Using multiscale texture and density features for near-term breast cancer risk analysis

Abstract

Purpose: To help improve efficacy of screening mammography by eventually establishing a new optimal personalized screening paradigm, the authors investigated the potential of using the quantitative multiscale texture and density feature analysis of digital mammograms to predict near-term breast cancer risk. Methods: The authors' dataset includes digital mammograms acquired from 340 women. Among them, 141 were positive and 199 were negative/benign cases. The negative digital mammograms acquired from the "prior" screening examinations were used in the study. Based on the intensity value distributions, five subregions at different scales were extracted from each mammogram. Five groups of features, including density and texture features, were developed and calculated on every one of the subregions. Sequential forward floating selection was used to search for the effective combinations. Using the selected features, a support vector machine (SVM) was optimized using a tenfold validation method to predict the risk of each woman having image-detectable cancer in the next sequential mammography screening. The area under the receiver operating characteristic curve (AUC) was used as the performance assessment index. Results: From a total number of 765 features computed from multiscale subregions, an optimal feature set of 12 features was selected. Applying this feature set, a SVM classifier yielded performance of AUC = 0.729 ± 0.021. The positive predictive value was 0.657 (92 of 140) and the negative predictive value was 0.755 (151 of 200). Conclusions:The study results demonstrated a moderately high positive association between risk prediction scores generated by the quantitative multiscale mammographic image feature analysis and the actual risk of a woman having an image-detectable breast cancer in the next subsequent examinations. C 2015 American Association of Physicists in Medicine. [http://dx

Breast density: why all the fuss?

“…These techniques were initially developed for use on digitised SFM, but have now been applied to DM and a number of fully automated methods have been developed using a cross-validation approach.Early prospective studies on SFM showed moderate relative risks with texture measures and no additional improvement in discrimination with subsequent addition of MD measures, confirmed in many retrospective studies using increasingly complex texture descriptors and combinations of TA features(173,174). Nielsen et al have developed a mammographic texture resemblance marker (MTR) which demonstrated good discrimination in two completely different cohorts, suggesting the measure is generalisable; the best discrimination was achieved with an aggregate of MTR and Cumulus PMD, with an area under the ROC curve (AUC) of 0.66(175).Studies using DM have attained AUCs of between 0.73(176) to 0.85(177), the latter using a complex combination of texture features. Some TA features may be more predictive of certain tumour subtypes (ER positivity or negativity) (178) and whereas MD does not appear to be predictive of risk in women with…”

Similarity measurement of lung masses for medical image retrieval using kernel based semisupervised distance metric

“…Consequently, more effectively choosing or adaptively optimizing these parameters with different image data sets still needs to be investigated in the future studies. Fifth, although we only extracted the texture features to represent the lung nodule, a number of studies used a large feature set to represent the cancer (e.g., 765 features were computed to represent the breast cancer, 32 and 849 features were computed to represent the lung nodule 33 ).…”

A Novel Breast Cancer Risk Assessment Scheme Design Using Dual View Mammograms

“…To achieve more reliable and accurate results, several studies including our previous researches have already investigated and tested the possibility of developing computerized schemes to automatically quantify mammographic density and to predict breast cancer risks using computed mammographic image features [9,10]. To the best of our knowledge, all these researches only used single view mammograms, i.e.…”