Screening mammography is a widely used approach for early breast cancer detection, effectively increasing the survival rate of affected patients. According to the Food and Drug Administration’s Mammography Quality Standards Act and Program statistics, approximately 39 million mammography procedures are performed in the United States each year. Therefore, breast cancer screening is among the most common radiological tasks. Interpretation of screening mammograms by a specialist radiologist includes primarily the review of breast positioning quality, which is a key factor affecting the sensitivity of mammography and thus the diagnostic performance. Each mammogram with inadequate positioning may lead to a missed cancer or, in case of false positive signal interpretation, to follow-up activities, increased emotional burden and potential over-therapy and must be repeated, requiring the return of the patient. In this study, we have developed deep convolutional neuronal networks to differentiate mammograms with inadequate breast positioning from the adequate ones. The aim of the proposed automated positioning quality evaluation is to assist radiology technologists in detecting poorly positioned mammograms during patient visits, improve mammography performance, and decrease the recall rate. The implemented models have achieved 96.5% accuracy in cranio-caudal view classification and 93.3% accuracy in mediolateral oblique view regarding breast positioning quality. In addition to these results, we developed a software module that allows the study to be applied in practice by presenting the implemented model predictions and informing the technologist about the missing quality criteria.

Most metastatic colorectal cancer (mCRC) patients succumb to refractory disease due to secondary chemotherapy resistance. To elucidate the molecular changes associated with secondary resistance, we recruited 64 patients with mCRC and hepatic metastases before standard first-line chemotherapy between 2014 and 2018. We subjected DNA from primary tumor specimens (P), hepatic metastasis specimens after treatment (M), and liquid biopsies (L) taken prior to (pre), during (intra), and after (post) treatment to next generation sequencing. We performed Nanostring expression analysis in P and M specimens. Comparative bioinformatics and statistical analysis revealed typical mutational patterns with frequent alterations in TP53, APC, and KRAS in P specimens (n = 48). P and pre-L (n = 42), as well as matched P and M (n = 30), displayed a similar mutation spectrum. In contrast, gene expression profiles classified P (n = 31) and M (n = 23), distinguishable by up-regulation of immune/cytokine receptor and autophagy programs. Switching of consensus molecular subtypes from P to M occurred in 58.3% of cases. M signature genes SFRP2 and SPP1 associated with inferior survival, as validated in an independent cohort. Molecular changes during first-line treatment were detectable by expression profiling rather than by mutational tumor and liquid biopsy analyses. SFRP2 and SPP1 may serve as biomarkers and/or actionable targets.