AI lesion risk score at different exposure settings
(2024) 17th International Workshop on Breast Imaging, IWBI 2024 In Proceedings of SPIE - The International Society for Optical Engineering 13174.- Abstract
Purpose: The purpose of this study was to investigate whether the lesion risk score provided by an AI system is influenced by the selection of exposure parameters. Methods: A breast phantom which contains a lesion, was imaged with digital mammography with different imaging conditions. The tube voltage, the dose level and the anode-filter combination were varied based on an exposure obtained with automatic exposure control. The organ dose for each image was extracted from the DICOM header. The images were analyzed with an AI system, which provided a lesion risk score (suspicion for malignancy) for each exposure condition. Correlations between the lesion risk score and the exposure conditions were investigated. Results: The results of the... (More)
Purpose: The purpose of this study was to investigate whether the lesion risk score provided by an AI system is influenced by the selection of exposure parameters. Methods: A breast phantom which contains a lesion, was imaged with digital mammography with different imaging conditions. The tube voltage, the dose level and the anode-filter combination were varied based on an exposure obtained with automatic exposure control. The organ dose for each image was extracted from the DICOM header. The images were analyzed with an AI system, which provided a lesion risk score (suspicion for malignancy) for each exposure condition. Correlations between the lesion risk score and the exposure conditions were investigated. Results: The results of the study showed that the organ dose had a strong impact on the lesion risk score. Reducing the organ dose to a low level resulted in that the AI system no longer detected the lesion. Conclusions: Images of suboptimal quality may result in inaccurate AI system performance. In our preliminary analysis, the breast phantom and the lesion were proven to be realistic enough for being analyzed by the AI system.
(Less)
- author
- Tingberg, Anders LU ; Dahlblom, Victor LU ; Bakic, Predrag LU ; Schurz, Haiko ; Strand, Fredrik ; Zackrisson, Sophia LU and Dustler, Magnus LU
- organization
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Artificial intelligence, Breast organ dose, Breast phantom, Digital mammography, Image quality, Suspicion for malignancy
- host publication
- 17th International Workshop on Breast Imaging, IWBI 2024
- series title
- Proceedings of SPIE - The International Society for Optical Engineering
- editor
- Giger, Maryellen L. ; Whitney, Heather M. ; Drukker, Karen and Li, Hui
- volume
- 13174
- article number
- 131741G
- publisher
- SPIE
- conference name
- 17th International Workshop on Breast Imaging, IWBI 2024
- conference location
- Chicago, United States
- conference dates
- 2024-06-09 - 2024-06-12
- external identifiers
-
- scopus:85195413668
- ISSN
- 1996-756X
- 0277-786X
- ISBN
- 9781510680203
- DOI
- 10.1117/12.3026976
- language
- English
- LU publication?
- yes
- id
- b8922b3c-ee64-4c0e-9bfc-7cfb3fa15922
- date added to LUP
- 2024-09-30 10:13:50
- date last changed
- 2024-10-28 14:20:51
@inproceedings{b8922b3c-ee64-4c0e-9bfc-7cfb3fa15922, abstract = {{<p>Purpose: The purpose of this study was to investigate whether the lesion risk score provided by an AI system is influenced by the selection of exposure parameters. Methods: A breast phantom which contains a lesion, was imaged with digital mammography with different imaging conditions. The tube voltage, the dose level and the anode-filter combination were varied based on an exposure obtained with automatic exposure control. The organ dose for each image was extracted from the DICOM header. The images were analyzed with an AI system, which provided a lesion risk score (suspicion for malignancy) for each exposure condition. Correlations between the lesion risk score and the exposure conditions were investigated. Results: The results of the study showed that the organ dose had a strong impact on the lesion risk score. Reducing the organ dose to a low level resulted in that the AI system no longer detected the lesion. Conclusions: Images of suboptimal quality may result in inaccurate AI system performance. In our preliminary analysis, the breast phantom and the lesion were proven to be realistic enough for being analyzed by the AI system.</p>}}, author = {{Tingberg, Anders and Dahlblom, Victor and Bakic, Predrag and Schurz, Haiko and Strand, Fredrik and Zackrisson, Sophia and Dustler, Magnus}}, booktitle = {{17th International Workshop on Breast Imaging, IWBI 2024}}, editor = {{Giger, Maryellen L. and Whitney, Heather M. and Drukker, Karen and Li, Hui}}, isbn = {{9781510680203}}, issn = {{1996-756X}}, keywords = {{Artificial intelligence; Breast organ dose; Breast phantom; Digital mammography; Image quality; Suspicion for malignancy}}, language = {{eng}}, publisher = {{SPIE}}, series = {{Proceedings of SPIE - The International Society for Optical Engineering}}, title = {{AI lesion risk score at different exposure settings}}, url = {{http://dx.doi.org/10.1117/12.3026976}}, doi = {{10.1117/12.3026976}}, volume = {{13174}}, year = {{2024}}, }