Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification
Abstract Background We investigated whether features derived from texture analysis (TA) can distinguish breast density (BD) in spiral photon-counting breast computed tomography (PC-BCT). Methods In this retrospective single-centre study, we analysed 10,000 images from 400 PC-BCT examinations of 200...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-07-01
|
| Series: | European Radiology Experimental |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s41747-022-00285-x |
| _version_ | 1817969566822694912 |
|---|---|
| author | Anna Landsmann Carlotta Ruppert Jann Wieler Patryk Hejduk Alexander Ciritsis Karol Borkowski Moritz C. Wurnig Cristina Rossi Andreas Boss |
| author_facet | Anna Landsmann Carlotta Ruppert Jann Wieler Patryk Hejduk Alexander Ciritsis Karol Borkowski Moritz C. Wurnig Cristina Rossi Andreas Boss |
| author_sort | Anna Landsmann |
| collection | DOAJ |
| description | Abstract Background We investigated whether features derived from texture analysis (TA) can distinguish breast density (BD) in spiral photon-counting breast computed tomography (PC-BCT). Methods In this retrospective single-centre study, we analysed 10,000 images from 400 PC-BCT examinations of 200 patients. Images were categorised into four-level density scale (a–d) using Breast Imaging Reporting and Data System (BI-RADS)-like criteria. After manual definition of representative regions of interest, 19 texture features (TFs) were calculated to analyse the voxel grey-level distribution in the included image area. ANOVA, cluster analysis, and multinomial logistic regression statistics were used. A human readout then was performed on a subset of 60 images to evaluate the reliability of the proposed feature set. Results Of the 19 TFs, 4 first-order features and 7 second-order features showed significant correlation with BD and were selected for further analysis. Multinomial logistic regression revealed an overall accuracy of 80% for BD assessment. The majority of TFs systematically increased or decreased with BD. Skewness (rho -0.81), as a first-order feature, and grey-level nonuniformity (GLN, -0.59), as a second-order feature, showed the strongest correlation with BD, independently of other TFs. Mean skewness and GLN decreased linearly from density a to d. Run-length nonuniformity (RLN), as a second-order feature, showed moderate correlation with BD, but resulted in redundant being correlated with GLN. All other TFs showed only weak correlation with BD (range -0.49 to 0.49, p < 0.001) and were neglected. Conclusion TA of PC-BCT images might be a useful approach to assess BD and may serve as an observer-independent tool. |
| first_indexed | 2024-04-13T20:22:54Z |
| format | Article |
| id | doaj.art-441fc6e307604b019c319366d89857a2 |
| institution | Directory Open Access Journal |
| issn | 2509-9280 |
| language | English |
| last_indexed | 2024-04-13T20:22:54Z |
| publishDate | 2022-07-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | European Radiology Experimental |
| spelling | doaj.art-441fc6e307604b019c319366d89857a22022-12-22T02:31:28ZengSpringerOpenEuropean Radiology Experimental2509-92802022-07-016111310.1186/s41747-022-00285-xRadiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classificationAnna Landsmann0Carlotta Ruppert1Jann Wieler2Patryk Hejduk3Alexander Ciritsis4Karol Borkowski5Moritz C. Wurnig6Cristina Rossi7Andreas Boss8Institute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Computational Physics, Zurich University of Applied SciencesInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Diagnostic Radiology, Hospital Lachen AGInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichInstitute of Diagnostic and Interventional Radiology, University Hospital ZurichAbstract Background We investigated whether features derived from texture analysis (TA) can distinguish breast density (BD) in spiral photon-counting breast computed tomography (PC-BCT). Methods In this retrospective single-centre study, we analysed 10,000 images from 400 PC-BCT examinations of 200 patients. Images were categorised into four-level density scale (a–d) using Breast Imaging Reporting and Data System (BI-RADS)-like criteria. After manual definition of representative regions of interest, 19 texture features (TFs) were calculated to analyse the voxel grey-level distribution in the included image area. ANOVA, cluster analysis, and multinomial logistic regression statistics were used. A human readout then was performed on a subset of 60 images to evaluate the reliability of the proposed feature set. Results Of the 19 TFs, 4 first-order features and 7 second-order features showed significant correlation with BD and were selected for further analysis. Multinomial logistic regression revealed an overall accuracy of 80% for BD assessment. The majority of TFs systematically increased or decreased with BD. Skewness (rho -0.81), as a first-order feature, and grey-level nonuniformity (GLN, -0.59), as a second-order feature, showed the strongest correlation with BD, independently of other TFs. Mean skewness and GLN decreased linearly from density a to d. Run-length nonuniformity (RLN), as a second-order feature, showed moderate correlation with BD, but resulted in redundant being correlated with GLN. All other TFs showed only weak correlation with BD (range -0.49 to 0.49, p < 0.001) and were neglected. Conclusion TA of PC-BCT images might be a useful approach to assess BD and may serve as an observer-independent tool.https://doi.org/10.1186/s41747-022-00285-xBreast densityBreast neoplasmsImage processing (computer-assisted)RadiomicsTomography (x-ray computed) |
| spellingShingle | Anna Landsmann Carlotta Ruppert Jann Wieler Patryk Hejduk Alexander Ciritsis Karol Borkowski Moritz C. Wurnig Cristina Rossi Andreas Boss Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification European Radiology Experimental Breast density Breast neoplasms Image processing (computer-assisted) Radiomics Tomography (x-ray computed) |
| title | Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification |
| title_full | Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification |
| title_fullStr | Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification |
| title_full_unstemmed | Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification |
| title_short | Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification |
| title_sort | radiomics in photon counting dedicated breast ct potential of texture analysis for breast density classification |
| topic | Breast density Breast neoplasms Image processing (computer-assisted) Radiomics Tomography (x-ray computed) |
| url | https://doi.org/10.1186/s41747-022-00285-x |
| work_keys_str_mv | AT annalandsmann radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT carlottaruppert radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT jannwieler radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT patrykhejduk radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT alexanderciritsis radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT karolborkowski radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT moritzcwurnig radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT cristinarossi radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification AT andreasboss radiomicsinphotoncountingdedicatedbreastctpotentialoftextureanalysisforbreastdensityclassification |