Optical pre-screening in breast screening programs: Can we identify women who benefit most from limited mammography resources?

Background: In excess of 60% of all cancers are detected in low and middle-income countries, with breast cancer (BC) the dominant malignancy for women. Incidence rates continue to climb, most noticeably in the less than 50-year-old population. Expansion of mammography infrastructure and resources is lacking, resulting in over 60% of women diagnosed with stage III/IV BC in the majority of these countries. Optical Breast Spectroscopy (OBS) was shown to correlate well with mammographic breast density (MBD). OBS could aid breast screening programs in low- and middle-income countries by lowering the number of mammographs required for complete population coverage. However, its performance needs to be tested in large population trails to ensure high sensitivity and acceptable specificity. Methods: For the planned studies in low- and middle-income countries in different continents, online methods need to be implemented to monitor the performance and data collection by these devices, operated by trained nurses. Based on existing datasets, procedures were developed to validate an individual woman's data integrity and to identify operator errors versus system malfunctions. Results: Using a dataset comprising spectra from 360 women collected by 2 instruments in different locations and with 3 different trained operators, automated methods were developed to identify 100% of the source or photodetector malfunctions as well as incorrect calibrations and 96% of instances of insufficient tissue contact. Conclusions: Implementing the dataset validation locally in each instrument and tethered to a cloud database will allow the planned clinical trials to proceed.