Adaptive Machine Learning Approach for Importance Evaluation of Multimodal Breast Cancer Radiomic Features.

Breast cancer holds the highest diagnosis rate among female tumors and is the leading cause of death among women. Quantitative analysis of radiological images shows the potential to address several medical challenges, including the early detection and classification of breast tumors. In the P.I.N.K study, 66 women were enrolled. Their paired Automated Breast Volume Scanner (ABVS) and Digital Breast Tomosynthesis (DBT) images, annotated with cancerous lesions, populated the first ABVS+DBT dataset. This enabled not only a radiomic analysis for the malignant vs. benign breast cancer classification, but also the comparison of the two modalities. For this purpose, the models were trained using a leave-one-out nested cross-validation strategy combined with a proper threshold selection approach. This approach provides statistically significant results even with medium-sized data sets. Additionally it provides distributional variables of importance, thus identifying the most informative radiomic features. The analysis proved the predictive capacity of radiomic models even using a reduced number of features. Indeed, from tomography we achieved AUC-ROC 89.9 % using 19 features and 92.1 % using 7 of them; while from ABVS we attained an AUC-ROC of 72.3 % using 22 features and 85.8 % using only 3 features. Although the predictive power of DBT outperforms ABVS, when comparing the predictions at the patient level, only 8.7% of lesions are misclassified by both methods, suggesting a partial complementarity. Notably, promising results (AUC-ROC ABVS-DBT 71.8 % - 74.1 % ) were achieved using non-geometric features, thus opening the way to the integration of virtual biopsy in medical routine.

The P.I.N.K. Study Approach for Supporting Personalized Risk Assessment and Early Diagnosis of Breast Cancer.

Breast cancer is a clear example of excellent survival when it is detected and properly treated in the early stage. Currently, screening of this cancer relies on mammography, which may be integrated by new imaging techniques for more exhaustive evaluation. The Personalized, Integrated, Network, Knowledge (P.I.N.K.) study is a longitudinal multicentric study involving several diagnostic centres across Italy, co-ordinated by the Italian National Research Council and co-funded by the Umberto Veronesi Foundation. Aim of the study is to evaluate the increased diagnostic accuracy in detecting cancers obtained with different combinations of imaging technologies, and find the most effective diagnostic pathway matching the characteristics of an individual patient. The study foresees the enrolment of 50,000 women over the age of 40 years presenting for breast examination and providing informed consent to data handling. So far, the 15 participating centres across Italy have recruited a total of 22,848 patients. Based on the analyses of the first 175 histopathological-proven breast cancers, mammographic sensitivity was estimated to be 61.7% (n = 108 cancers), whereas diagnostic accuracy increased by 35.5% (n = 44 cancers) when mammography was integrated with other imaging modalities (ultrasound and/or digital breast tomosynthesis). Increase was mainly determined by ultrasound alone. Given the ongoing data collection and recruitment, the number of cancers detected is too low to allow any further in-depth analysis to explore links to patient characteristics. Past studies show that the uniform approach of population screening guidelines should be revised in favour of more personalised regimens, where known standards are integrated by imaging techniques most suitable for the individual's characteristics. With the ultimate goal of identifying early breast cancer detection strategies, our preliminary results suggest that integrated diagnostic approach could lead to a paradigm shift from an age-based regimen toward more specific and effective risk-based personalised screening regimens, in order to reduce mortality from breast cancer.