Image-based analysis of breast tumour growth rate may help optimize breast cancer screening and diagnosis. It may improve the identification of aggressive tumours and suggest optimal screening intervals. Virtual clinical trial (VCT) is a simulation-based method used to evaluate and optimize medical imaging systems and design clinical trials. Our work is motivated by desire to simulate multiple screening rounds with growing tumours. We have developed a model to simulate tumours with various growth rates; this study aims at evaluating the model. We used clinical data on tumour volume doubling times (TVDT) from our previous study, to fit a probability distribution (“clinical fit”). Growing tumours were inserted into 30 virtual breasts (“simulated cohort”). Based on the clinical fit we simulated two successive screening rounds for each virtual breast. TVDT from clinical and simulated images were compared. Tumour size was measured from simulated mammograms by a radiologist in three repeated sessions, to estimate TVDT (“estimated TVDT”). Reproducibility of measured sizes decreased slightly for small tumours. The mean TVDT from the clinical fit was 297±169 days, whereas the simulated cohort had 322±217 days, and the average estimated TVDT 340 ± 287 days. The median difference between the simulated and estimated TVDT was 12 days (4% of the mean clinical TVDT). Comparisons between other data sets suggest no significant difference (p>0.5). The proposed tumour growth model suggested close agreement with clinical results, supporting potential use in VCTs of temporal breast imaging.
Artificial intelligence (AI) applications are increasingly seeing use in breast imaging, particularly to assist in or automate the reading of mammograms. Another novel technique is mechanical imaging (MI) which estimates the relative stiffness of suspicious breast abnormalities by measuring the distribution of pressure on the compressed breast. This study investigates the feasibility of combining AI and MI information in breast imaging to provide further diagnostic information. Forty-six women recalled from screening were included in the analysis. Mammograms with findings scored on a suspiciousness scale by an AI tool, and corresponding pressure distributions were collected for each woman. The cases were divided into three groups by diagnosis; biopsy-proven cancer, biopsy-proven benign and non-biopsied, very likely benign. For all three groups, the relative increase of pressure at the location of the finding marked most suspicious by the AI software was recorded. A significant correlation between the relative pressure increase at the AI finding and the AI score was established in the group with cancer (p=0.043), but neither group of healthy women showed such a correlation. This study suggests that AI and MI indicate independent markers for breast cancer. The combination of these two methods has the potential to increase the accuracy of mammography screening, but further research is needed.