Assessment of breast positioning criteria in mammographic screening: Agreement between artificial intelligence software and radiographers.

OBJECTIVES: To determine the agreement between artificial intelligence software (AI) and radiographers in assessing breast positioning criteria for mammograms from standard digital mammography and digital breast tomosynthesis. METHODS: Assessment of breast positioning was performed by AI and by four radiographers in pairs of two on 156 examinations of women screened in Bergen, April to September 2019, as part of BreastScreen Norway. Ten criteria were used; three for craniocaudal and seven for mediolateral-oblique view. The criteria evaluated the appearance of the nipple, breast rotation, pectoral muscle, inframammary fold and pectoral nipple line. Intraclass correlation and Cohen's kappa coefficient (κ) were used to investigate the correlation and agreement between the radiographer's assessments and AI. RESULTS: The intraclass correlation for the pectoral nipple line between the radiographers and AI was >0.92. A substantial to almost perfect agreement (κ > 0.69) was observed between the radiographers and AI on the nipple in profile criterion. We observed a slight to moderate agreement for the other criteria (κ = 0.06-0.52) and generally a higher agreement between the two pairs of radiographers (mean κ = 0.70) than between the radiographers and AI (mean κ = 0.41). CONCLUSIONS: AI has great potential in evaluating breast position criteria in mammography by reducing subjectivity. However, varying agreement between radiographers and AI was observed. Standardized and evidence-based criteria for definitions, understandings and assessment methods are needed to reach optimal image quality in mammography.

Screening at stationary versus mobile units in BreastScreen Norway.

Objective: To compare breast characteristics, compression parameters, and early performance measures (rates of recall, screen-detected and interval breast cancer, and histopathologic tumour characteristics) for mammographic screening at a stationary versus mobile screening unit. Methods: Results from 92,408 mammographic screening examinations performed as part of BreastScreen Norway during 2008­2017 at either a stationary (n = 52,620) or mobile (n = 39,788) unit in Hordaland county were compared using descriptive statistics and generalized estimating equations. A generalized estimating equation for a binary outcome was used to estimate crude and adjusted odds ratios with 95% confidence intervals for the outcomes of interest. Adjusted generalized estimating equation models included age, breast volume, and density grade as covariates. Results: Screening at the stationary unit was performed on smaller breasts with higher mammographic density, using lower compression force but higher pressure than at the mobile unit. Using the stationary screening unit as reference, for women screened at the mobile unit, the adjusted odds ratio was: for recall 0.94 (95% CI: 0.87--1.01), screen-detected breast cancer 0.92 (95% CI: 0.78--1.10), and interval breast cancer 1.17 (95% CI: 0.83­1.64). Conclusions: The quality of care did not differ for women screened at the stationary versus the mobile unit, but there were differences between the women who attended the two units. Sociodemographic factors should be included in future analyses to fully understand the risk of breast cancer among women residing in urban versus rural areas.