Novel mammographic image features differentiate between interval and screen-detected breast cancer: a case-case study

Explainable Multimedia Feature Fusion for Medical Applications

Due to the exponential growth of medical information in the form of, e.g., text, images, Electrocardiograms (ECGs), X-rays, and multimedia, the management of a patient’s data has become a huge challenge. In particular, the extraction of features from various different formats and their representation in a homogeneous way are areas of interest in medical applications. Multimedia Information Retrieval (MMIR) frameworks, like the Generic Multimedia Analysis Framework (GMAF), can contribute to solving this problem, when adapted to special requirements and modalities of medical applications. In this paper, we demonstrate how typical multimedia processing techniques can be extended and adapted to medical applications and how these applications benefit from employing a Multimedia Feature Graph (MMFG) and specialized, efficient indexing structures in the form of Graph Codes. These Graph Codes are transformed to feature relevant Graph Codes by employing a modified Term Frequency Inverse Document Frequency (TFIDF) algorithm, which further supports value ranges and Boolean operations required in the medical context. On this basis, various metrics for the calculation of similarity, recommendations, and automated inferencing and reasoning can be applied supporting the field of diagnostics. Finally, the presentation of these new facilities in the form of explainability is introduced and demonstrated. Thus, in this paper, we show how Graph Codes contribute new querying options for diagnosis and how Explainable Graph Codes can help to readily understand medical multimedia formats.

Mammography features for early markers of aggressive breast cancer subtypes and tumor characteristics: A population-based cohort study

Current breast cancer risk models identify mostly less aggressive tumors, although only women developing fatal breast cancer will greatly benefit from early identification. Here, we evaluated the use of mammography features (microcalcification clusters, computer-generated Breast Imaging Reporting and Data System [cBIRADS] density and lack of breast density reduction) as early markers of aggressive subtypes and tumor characteristics. Mammograms were retrieved from a population-based cohort of women that were diagnosed with breast cancer from 2001 to 2008 in Stockholm-Gotland County, Sweden. Tumor and patient characteristics were obtained from Stockholm Breast Cancer Quality Register and the Swedish Cancer Registry. Multinomial logistic regression was used to individually model each mammographic feature as a function of molecular subtypes, tumor characteristics and detection mode. A total of 4546 women with invasive breast cancer were included in the study. Women with microcalcification clusters in the affected breast were more likely to have human epidermal growth factor receptor 2 subtype (odds ratio [OR] 1.78; 95% confidence interval [CI] 1.24-2.54) and potentially less likely to have basal subtype (OR 0.54; 0.30-0.96) compared to Luminal A subtype. High mammographic cBIRADS showed association with larger tumor size and interval vs screen-detected cancers. Lack of density reduction was associated with interval vs screen-detected cancers (OR 1.43; 1.11-1.83) and potentially of Luminal B subtype vs Luminal A subtype (OR 1.76; 1.04-2.99). In conclusion, microcalcification clusters, cBIRADS density and lack of breast density reduction could serve as early markers of particular subtypes and tumor characteristics of breast cancer. This information has the potential to be integrated into risk models to identify women at risk for developing aggressive breast cancer in need of supplemental screening.

Deep learning networks find unique mammographic differences in previous negative mammograms between interval and screen-detected cancers: a case-case study

Background

To determine if mammographic features from deep learning networks can be applied in breast cancer to identify groups at interval invasive cancer risk due to masking beyond using traditional breast density measures.

Methods

Full-field digital screening mammograms acquired in our clinics between 2006 and 2015 were reviewed. Transfer learning of a deep learning network with weights initialized from ImageNet was performed to classify mammograms that were followed by an invasive interval or screen-detected cancer within 12 months of the mammogram. Hyperparameter optimization was performed and the network was visualized through saliency maps. Prediction loss and accuracy were calculated using this deep learning network. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values were generated with the outcome of interval cancer using the deep learning network and compared to predictions from conditional logistic regression with errors quantified through contingency tables.

Results

Pre-cancer mammograms of 182 interval and 173 screen-detected cancers were split into training/test cases at an 80/20 ratio. Using Breast Imaging-Reporting and Data System (BI-RADS) density alone, the ability to correctly classify interval cancers was moderate (AUC = 0.65). The optimized deep learning model achieved an AUC of 0.82. Contingency table analysis showed the network was correctly classifying 75.2% of the mammograms and that incorrect classifications were slightly more common for the interval cancer mammograms. Saliency maps of each cancer case found that local information could highly drive classification of cases more than global image information.

Conclusions

Pre-cancerous mammograms contain imaging information beyond breast density that can be identified with deep learning networks to predict the probability of breast cancer detection.

A review of the influence of mammographic density on breast cancer clinical and pathological phenotype

PURPOSE

It is well established that high mammographic density (MD), when adjusted for age and body mass index, is one of the strongest known risk factors for breast cancer (BC), and also associates with higher incidence of interval cancers in screening due to the masking of early mammographic abnormalities. Increasing research is being undertaken to determine the underlying histological and biochemical determinants of MD and their consequences for BC pathogenesis, anticipating that improved mechanistic insights may lead to novel preventative or treatment interventions. At the same time, technological advances in digital and contrast mammography are such that the validity of well-established relationships needs to be re-examined in this context.

METHODS

With attention to old versus new technologies, we conducted a literature review to summarise the relationships between clinicopathologic features of BC and the density of the surrounding breast tissue on mammography, including the associations with BC biological features inclusive of subtype, and implications for the clinical disease course encompassing relapse, progression, treatment response and survival.

RESULTS AND CONCLUSIONS

There is reasonable evidence to support positive relationships between high MD (HMD) and tumour size, lymph node positivity and local relapse in the absence of radiotherapy, but not between HMD and LVI, regional relapse or distant metastasis. Conflicting data exist for associations of HMD with tumour location, grade, intrinsic subtype, receptor status, second primary incidence and survival, which need further confirmatory studies. We did not identify any relationships that did not hold up when data involving newer imaging techniques were employed in analysis.

Derived mammographic masking measures based on simulated lesions predict the risk of interval cancer after controlling for known risk factors: a case-case analysis

PURPOSE

Women with radiographically dense or texturally complex breasts are at increased risk for interval cancer, defined as cancers diagnosed after a normal screening examination. The purpose of this study was to create masking measures and apply them to identify interval risk in a population of women who experienced either screen-detected or interval cancers after controlling for breast density.

METHODS

We examined full-field digital screening mammograms acquired from 2006 to 2015. Examinations associated with 182 interval cancers were matched to 173 screen-detected cancers on age, race, exam date and time since last imaging examination. Local Image Quality Factor (IQF) values were calculated and used to create IQF maps that represented mammographic masking. We used various statistics to define global masking measures of these maps. Association of these masking measures with interval cancer vs screen-detected cancer was estimated using conditional logistic regression in a univariate and adjusted model for Breast Imaging-Reporting and Data System (BI-RADS) density. Receiver operator curves were calculated in each case to compare specificity vs sensitivity, and area under those curves were generated. Proportion of screen-detected cancer was estimated for stratifications of IQF features.

RESULTS

Several masking features showed significant association with interval compared to screen-detected cancers after adjusting for BI-RADS density (up to P = 2.52E-6), and the 10th percentile of the IQF value (P = 1.72E-3) showed the strongest improvement in the area under the receiver operator curve, increasing from 0.65 using only BI-RADS density to 0.69. The highest masking group had a 32% proportion of screen-detected cancers while the low masking group had a 69% proportion.

CONCLUSIONS

We conclude that computer vision methods using model observers may improve quantifying the probability of breast cancer detection beyond using breast density alone.

E-Science technologies in a workflow for personalized medicine using cancer screening as a case study

Objective

We provide an e-Science perspective on the workflow from risk factor discovery and classification of disease to evaluation of personalized intervention programs. As case studies, we use personalized prostate and breast cancer screenings.

Materials and Methods

We describe an e-Science initiative in Sweden, e-Science for Cancer Prevention and Control (eCPC), which supports biomarker discovery and offers decision support for personalized intervention strategies. The generic eCPC contribution is a workflow with 4 nodes applied iteratively, and the concept of e-Science signifies systematic use of tools from the mathematical, statistical, data, and computer sciences.

Results

The eCPC workflow is illustrated through 2 case studies. For prostate cancer, an in-house personalized screening tool, the Stockholm-3 model (S3M), is presented as an alternative to prostate-specific antigen testing alone. S3M is evaluated in a trial setting and plans for rollout in the population are discussed. For breast cancer, new biomarkers based on breast density and molecular profiles are developed and the US multicenter Women Informed to Screen Depending on Measures (WISDOM) trial is referred to for evaluation. While current eCPC data management uses a traditional data warehouse model, we discuss eCPC-developed features of a coherent data integration platform.

Discussion and Conclusion

E-Science tools are a key part of an evidence-based process for personalized medicine. This paper provides a structured workflow from data and models to evaluation of new personalized intervention strategies. The importance of multidisciplinary collaboration is emphasized. Importantly, the generic concepts of the suggested eCPC workflow are transferrable to other disease domains, although each disease will require tailored solutions.