strategies to prevent errors, breast imaging radiologists must first understand the underlying causes of the errors that occur in the breast imaging setting. Errors in breast imaging can be related to errors in interpretation, improper workup of imaging findings, procedural errors, or errors in communication to the patient or other medical staff. The Yorkshire contributory factors framework was developed to identify factors that contribute to the errors in a hospital setting and can be adapted for use in the breast imaging setting. Within this framework, active failures refer to errors that directly affect the patient. Active errors include slips (including biases), lapses, and mistakes. The framework describes how active errors often result from factors that occur uphill from these active errors at different levels within the system. Once error causes are understood, there are concrete strategies and tools that breast imaging radiologists can implement to decrease adverse events, reduce medical errors, and promote a safety environment in the breast imaging clinic. Error mitigation tools can be summarized using the acronym SAFE, which includes support the team, ask questions, focus on a task, and effectively communicate/ensure equipment optimization/safe environment. Knowledge of errors commonly seen in a breast imaging clinic represent an opportunity for constructive changes and, ultimately, improved health care delivery.

Breast MRI: False-Negative Results and Missed Opportunities

Breast MRI is the most sensitive modality for the detection of breast cancer. However, false-negative cases may occur, in which the cancer is not visualized at MRI and is instead diagnosed with another imaging modality. The authors describe the causes of false-negative breast MRI results, which can be categorized broadly as secondary to perceptual errors or cognitive errors, or nonvisualization secondary to nonenhancement of the tumor. Tips and strategies to avoid these errors are discussed. Perceptual errors occur when an abnormality is not prospectively identified, yet the examination is technically adequate. Careful development of thorough search patterns is critical to avoid these errors. Cognitive errors occur when an abnormality is identified but misinterpreted or mischaracterized as benign. The radiologist may avoid these errors by utilizing all available prior examinations for comparison, viewing images in all planes to better assess the margins and shapes of abnormalities, and appropriately integrating all available information from the contrast-enhanced, T2-weighted, and T1-weighted images as well as the clinical history. Despite this, false-negative cases are inevitable, as certain subtypes of breast cancer, including ductal carcinoma in situ, invasive lobular carcinoma, and certain well-differentiated invasive cancers, may demonstrate little to no enhancement at MRI, owing to differences in angiogenesis and neovascularity. MRI is a valuable diagnostic tool in breast imaging. However, MRI should continue to be used as a complementary modality, with mammography and US, in the detection of breast cancer. ^©RSNA, 2021.