Magnus Dustler
Researcher
Intra- and inter-rater reliability of compressed breast thickness, applied force, and pressure distribution in screening mammography
Author
Summary, in English
Background
Ensuring equivalent and reproducible breast compression between mammographic screening rounds is important for the diagnostic performance of mammography, yet the extent to which screening mammography positioning and compression is reproducible for the individual woman is unknown.
Purpose
To investigate the intra- and inter-rater reliability of breast compression in screening mammography.
Materials and Methods
Eleven breast-healthy women participated in the study. Two experienced radiographers independently positioned and compressed the breasts of each participant in two projections—right craniocaudal and left mediolateral oblique—and at two time points. The spatial pressure distribution on the compressed breast was measured using a pressure sensor matrix. Applied force, compressed breast thickness, force in field of view, contact area, mean pressure, and center of mass (anterio-posterior and mediolateral axes) were measured. The reliabilities of the measures between the time points for each radiographer (intra-rater reliability) and between the radiographers (inter-rater reliability) were analyzed using the intraclass correlation coefficient (ICC).
Results
Intra- and inter-rater reliabilities from both projections demonstrated good to excellent ICCs (≥0.82) for compressed breast thickness, contact area, and anterio-posterior center of mass. The other measures produced ICCs that varied from poor (≤0.42) to excellent (≥0.93) between time points and between radiographers.
Conclusion
Intra- and inter-rater reliability of breast compression was consistently high for compressed breast thickness, contact area, and anterio-posterior center of mass but low for mediolateral center of mass and applied force. Further research is needed to establish objective and clinically useful parameters for the standardization of breast compression.
Ensuring equivalent and reproducible breast compression between mammographic screening rounds is important for the diagnostic performance of mammography, yet the extent to which screening mammography positioning and compression is reproducible for the individual woman is unknown.
Purpose
To investigate the intra- and inter-rater reliability of breast compression in screening mammography.
Materials and Methods
Eleven breast-healthy women participated in the study. Two experienced radiographers independently positioned and compressed the breasts of each participant in two projections—right craniocaudal and left mediolateral oblique—and at two time points. The spatial pressure distribution on the compressed breast was measured using a pressure sensor matrix. Applied force, compressed breast thickness, force in field of view, contact area, mean pressure, and center of mass (anterio-posterior and mediolateral axes) were measured. The reliabilities of the measures between the time points for each radiographer (intra-rater reliability) and between the radiographers (inter-rater reliability) were analyzed using the intraclass correlation coefficient (ICC).
Results
Intra- and inter-rater reliabilities from both projections demonstrated good to excellent ICCs (≥0.82) for compressed breast thickness, contact area, and anterio-posterior center of mass. The other measures produced ICCs that varied from poor (≤0.42) to excellent (≥0.93) between time points and between radiographers.
Conclusion
Intra- and inter-rater reliability of breast compression was consistently high for compressed breast thickness, contact area, and anterio-posterior center of mass but low for mediolateral center of mass and applied force. Further research is needed to establish objective and clinically useful parameters for the standardization of breast compression.
Department/s
- Integrative Health Research
- LUCC: Lund University Cancer Centre
- Radiology Diagnostics, Malmö
- Medical Radiation Physics, Malmö
Publishing year
2021-12-09
Language
English
Publication/Series
Acta Radiologica Open
Volume
10
Issue
12
Document type
Journal article
Publisher
SAGE Publications
Topic
- Radiology, Nuclear Medicine and Medical Imaging
Status
Published
Research group
- Integrative Health Research
- Radiology Diagnostics, Malmö
- Medical Radiation Physics, Malmö
ISBN/ISSN/Other
- ISSN: 2058-4601