Reassessment of breast cancers missed during routine screening mammography: a community-based study

Contrast-enhanced mammography in the management of breast architectural distortions and avoidance of unnecessary biopsies

BACKGROUND

To assess contrast-enhanced mammography (CEM) in the management of BI-RADS3 breast architectural distortions (AD) in digital breast tomosynthesis (DBT).

METHODS

We retrospectively reviewed 328 women with 332 ADs detected on DBT between 2017 and 2021 and selected those classified as BI-RADS3 receiving CEM as problem-solving. In CEM recombined images, we evaluated AD's contrast enhancement (CE) according to its presence/absence, type, and size. AD with enhancement underwent imaging-guided biopsy while AD without enhancement follow-up or biopsy if detected in high/intermediate-risk women.

RESULTS

AD with enhancement were 174 (52.4%): 72 (41.4%) were malignant lesions, 102 (59.6%) false positive results: 28 (16%) B3 lesions, and 74 (42.5%) benign lesions. AD without enhancement were 158 (47.6%): 26 (16.5%) were subjected to biopsy (1 malignant and 25 benign) while the other 132 cases were sent to imaging follow-up, still negative after two years. CEM's sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and accuracy were 98.63%, 60.62%, 41.38%, 99.37%, and 68.98%. The AUC determined by ROC was 0.796 (95% CI, 0.749-0.844).

CONCLUSION

CEM has high sensitivity and NPV in evaluating BI-RADS3 AD and can be a complementary tool in assessing AD, avoiding unnecessary biopsies without compromising cancer detection.

The Correlation between Morpho-Dynamic Contrast-Enhanced Mammography (CEM) Features and Prognostic Factors in Breast Cancer: A Single-Center Retrospective Analysis

Breast cancer, a major contributor to female mortality globally, presents challenges in detection, prompting exploration beyond digital mammography. Contrast-Enhanced Mammography (CEM), integrating morphological and functional information, emerges as a promising alternative, offering advantages in cost-effectiveness and reduced anxiety compared to MRI. This study investigates CEM's correlation with breast cancer prognostic factors, encompassing histology, grade, and molecular markers. In a retrospective analysis involving 114 women, CEM revealed diverse lesion characteristics. Statistical analyses identified correlations between specific CEM features, such as spiculated margins and irregular shape, and prognostic factors like tumor grade and molecular markers. Notably, spiculated margins predicted lower grade and HER2 status, while irregular shape correlated with PgR and Ki-67 status. The study emphasizes CEM's potential in predicting breast cancer prognosis, shedding light on tumor behavior. Despite the limitations, including sample size and single-observer analysis, the findings advocate for CEM's role in stratifying breast cancers based on biological characteristics. CEM features, particularly spiculated margins, irregular shape, and enhancement dynamics, may serve as valuable indicators for personalized treatment decisions. Further research is crucial to validate these correlations and enhance CEM's clinical utility in breast cancer assessment.

Identification and diagnosis of mammographic malignant architectural distortion using a deep learning based mask regional convolutional neural network

Background

Architectural distortion (AD) is a common imaging manifestation of breast cancer, but is also seen in benign lesions. This study aimed to construct deep learning models using mask regional convolutional neural network (Mask-RCNN) for AD identification in full-field digital mammography (FFDM) and evaluate the performance of models for malignant AD diagnosis.

Methods

This retrospective diagnostic study was conducted at the Second Affiliated Hospital of Guangzhou University of Chinese Medicine between January 2011 and December 2020. Patients with AD in the breast in FFDM were included. Machine learning models for AD identification were developed using the Mask RCNN method. Receiver operating characteristics (ROC) curves, their areas under the curve (AUCs), and recall/sensitivity were used to evaluate the models. Models with the highest AUCs were selected for malignant AD diagnosis.

Results

A total of 349 AD patients (190 with malignant AD) were enrolled. EfficientNetV2, EfficientNetV1, ResNext, and ResNet were developed for AD identification, with AUCs of 0.89, 0.87, 0.81 and 0.79. The AUC of EfficientNetV2 was significantly higher than EfficientNetV1 (0.89 vs. 0.78, P=0.001) for malignant AD diagnosis, and the recall/sensitivity of the EfficientNetV2 model was 0.93.

Conclusion

The Mask-RCNN-based EfficientNetV2 model has a good diagnostic value for malignant AD.

Breast Cancer in Asia: Incidence, Mortality, Early Detection, Mammography Programs, and Risk-Based Screening Initiatives

Simple Summary

Nearly all breast cancer patients survive for more than five years when the tumor is found early and in the localized stage. Regular clinical breast examinations, mammograms, and monthly self-exams of the breasts all contribute to early detection. However, late-stage breast cancers are common in many Asian countries. Low-income countries suffer from a lack of resources for breast cancer screening. High-income countries, on the other hand, are not benefiting fully from national breast screening programs due to an underutilization of the preventive healthcare services available. Existing reviews on Asian breast cancers are heavily focused on risk factors. The question of whether we should adopt or adapt the knowledge generated from non-Asian breast cancers would benefit from an extension into screening guidelines. In addition, several Asian countries are piloting studies that move away from the age-based screening paradigm.

Abstract

Close to half (45.4%) of the 2.3 million breast cancers (BC) diagnosed in 2020 were from Asia. While the burden of breast cancer has been examined at the level of broad geographic regions, literature on more in-depth coverage of the individual countries and subregions of the Asian continent is lacking. This narrative review examines the breast cancer burden in 47 Asian countries. Breast cancer screening guidelines and risk-based screening initiatives are discussed.

Evaluation of the Combination of Artificial Intelligence and Radiologist Assessments to Interpret Malignant Architectural Distortion on Mammography

Purpose

To compare the mammographic malignant architectural distortion (AD) detection performance of radiologists who read mammographic examinations unaided versus those who read these examinations with the support of artificial intelligence (AI) systems.

Material and Methods

This retrospective case-control study was based on a double-reading of clinical mammograms between January 2011 and December 2016 at a large tertiary academic medical center. The study included 177 malignant and 90 benign architectural distortion (AD) patients. The model was built based on the ResNeXt-50 network. Algorithms used deep learning convolutional neural networks, feature classifiers, image analysis algorithms to depict AD and output a score that translated to malignant. The accuracy for malignant AD detection was evaluated using area under the curve (AUC).

Results

The overall AUC was 0.733 (95% CI, 0.673-0.792) for Reader First-1, 0.652 (95% CI, 0.586-0.717) for Reader First-2, and 0.655 (95% CI, 0.590-0.719) for Reader First-3. and the overall AUCs for Reader Second-1, 2, 3 were 0.875 (95% CI, 0.830-0.919), 0.882 (95% CI, 0.839-0.926), 0.884 (95% CI, 0.841-0.927),respectively. The AUCs for all the reader-second radiologists were significantly higher than those for all the reader-first radiologists (Reader First-1 vs. Reader Second-1, P= 0.004). The overall AUC was 0.792 (95% CI, 0.660-0.925) for AI algorithms. The combination assessment of AI algorithms and Reader First-1 achieved an AUC of 0.880 (95% CI, 0.793-0.968), increased than the Reader First-1 alone and AI algorithms alone. AI algorithms alone achieved a specificity of 61.1% and a sensitivity of 80.6%. The specificity for Reader First-1 was 55.5%, and the sensitivity was 86.1%. The results of the combined assessment of AI and Reader First-1 showed a specificity of 72.7% and sensitivity of 91.7%. The performance showed significant improvements compared with AI alone (p<0.001) as well as the reader first-1 alone (p=0.006).

Conclusion

While the single AI algorithm did not outperform radiologists, an ensemble of AI algorithms combined with junior radiologist assessments were found to improve the overall accuracy. This study underscores the potential of using machine learning methods to enhance mammography interpretation, especially in remote areas and primary hospitals.

Diet-Driven Inflammation and Insulinemia and Risk of Interval Breast Cancer

Interval breast cancers (IBCs) emerge after a non-suspicious mammogram and before the patient's next scheduled screen. Risk factors associated with IBC have not been identified. This study evaluated if the empirical dietary inflammatory pattern (EDIP) or empirical dietary index for hyperinsulinemia (EDIH) scores are associated with IBC compared to screen-detected breast cancer. Data were from women 50-79 years-old in the Women's Health Initiative cohort who completed food frequency questionnaires at baseline (1993-98) and were followed through March 31, 2019 for breast cancer detection. Women were identified as having either IBC diagnosed within 1-year after their last negative screening mammogram (N = 317) or screen-detected breast cancer (N = 1,928). Multivariable-adjusted logistic regression analyses were used to estimate odds ratios for risk of IBC compared to screen-detected cancer in dietary index tertiles. No associations were observed between EDIP or EDIH and IBC. Odds ratios comparing the highest to the lowest dietary index tertile were 1.08; 95%CI, 0.78-1.48 for EDIP and 0.92; 95%CI, 0.67-1.27 for EDIH. The null associations persisted when stratified by BMI categories. Findings suggest that diet-driven inflammation or insulinemia may not be substantially associated with IBC risk among postmenopausal women. Future studies are warranted to identify modifiable factors for IBC prevention.

SCREENING MAMMOGRAPHY: DIAGNOSTIC EFFICACY-ISSUES AND CONSIDERATIONS FOR THE 2020S

Diagnostic efficacy in medical imaging is ultimately a reflection of radiologist performance. This can be influenced by numerous factors, some of which are patient related, such as the physical size and density of the breast, and machine related, where some lesions are difficult to visualise on traditional imaging techniques. Other factors are human reader errors that occur during the diagnostic process, which relate to reader experience and their perceptual and cognitive oversights. Given the large-scale nature of breast cancer screening, even small increases in diagnostic performance equate to large numbers of women saved. It is important to identify the causes of diagnostic errors and how detection efficacy can be improved. This narrative review will therefore explore the various factors that influence mammographic performance and the potential solutions used in an attempt to ameliorate the errors made.

Breast cancer risk stratification for mammographic screening: A nation-wide screening cohort of 24,431 women in Singapore

Background

Breast cancer incidence is increasing in Asia. However, few women in Singapore attend routine mammography screening. We aim to identify women at high risk of breast cancer who will benefit most from regular screening using the Gail model and information from their first screen (recall status and mammographic density).

Methods

In 24,431 Asian women (50–69 years) who attended screening between 1994 and 1997, 117 developed breast cancer within 5 years of screening. Cox proportional hazard models were used to study the associations between risk classifiers (Gail model 5‐year absolute risk, recall status, mammographic density), and breast cancer occurrence. The efficacy of risk stratification was evaluated by considering sensitivity, specificity, and the proportion of cancers identified.

Results

Adjusting for information from first screen attenuated the hazard ratios (HR) associated with 5‐year absolute risk (continuous, unadjusted HR [95% confidence interval]: 2.3 [1.8–3.1], adjusted HR: 1.9 [1.4–2.6]), but improved the discriminatory ability of the model (unadjusted AUC: 0.615 [0.559–0.670], adjusted AUC: 0.703 [0.653–0.753]). The sensitivity and specificity of the adjusted model were 0.709 and 0.622, respectively. Thirty‐eight percent of all breast cancers were detected in 12% of the study population considered high risk (top five percentile of the Gail model 5‐year absolute risk [absolute risk ≥1.43%], were recalled, and/or mammographic density ≥50%).

Conclusion

The Gail model is able to stratify women based on their individual breast cancer risk in this population. Including information from the first screen can improve prediction in the 5 years after screening. Risk stratification has the potential to pick up more cancers.

Retrospective radiological review and classification of interval breast cancers within population-based breast screening programmes for the purposes of open disclosure: A systematic review

INTRODUCTION

Interval breast cancers occur following a negative breast screening mammogram and before the next scheduled appointment within screening programmes. Radiological review classifies them as cancers that develop between screens, cancers with no obvious malignant abnormalities on prior screens or cancers not detected at screening. This study aimed to systematically review published literature on the occurrence of open disclosure following interval cancer radiological reviews by breast screening programmes internationally in a retrospective setting and examine methodologies used for radiological reviews for the purposes of disclosure.

METHODS

A search for relevant articles published (January 2000 - May 2019) was conducted according to PICO and PRISMA guidelines. The databases Pubmed, Scopus, Google Scholar, Cinahl, Web of Science, Embase, Science Direct and Global Health were searched. Relevant studies were reviewed if they had completed a retrospective review and classification of interval breast cancers.

RESULTS

Of 46 relevant articles included, no study was identified that conducted a retrospective review purposely for open disclosure. Retrospective reviews were conducted for audit/quality assurance, and research including for radiologist education and learning. Variation in methodology was found across review type (non-blinded/semi-informed approach), number of reviewers and classification categories. The proportion of false negative cancers classified among the studies ranged from 4 to 40 %.

DISCUSSION

Variation among radiological review practices were observed, which likely impacts classification results. To ensure standardised classification of interval breast cancers are employed for the purposes of open disclosure in screening settings, reproducible and consistent methodology is required.

Is axillary imaging for invasive lobular carcinoma accurate in determining clinical node staging?

PURPOSE

Preoperative evaluation of clinical N-stage (cN) is difficult in breast cancer patients with invasive lobular carcinoma (ILC). Our goal was to assess the predictive value of axillary imaging in ILC by comparing imaging cN and pathologic N-stage (pN).

METHODS

A single-institution retrospective review was performed for newly diagnosed stage I-III ILC patients undergoing preoperative breast imaging from 2011 to 2016. Clinicopathologic factors; mammogram, MRI, and ultrasound findings; and surgical pathology data were reviewed. Sub-analysis for pN2-N3 patients was performed to determine imaging sensitivity for patients with a larger nodal disease burden. Statistical analysis included sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of each imaging modality.

RESULTS

Of the total 349 patients included, 70.5% were cN0, and 62% were pN0 (p = 0.03). For all patients, mammogram sensitivity was 7%, specificity 97%, PPV 50%, NPV 72%; ultrasound sensitivity was 26%, specificity 86%, PPV 52%, NPV 67%; MRI sensitivity was 7%, specificity 98%, PPV 80%, NPV 51%. For pN2/N3 patients, 38% were identified as cN0. Mammogram sensitivity was 10%; ultrasound 42%; MRI 65%. Pathology evaluation of N2/N3 patients indicated LN were replaced with ILC but maintained normal architecture. The average largest pathologic tumor deposit (1.5 ± 0.8 cm) correlated with average largest imaging LN size (1.4 ± 0.6 cm) (p = 0.58).

CONCLUSION

A statistically significant difference between clinical and pathologic N-stage exists for ILC patients. MRI was most sensitive for identification of pN2-N3 patients and should be considered part of routine axillary imaging evaluation for ILC patients.

Satisfaction of Search in Periapical Radiograph Interpretation

INTRODUCTION

Several studies in radiology and medicine have evaluated the satisfaction of search (SOS) error effect in chest radiography, abdominal radiography, osteoradiology, and patients with multiple trauma. No research to date has been published evaluating the possible existence of the SOS error phenomenon made during dental periapical radiograph interpretations. The purpose of the present pilot study was to determine if an SOS error effect exists when dental clinicians interpret periapical radiographs. The null hypothesis was that the detection accuracy will be the same or will improve for the detection of native lesions in the presence of an added abnormality. The alternative hypothesis is that there will be a decrease in detection accuracy for native lesions in the presence of an added abnormality.

METHODS

Six images were selected to be part of the present experiment. One of the 6 images served as the positive control, and another image served as the negative control. Four images, each including a single subtle carious lesion, were selected to represent the experimental images. The single subtle carious lesion present within the 4 experimental radiographs served as the native pathology, and an abnormality such as a periapical radiolucency, resorption, inadequate nonideal root canal obturation material, or recurrent carious lesion was artificially inserted into the image as the added pathology. Thus, the second set of images consisted of the same 4 images containing the native pathology including an added pathology that was inserted into the image using Adobe Photoshop CS6 (Adobe, Inc, San Jose, CA). Purposive sampling was obtained from 16 examiners including residents from endodontics and periodontics as well as alumni and faculty from the Saint Louis University Center for Advanced Dental Education, St Louis, MO. Each observer participated as a subject during 2 time-separated sessions. Each session was separated by a minimum period of 3 months' duration in order to prevent memory bias. Before starting each interpretation session, the participants were given verbal instructions. Subjects were instructed to provide a location (by tooth number), identify, and rate the presence of all suspected pathology using a Likert scale of 1-5 (1: definitely normal, 2: probably normal, 3: possibly abnormal, 4: probably abnormal, and 5: definitely abnormal). In the second session, the radiographs that were initially presented containing only the native lesion were presented again with the added abnormality, and vice versa. The observers' reports and confidence ratings were recorded and analyzed. Ratings of 3-5 were considered as being positive for the presence of pathology.

RESULTS

A true SOS error occurs when the presence of the native lesion is reported correctly without an added abnormality but is not reported (missed) in the presence of an added abnormality. In our study, a true SOS error occurred in 13 of the 64 interpretation sets (20.31%). There was a total of 64 expected native lesions present within the 4 native images viewed by 16 observers. In the 4 added images, there was a total of 64 expected added findings. In the images containing only native lesions, the observers reported 30 of the 64 expected native lesions. In the images containing an artificially added abnormality, the observers reported 58 of the 64 expected added abnormalities and 25 of the 64 expected native lesions. Observers reported fewer native lesions in the presence of an added abnormality.

CONCLUSIONS

The current investigation demonstrated the existence of the SOS effect during periapical radiographic interpretations. In 20.31% of interpretations, a true SOS error occurred. This study is clinically relevant because it can help clinicians in reducing false-negative errors made during radiographic interpretation, thus preventing misdiagnosis.

Observational Study to Evaluate the Clinical Efficacy of Thermalytix for Detecting Breast Cancer in Symptomatic and Asymptomatic Women

PURPOSE

To evaluate the sensitivity and specificity of Thermalytix, an artificial intelligence-based computer-aided diagnostics (CADx) engine, to detect breast malignancy by comparing the CADx output with the final diagnosis derived using standard screening modalities.

METHODS

This multisite observational study included 470 symptomatic and asymptomatic women who presented for a breast health checkup in two centers. Among them, 238 women had symptoms such as breast lump, nipple discharge, or breast pain, and the rest were asymptomatic. All participants underwent a Thermalytix test and one or more standard-of-care tests for breast cancer screening, as recommended by the radiologists. Results from Thermalytix and standard modalities were obtained independently in a blinded fashion for comparison. The ground truth used for analysis (normal or malignant) was the final impression of an expert clinician based on the symptoms and the available reports of standard modalities (mammography, ultrasonography, elastography, biopsy, fine-needle aspiration cytology, and so on).

RESULTS

For the 470 women, Thermalytix resulted in a sensitivity of 91.02% (symptomatic, 89.85%; asymptomatic, 100%) and specificity of 82.39% (symptomatic, 69.04%; asymptomatic, 92.41%) in detection of breast malignancy. Thermalytix showed an overall area under the curve (AUC) of 0.90, with an AUC of 0.82 for symptomatic and 0.98 for asymptomatic women.

CONCLUSION

High sensitivity and high AUC of Thermalytix in women of all age groups demonstrates the efficacy of the tool for breast cancer screening. Thermalytix, with its automated scoring and image annotations of potential malignancies and vascularity, can assist the clinician in better decision making and improve quality of care in an affordable and radiation-free manner. Thus, we believe Thermalytix is poised to be a promising modality for breast cancer screening.

The Role of Cognitive Bias in Breast Radiology Diagnostic and Judgment Errors

Cognitive bias is an unavoidable aspect of human decision-making. In breast radiology, these biases contribute to missed or erroneous diagnoses and mistaken judgments. This article introduces breast radiologists to eight cognitive biases commonly encountered in breast radiology: anchoring, availability, commission, confirmation, gambler's fallacy, omission, satisfaction of search, and outcome. In addition to illustrative cases, this article offers suggestions for radiologists to better recognize and counteract these biases at the individual level and at the organizational level.

Missed Breast Cancer: Effects of Subconscious Bias and Lesion Characteristics

Medical errors are a substantial cause of morbidity and mortality and the third leading cause of death in the United States. Errors resulting in missed breast cancer are the most common reason for medical malpractice lawsuits against all physicians. Missed breast cancers are breast malignancies that are detectable at retrospective review of a previously obtained mammogram that was prospectively reported as showing negative, benign, or probably benign findings. Investigators in prior studies have found that up to 35% of both interval cancers and screen-detected cancers could be classified as missed. As such, in conjunction with having awareness of the most common misleading appearances of breast cancer, it is important to understand the cognitive processes and unconscious biases that can impact image interpretation, thereby helping to decrease the number of missed breast cancers. The various cognitive processes that lead to unconscious bias in breast imaging, such as satisfaction of search, inattention blindness, hindsight, anchoring, premature closing, and satisfaction of reporting, are outlined in this pictorial review of missed breast cancers. In addition, strategies for reducing the rates of these missed cancers are highlighted. The most commonly missed and misinterpreted lesions, including stable lesions, benign-appearing masses, one-view findings, developing asymmetries, subtle calcifications, and architectural distortion, also are reviewed. This information will help illustrate why and how breast cancers are missed and aid in the development of appropriate minimization strategies in breast imaging. ^©RSNA, 2020.

Breast Cancer Screening Program in Lithuania: Interval Cancers and Program Sensitivity After 7 Years of Mammography Screening

OBJECTIVE

Analysis of interval cancers is critical in determining the sensitivity of screening and represents an objective measure of the quality of mammography screening program (MSP).

METHODS

Period analyzed: from 2006 to 2012. The rate of screen-detected, interval cancers and program sensitivity were measured. A comparison of screen-detected and interval cancers was performed.

RESULTS

During the period of the study, 429 473 women were screened and 1297 were found to have cancer. The overall screen-detected cancer rate was 30.2 per 10 000 women screened. Four hundred thirty-one case of interval cancers have occurred during the period of the study. The interval cancer ratio (ICR) was 0.25. Overall sensitivity of MSP amounted to 75.1%. Slightly lower sensitivity was found among the youngest age-group, especially for those with lobular cancers. Interval cancers were bigger in size, more often with metastases in lymph nodes, than screen-detected cancers, but these differences were not statistically significant.

CONCLUSIONS

Overall program sensitivity in Lithuania is about 75%, ICR is 0.25, and these parameters are comparable to other European countries.

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.

A computer-aided detection of the architectural distortion in digital mammograms using the fractal dimension measurements of BEMD

Achieving a high performance for the detection and characterization of architectural distortion in screening mammograms is important for an efficient breast cancer early detection. Viewing a mammogram image as a rough surface that can be described using the fractal theory is a well-recognized approach. This paper presents a new fractal-based computer-aided detection (CAD) algorithm for characterizing various breast tissues in screening mammograms with a particular focus on distinguishing between architectural distortion and normal breast parenchyma. The proposed approach is based on two underlying assumptions: (i) monitoring the variation pattern of fractal dimension, with the changes of the image resolution, is a useful tool to distinguish textural patterns of breast tissue, (ii) the bidimensional empirical mode decomposition (BEMD) algorithm appropriately generates a multiresolution representation of the mammogram. The proposed CAD has been tested using different validation datasets of mammographic regions of interest (ROIs) extracted from the Digital Database for Screening Mammography (DDSM) database. The validation ROI datasets contain architectural distortion (AD), normal breast tissue, and AD surrounding tissue. The highest classification performance, in terms of area under the receiver operating characteristic curve, of Az = 0.95 was achieved when the proposed approach applied to distinguish 187 architectural distortion depicting regions from 2191 normal breast parenchyma regions. The obtained results validate the underlying hypothesis and demonstrate that effectiveness of capturing the variation of the fractal dimension measurements within an appropriate multiscale representation of the digital mammogram. Results also reveal that this tool has the potential of prescreening other key and common mammographic signs of early breast cancer.

Recent Advances in Breast-Specific Imaging

Background

Imaging the breast is a vital component not only for breast cancer screening, but also for diagnosis, evaluation, treatment, and follow-up of patients with breast cancer.

Methods

The author reviews recent advances and also provides her personal experience in describing the status of digital mammography, computer-aided detection, dedicated magnetic resonance imaging (MRI), and positron-emission mammography for evaluating the breast.

Results

Full-field digital mammography is superior to standard mammography in women under 50 years of age and in those with dense breasts. Computer-aided detection assists inexperienced mammographers and enhances detection of microcalcifications in dense breasts. Breast MRI is useful in preoperative evaluation, clarification of indeterminate mammograms, and follow-up of BRCA mutation carriers. The specificity of MRI remains problematic, however. Positron-emission mammography promises enhanced detection of ductal carcinoma in situ (DCIS), even when not associated with microcalcifications, and should aid surgical planning.

Conclusions

These four significant advances in breast imaging have all improved the sensitivity of detecting breast abnormalities. Cost issues, however, may limit the widespread application of these advances.

Biologic Profiles of Invasive Breast Cancers Detected Only With Digital Breast Tomosynthesis

OBJECTIVE

The purpose of this study was to analyze the clinicopathologic and immunohistochemical features of invasive breast cancers detected only with digital breast tomosynthesis (DBT), compared with those of cancers detected with both DBT and full-field digital mammography (FFDM).

MATERIALS AND METHODS

The medical records of 261 women (108 without and 153 with symptoms) with invasive breast cancers who underwent FFDM and DBT between April 2015 and June 2016 were retrospectively reviewed. To assess detectability, all DBT and FFDM images were reviewed independently by three radiologists blinded to clinicopathologic information. The reference standard was established by an unblinded consensus review of all images. Clinicopathologic and immunohistochemical features were analyzed according to the detectability status.

RESULTS

Of the 261 cancers, 223 (85.4%) were detected with both DBT and FFDM (both-detected group). Twenty-four cancers (9.2%) not detected with FFDM (DBT-only group) were classified by DBT as a mass (58.3%), architectural distortion (33.3%), or asymmetry (8.3%). The remaining 14 cancers (5.4%) were not detected with either DBT or FFDM (both-occult group). On multivariate analysis, a dense breast parenchyma (p = 0.007), small tumor size (≤ 2 cm; p = 0.027), and luminal A-like subtype (estrogen receptor positive or progesterone receptor positive or both, human epidermal growth factor receptor 2 negative, and Ki-67 expression < 14%; p = 0.008) were significantly associated with the DBT-only group. For 108 screening-detected cancers, a dense breast parenchyma (p = 0.007) and luminal A-like subtype (p = 0.008) also maintained significance.

CONCLUSION

The addition of DBT to FFDM in screening would aid in the detection of less-aggressive subtypes of invasive breast cancers in women with dense breasts.

Fractal analysis of visual search activity for mass detection during mammographic screening

PURPOSE

The objective of this study was to assess the complexity of human visual search activity during mammographic screening using fractal analysis and to investigate its relationship with case and reader characteristics.

METHODS

The study was performed for the task of mammographic screening with simultaneous viewing of four coordinated breast views as typically done in clinical practice. Eye-tracking data and diagnostic decisions collected for 100 mammographic cases (25 normal, 25 benign, 50 malignant) from 10 readers (three board certified radiologists and seven Radiology residents), formed the corpus for this study. The fractal dimension of the readers' visual scanning pattern was computed with the Minkowski-Bouligand box-counting method and used as a measure of gaze complexity. Individual factor and group-based interaction ANOVA analysis was performed to study the association between fractal dimension, case pathology, breast density, and reader experience level. The consistency of the observed trends depending on gaze data representation was also examined.

RESULTS

Case pathology, breast density, reader experience level, and individual reader differences are all independent predictors of the complexity of visual scanning pattern when screening for breast cancer. No higher order effects were found to be significant.

CONCLUSIONS

Fractal characterization of visual search behavior during mammographic screening is dependent on case properties and image reader characteristics.

Missed Breast Cancer: What Can We Learn?

Multiple studies have shown that screening mammography helps to reduce mortality and morbidity from advanced breast cancer. However mammography does have its own limitations, and unfortunately, there are a fair number of false-negative mammograms. We are all aware that the sensitivity of mammography is inversely proportional to the breast density. With many states passing mandatory breast density reporting legislation, there has been an emphasis on using additional and alternative screening methods such as whole breast ultrasound and screening magnetic resonance imaging. Many cancers are simply not detected on mammography, even in retrospect. However, many of the breast cancers are actually visible retrospectively on the prior mammogram. It is these small and often subtle cancers that are perceptible but often missed, that provide a valuable learning opportunity. Studying the imaging findings of cancers that went undetected is a good learning exercise for the radiologist to identify common patterns and mistakes that lead to a missed cancer. This allows the radiologist to improve mammographic sensitivity and overall diagnostic accuracy. This article discusses some of the limitations of mammography, common sources of error which may lead to an undetected cancer, and also discuss a few pearls to prevent these common errors.

In the digital era, architectural distortion remains a challenging radiological task

AIM

To compare readers' performance in detecting architectural distortion (AD) compared with other breast cancer types using digital mammography.

MATERIALS AND METHODS

Forty-one experienced breast screen readers (20 US and 21 Australian) were asked to read a single test set of 30 digitally acquired mammographic cases. Twenty cases had abnormal findings (10 with AD, 10 non-AD) and 10 cases were normal. Each reader was asked to locate and rate any abnormalities. Lesion and case-based performance was assessed. For each collection of readers (US; Australian; combined), jackknife free-response receiver operating characteristic (JAFROC), figure of merit (FOM), and inferred receiver operating characteristic (ROC), area under curve (Az) were calculated using JAFROC v.4.1 software. Readers' sensitivity, location sensitivity, JAFROC, FOM, ROC, Az scores were compared between cases groups using Wilcoxon's signed ranked test statistics.

RESULTS

For lesion-based analysis, significantly lower location sensitivity (p=0.001) was shown on AD cases compared with non-AD cases for all reader collections. The case-based analysis demonstrated significantly lower ROC Az values (p=0.02) for the first collection of readers, and lower sensitivity for the second collection of readers (p=0.04) and all-readers collection (p=0.008), for AD compared with non-AD cases.

CONCLUSIONS

The current work demonstrates that AD remains a challenging task for readers, even in the digital era.

Detection of mammographically occult architectural distortion on digital breast tomosynthesis screening: initial clinical experience

OBJECTIVE

Digital breast tomosynthesis (DBT) has been shown to improve the sensitivity of screening mammography. DBT may have the most potential impact in cases of subtle mammographic findings such as architectural distortion (AD). The objective of our study was to determine whether DBT provides better visualization of AD than digital mammography (DM) and whether sensitivity for cancer detection is increased by the addition of DBT as it relates to cases of mammographically occult AD.

MATERIALS AND METHODS

Retrospective review of BI-RADS category 0 reports from 9982 screening DM examinations with adjunct DBT were searched for the term "architectural distortion" and were reviewed in consensus by three radiologists. ADs were classified by whether they were seen better on DM or DBT, were seen equally well on both, or were occult on either modality. The electronic medical record was reviewed to identify additional imaging studies, biopsy results, and surgical excision pathology results.

RESULTS

Review identified 26 cases of AD, 19 (73%) of which were seen only on the DBT images. Of the remaining seven ADs, six were seen better on DBT than DM. On diagnostic workup, nine lesions were assigned to BI-RADS category 4 or 5. Surgical pathology revealed two invasive carcinomas, two ductal carcinoma in situ lesions, three radial scars, and two lesions showing atypia. The cancer detection rate of DBT in mammographically occult AD was 21% (4/19). The positive predictive value of biopsy was 44%.

CONCLUSION

DBT provides better visualization of AD than DM and identifies a subset of ADs that are occult on DM. Identification of additional ADs on DBT increases the cancer detection rate.

Investigating the link between radiologists' gaze, diagnostic decision, and image content

OBJECTIVE

To investigate machine learning for linking image content, human perception, cognition, and error in the diagnostic interpretation of mammograms.

METHODS

Gaze data and diagnostic decisions were collected from three breast imaging radiologists and three radiology residents who reviewed 20 screening mammograms while wearing a head-mounted eye-tracker. Image analysis was performed in mammographic regions that attracted radiologists' attention and in all abnormal regions. Machine learning algorithms were investigated to develop predictive models that link: (i) image content with gaze, (ii) image content and gaze with cognition, and (iii) image content, gaze, and cognition with diagnostic error. Both group-based and individualized models were explored.

RESULTS

By pooling the data from all readers, machine learning produced highly accurate predictive models linking image content, gaze, and cognition. Potential linking of those with diagnostic error was also supported to some extent. Merging readers' gaze metrics and cognitive opinions with computer-extracted image features identified 59% of the readers' diagnostic errors while confirming 97.3% of their correct diagnoses. The readers' individual perceptual and cognitive behaviors could be adequately predicted by modeling the behavior of others. However, personalized tuning was in many cases beneficial for capturing more accurately individual behavior.

CONCLUSIONS

There is clearly an interaction between radiologists' gaze, diagnostic decision, and image content which can be modeled with machine learning algorithms.

The adjunctive digital breast tomosynthesis in diagnosis of breast cancer

Purpose. To compare the diagnostic performance of digital breast tomosynthesis (DBT) and digital mammography (DM) for breast cancers. Materials and Methods. Fifty-seven female patients with pathologically proved breast cancer were enrolled. Three readers gave a subjective assessment superiority of the index lesions (mass, focal asymmetry, architectural distortion, or calcifications) and a forced BIRADS score, based on DM reading alone and with additional DBT information. The relevance between BIRADS category and index lesions of breast cancer was compared by chi-square test. Result. A total of 59 breast cancers were reviewed, including 17 (28.8%) mass lesions, 12 (20.3%) focal asymmetry/density, 6 (10.2%) architecture distortion, 23 (39.0%) calcifications, and 1 (1.7%) intracystic tumor. Combo DBT was perceived to be more informative in 58.8% mass lesions, 83.3% density, 94.4% architecture distortion, and only 11.6% calcifications. As to the forced BIRADS score, 84.4% BIRADS 0 on DM was upgraded to BIRADS 4 or 5 on DBT, whereas only 27.3% BIRADS 4A on DM was upgraded on DBT, as BIRADS 4A lesions were mostly calcifications. A significant P value (<0.001) between the BIRADS category and index lesions was noted. Conclusion. Adjunctive DBT gives exquisite information for mass lesion, focal asymmetry, and/or architecture distortion to improve the diagnostic performance in mammography.

Quantitative measures confirm the inverse relationship between lesion spiculation and detection of breast masses

OBJECTIVE

To identify specific mammographic appearances that reduce the mammographic detection of breast cancer.

MATERIALS AND METHODS

This study received institutional board review approval and all readers gave informed consent. A set of 60 mammograms each consisting of craniocaudal and mediolateral oblique projections were presented to 129 mammogram Breastscreen readers. The images consisted of 20 positive cases with single and multicentric masses in 16 and 4 cases, respectively (resulting in a total of 24 cancers), and readers were asked to identify and locate the lesions. Each lesion was then ranked according to a detectability rating (ie, the number of observers who correctly located the lesion divided by the total number of observers), and this was correlated with breast density, lesion size, and various descriptors of lesion shape and texture.

RESULTS

Negative and positive correlations between lesion detection and density (r = -0.64, P = .007) and size (r = 0.65, P = .005), respectively, were demonstrated. In terms of lesion size and shape, there were significant correlations between the probability of detection and area (r = 0.43, P = .04), perimeter (r = 0.66, P = .0004), lesion elongation (r = 0.49, P = .02), and lesion nonspiculation (r = 0.78, P < .0001).

CONCLUSIONS

The results of this study have identified specific lesion characteristics associated with shape that may contribute to reduced cancer detection. Mammographic sensitivity may be adversely affected without appropriate attention to spiculation.

New image processing technique for evaluating breast microcalcifications: a comparative study

OBJECTIVES

The purpose of this study was to evaluate a new commercial image processing technique (MicroPure; Toshiba America Medical Systems, Tustin, CA) for identifying breast microcalcifications compared to gray scale ultrasound imaging (US) using mammography as the reference standard.

METHODS

Twenty women, with breast calcifications identified mammographically, underwent gray scale US and MicroPure examinations of the breast. Still images and digital clips of the target area were acquired using gray scale US and MicroPure (at 3 different sensitivity levels: 0, 1, and 2). The images were analyzed by 4 independent and blinded readers (2 radiologists and 2 physicists) to determine the number of calcifications as well as to score image quality and artifacts.

RESULTS

For all 4 readers, there were significantly more calcifications seen with MicroPure (at the 2 highest sensitivity levels) compared to gray scale US (P < .009). Agreement between readers consistently increased from gray scale US to MicroPure imaging (gray scale intraclass correlation coefficient, 0.02-0.44; versus MicroPure intraclass correlation coefficient, 0.34-0.71). The agreement improved between mammography and MicroPure (13.2%-28.3%) when compared with mammography and gray scale US (1.7%-5.2%); the 2 radiologists saw a bigger improvement. Two readers preferred the MicroPure image quality over gray scale US (P < .001) and vice versa for the other 2 readers(P < .001). All 4 readers saw fewer artifacts with MicroPure (at level 2) than with gray scale US (P < .02).

CONCLUSIONS

MicroPure imaging identified significantly more breast microcalcifications than gray scale US.

Measures of angular spread and entropy for the detection of architectural distortion in prior mammograms

PURPOSE

Architectural distortion is an important sign of early breast cancer. We present methods for computer-aided detection of architectural distortion in mammograms acquired prior to the diagnosis of breast cancer in the interval between scheduled screening sessions.

METHODS

Potential sites of architectural distortion were detected using node maps obtained through the application of a bank of Gabor filters and linear phase portrait modeling. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases, including 301 true-positive ROIs, and from 52 mammograms of 13 normal cases. Each ROI was represented by three types of entropy measures of angular histograms composed with the Gabor magnitude response, angle, coherence, orientation strength, and the angular spread of power in the Fourier spectrum, including Shannon's entropy, Tsallis entropy for nonextensive systems, and Rényi entropy for extensive systems.

RESULTS

Using the entropy measures with stepwise logistic regression and the leave-one-patient-out method for feature selection and cross-validation, an artificial neural network resulted in an area under the receiver operating characteristic curve of 0.75. Free-response receiver operating characteristics indicated a sensitivity of 0.80 at 5.2 false positives (FPs) per patient.

CONCLUSION

The proposed methods can detect architectural distortion in prior mammograms taken 15 months (on the average) before clinical diagnosis of breast cancer, with a high sensitivity and a moderate number of FPs per patient. The results are promising and may be improved with additional features to characterize subtle abnormalities and larger databases including prior mammograms.

Malpractice claims following screening mammography in The Netherlands

Although malpractice lawsuits are frequently related to a delayed breast cancer diagnosis in symptomatic patients, information on claims at European screening mammography programs is lacking. We determined the type and frequency of malpractice claims at a Dutch breast cancer screening region. We included all 85,274 women (351,009 screens) who underwent biennial screening mammography at a southern breast screening region in The Netherlands between 1997 and 2009. Two screening radiologists reviewed the screening mammograms of all screen detected cancers and interval cancers and determined whether the cancer had been missed at the previous screen or at the latest screen, respectively. We analyzed all correspondence between the screening organization, clinicians and screened women, and collected complaints and claims until September 2011. At review, 20.9% (308/1,475) of screen detected cancers and 24.3% (163/670) of interval cancers were considered to be missed at a previous screen. A total of 19 women (of which 2, 6 and 11 women had been screened between 1997 and 2001 (102,439 screens), 2001 and 2005 (114,740 screens) and 2005 and 2009 (133,830 screens), respectively) had contacted the screening organization for additional information about their screen detected cancer or interval cancer, but filed no claim. Three other women directly initiated an insurance claim for financial compensation of their interval cancer without previously having contacted the screening organization. We conclude that screening-related claims were rarely encountered, although many screen detected cancers and interval cancers had been missed at a previous screen. A small but increasing proportion of women sought additional information about their breast cancer from the screening organization.

Computer-aided detection of architectural distortion in prior mammograms of interval cancer

Architectural distortion is an important sign of breast cancer, but because of its subtlety, it is a common cause of false-negative findings on screening mammograms. This paper presents methods for the detection of architectural distortion in mammograms of interval cancer cases taken prior to the detection of breast cancer using Gabor filters, phase portrait analysis, fractal analysis, and texture analysis. The methods were used to detect initial candidates for sites of architectural distortion in prior mammograms of interval cancer and also normal control cases. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval cancer cases, including 301 ROIs related to architectural distortion, and from 52 prior mammograms of 13 normal cases. For each ROI, the fractal dimension and Haralick's texture features were computed. Feature selection was performed separately using stepwise logistic regression and stepwise regression. The best results achieved, in terms of the area under the receiver operating characteristics curve, with the features selected by stepwise logistic regression are 0.76 with the Bayesian classifier, 0.73 with Fisher linear discriminant analysis, 0.77 with an artificial neural network based on radial basis functions, and 0.77 with a support vector machine. Analysis of the performance of the methods with free-response receiver operating characteristics indicated a sensitivity of 0.80 at 7.6 false positives per image. The methods have good potential in detecting architectural distortion in mammograms of interval cancer cases.

The mammographic correlations with Basal-like phenotype of invasive breast cancer

RATIONALE AND OBJECTIVES

Mammography contributes to the improvement of breast carcinoma survival through early detection and treatment of breast lesions. The basal-like phenotype has been found to be an independent poor prognostic factor for breast cancer. The aim of this study was to determine the mammographic correlates of the basal-like phenotype of invasive breast cancer, and to more precisely predict patient outcome and those individuals who will be responsive to a specific therapeutic regimen.

MATERIALS AND METHODS

The mammographic findings in 267 patients with operable breast cancer were correlated with the basal-like subtype identified using immunohistochemical assessment of breast cancer cases, including estrogen receptor, progesterone receptor, HER-2/neu status, cytokeratin (CK5/6), and epidermal growth factor receptor.

RESULTS

Of the 267 invasive breast cancers, 40 (15%) were of the basal-like phenotype. Basal-phenotype tumors were significantly more likely to manifest as a mass (P = .002), most of which were indistinct margin (P =.035), at mammography, and architecture distortion at mammography (P = .002).

CONCLUSION

The mammographic appearances of basal-like tumors, more mass and architecture distortion, suggest more rapid carcinogenesis. Additional studies are warranted to further refine prognosis, and to optimize treatment in patients with basal-like breast cancer.

Evaluation of clinical image processing algorithms used in digital mammography

Screening is the only proven approach to reduce the mortality of breast cancer, but significant numbers of breast cancers remain undetected even when all quality assurance guidelines are implemented. With the increasing adoption of digital mammography systems, image processing may be a key factor in the imaging chain. Although to our knowledge statistically significant effects of manufacturer-recommended image processings have not been previously demonstrated, the subjective experience of our radiologists, that the apparent image quality can vary considerably between different algorithms, motivated this study. This article addresses the impact of five such algorithms on the detection of clusters of microcalcifications. A database of unprocessed (raw) images of 200 normal digital mammograms, acquired with the Siemens Novation DR, was collected retrospectively. Realistic simulated microcalcification clusters were inserted in half of the unprocessed images. All unprocessed images were subsequently processed with five manufacturer-recommended image processing algorithms (Agfa Musica 1, IMS Raffaello Mammo 1.2, Sectra Mamea AB Sigmoid, Siemens OPVIEW v2, and Siemens OPVIEW v1). Four breast imaging radiologists were asked to locate and score the clusters in each image on a five point rating scale. The free-response data were analyzed by the jackknife free-response receiver operating characteristic (JAFROC) method and, for comparison, also with the receiver operating characteristic (ROC) method. JAFROC analysis revealed highly significant differences between the image processings (F = 8.51, p < 0.0001), suggesting that image processing strongly impacts the detectability of clusters. Siemens OPVIEW2 and Siemens OPVIEW1 yielded the highest and lowest performances, respectively. ROC analysis of the data also revealed significant differences between the processing but at lower significance (F = 3.47, p = 0.0305) than JAFROC. Both statistical analysis methods revealed that the same six pairs of modalities were significantly different, but the JAFROC confidence intervals were about 32% smaller than ROC confidence intervals. This study shows that image processing has a significant impact on the detection of microcalcifications in digital mammograms. Objective measurements, such as described here, should be used by the manufacturers to select the optimal image processing algorithm.

Characterization and classification of tumor lesions using computerized fractal-based texture analysis and support vector machines in digital mammograms

OBJECTIVE

This paper presents a detailed study of fractal-based methods for texture characterization of mammographic mass lesions and architectural distortion. The purpose of this study is to explore the use of fractal and lacunarity analysis for the characterization and classification of both tumor lesions and normal breast parenchyma in mammography.

MATERIALS AND METHODS

We conducted comparative evaluations of five popular fractal dimension estimation methods for the characterization of the texture of mass lesions and architectural distortion. We applied the concept of lacunarity to the description of the spatial distribution of the pixel intensities in mammographic images. These methods were tested with a set of 57 breast masses and 60 normal breast parenchyma (dataset1), and with another set of 19 architectural distortions and 41 normal breast parenchyma (dataset2). Support vector machines (SVM) were used as a pattern classification method for tumor classification.

RESULTS

Experimental results showed that the fractal dimension of region of interest (ROIs) depicting mass lesions and architectural distortion was statistically significantly lower than that of normal breast parenchyma for all five methods. Receiver operating characteristic (ROC) analysis showed that fractional Brownian motion (FBM) method generated the highest area under ROC curve (A ( z ) = 0.839 for dataset1, 0.828 for dataset2, respectively) among five methods for both datasets. Lacunarity analysis showed that the ROIs depicting mass lesions and architectural distortion had higher lacunarities than those of ROIs depicting normal breast parenchyma. The combination of FBM fractal dimension and lacunarity yielded the highest A ( z ) value (0.903 and 0.875, respectively) than those based on single feature alone for both given datasets. The application of the SVM improved the performance of the fractal-based features in differentiating tumor lesions from normal breast parenchyma by generating higher A ( z ) value.

CONCLUSION

FBM texture model is the most appropriate model for characterizing mammographic images due to self-affinity assumption of the method being a better approximation. Lacunarity is an effective counterpart measure of the fractal dimension in texture feature extraction in mammographic images. The classification results obtained in this work suggest that the SVM is an effective method with great potential for classification in mammographic image analysis.

Evaluating the effect of image preprocessing on an information-theoretic CAD system in mammography

RATIONALE AND OBJECTIVES

In our earlier studies, we reported an evidence-based computer-assisted decision (CAD) system for location-specific interrogation of mammograms. A content-based image retrieval framework with information theoretic (IT) similarity measures serves as the foundation for this system. Specifically, the normalized mutual information (NMI) was shown to be the most effective similarity measure for reduction of false-positive marks generated by other prescreening mass detection schemes. The objective of this work was to investigate the importance of image filtering as a possible preprocessing step in our IT-CAD system.

MATERIALS AND METHODS

Different filters were applied, each one aiming to compensate for known limitations of the NMI similarity measure. The study was based on a region-of-interest database that included true masses and false-positive regions from digitized mammograms.

RESULTS

Receiver-operating characteristics (ROC) analysis showed that IT-CAD is affected slightly by image filtering. Modest, yet statistically significant, performance gain was observed with median filtering (overall ROC area index A(z) improved from 0.78 to 0.82). However, Gabor filtering improved performance for the high-sensitivity portion of the ROC curve where a typical false-positive reduction scheme should operate (partial ROC area index (0.90)A(z) improved from 0.33 to 0.37). Fusion of IT-CAD decisions from different filtering schemes markedly improved performance (A(z) = 0.90 and (0.90)A(z) = 0.55). At 95% sensitivity, the system's specificity improved by 36.6%.

CONCLUSIONS

Additional improvement in false-positive reduction can be achieved by incorporating image filtering as a preprocessing step in our IT-CAD system.

Errors in cancer diagnosis: current understanding and future directions

PURPOSE

Errors in cancer diagnosis are likely the most harmful and expensive types of diagnostic errors. We reviewed the literature to understand the prevalence, origins, and prevention of errors in cancer diagnosis, focusing on common cancers for which early diagnosis offers clear benefit (melanoma and cancers of the breast, colon, and lung).

METHODS

We searched the Cochrane Library and PubMed from 1966 until April 2007 for publications that met our review criteria and manually searched references of key publications. Our search yielded 110 studies, of which nine were prospective studies and the remaining were retrospective studies.

RESULTS

Errors in cancer diagnosis were not uncommon in autopsy studies and were associated with significant harm and expense in malpractice claims. Literature on prevalence was scant. For each type of cancer, we classified preventable errors according to their origins in patient-physician encounters in the clinic setting, diagnostic test or procedure performance, pathologic confirmation of diagnosis, follow-up of patient or test result, or patient-related delays.

CONCLUSION

The literature reflects advanced knowledge of contributory factors and prevention for diagnostic errors related to the performance of procedures and imaging tests and emerging understanding of pathology errors. However, prospective studies are few, as are studies of diagnostic errors arising from the clinical encounter and patient follow-up. Future research should examine further the system and cognitive problems that lead to the many contributory factors we identified, and address interdisciplinary interventions to prevent errors in cancer diagnosis.

A computer aided detection system for digital mammograms based on radial basis functions and feature extraction techniques

An intelligent Computer-Aided Detection system (CAD) can be very helpful in detecting and diagnosing breast abnormalities earlier and faster than typical screening programs. In this paper, a system based on Radial Basis neural networks coupled with feature extraction techniques for detecting breast abnormalities in digital mammograms is presented. Suspicious regions are identified following a run of the trained neural network. Within this work, 322 breast images from the MIAS database are considered. Five co-occurrence matrices are constructed at different distances for each suspicious region. A number of statistical features are used to train and test the Radial Basis neural network presented. An average recognition rate of 87% was achieved. Using Receiver Operating Characteristic (ROC) analysis, the overall sensitivity of the technique measured by Az was found to be 0.91.