Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach

H. S. Vries; G. D. Van Praagh; P. H. Nienhuis; O. Bouhali; R. H.J.A. Slart; L. Alic

doi:10.1109/CBMS58004.2023.00334

Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach

H. S. Vries, G. D. Van Praagh, P. H. Nienhuis, O. Bouhali, R. H.J.A. Slart, L. Alic

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Background This work aims to investigate the feasibility of an explainable machine learning model based on radiomics features to differentiate between giant cell arteritis (GCA) and atherosclerosis in aortic [F-18]FDG-PET scans.Method Twenty [F-18]FDG-PET scans (ten of patients with GCA, ten with atherosclerosis) were retrospectively included. The aorta was delineated into four segments (ascending, arch, descending, and abdominal aorta). In total, 93 radiomic features and two quantitative features were extracted from each of the 80 segments. Four different feature selection methods and four classifiers were used to identify important features for the machine learning model and determine the probability. The model's performance was evaluated using accuracy and AUC. To enhance explainability of the model, feature importance was determined, and an occlusion sensitivity map of the aorta was created.Results The combination of the first-order skewness, GLDM dependence non-uniformity, and GLRLM run entropy features showed the highest accuracy and AUC of, 0.90 +/- 0.08 and 0.960 +/- 0.029, respectively.Conclusion This study demonstrated the potential of an explainable radiomics-based machine learning model for the differentiation between GCA and atherosclerosis in [F-18]FDG-PET scans.

Original language	English
Title of host publication	2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms
Editors	JR Almeida, M Spiliopoulou, JAB Andrades, G Placidi, AR Gonzalez, R Sicilia, B Kane
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	870-875
Number of pages	6
ISBN (Electronic)	9798350312249
DOIs	https://doi.org/10.1109/CBMS58004.2023.00334
Publication status	Published - 2023
Externally published	Yes
Event	36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023 - L�Aquila, Italy Duration: 22 Jun 2023 → 24 Jun 2023

Publication series

Name	Ieee International Symposium On Computer-based Medical Systems

Conference

Conference	36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023
Country/Territory	Italy
City	L�Aquila
Period	22/06/23 → 24/06/23

Keywords

Atherosclerosis
Explainable machine learning
Giant cell arteritis
Radiomics
[f-18]fdg-pet

Access to Document

10.1109/CBMS58004.2023.00334

Cite this

Vries, H. S., Van Praagh, G. D., Nienhuis, P. H., Bouhali, O., Slart, R. H. J. A., & Alic, L. (2023). Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach. In JR. Almeida, M. Spiliopoulou, JAB. Andrades, G. Placidi, AR. Gonzalez, R. Sicilia, & B. Kane (Eds.), 2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms (pp. 870-875). (Ieee International Symposium On Computer-based Medical Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CBMS58004.2023.00334

Vries, H. S. ; Van Praagh, G. D. ; Nienhuis, P. H. et al. / Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET : An explainable machine learning approach. 2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms. editor / JR Almeida ; M Spiliopoulou ; JAB Andrades ; G Placidi ; AR Gonzalez ; R Sicilia ; B Kane. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 870-875 (Ieee International Symposium On Computer-based Medical Systems).

@inproceedings{8200b1842cfd45809df800c600d68a4a,

title = "Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach",

abstract = "Background This work aims to investigate the feasibility of an explainable machine learning model based on radiomics features to differentiate between giant cell arteritis (GCA) and atherosclerosis in aortic [F-18]FDG-PET scans.Method Twenty [F-18]FDG-PET scans (ten of patients with GCA, ten with atherosclerosis) were retrospectively included. The aorta was delineated into four segments (ascending, arch, descending, and abdominal aorta). In total, 93 radiomic features and two quantitative features were extracted from each of the 80 segments. Four different feature selection methods and four classifiers were used to identify important features for the machine learning model and determine the probability. The model's performance was evaluated using accuracy and AUC. To enhance explainability of the model, feature importance was determined, and an occlusion sensitivity map of the aorta was created.Results The combination of the first-order skewness, GLDM dependence non-uniformity, and GLRLM run entropy features showed the highest accuracy and AUC of, 0.90 +/- 0.08 and 0.960 +/- 0.029, respectively.Conclusion This study demonstrated the potential of an explainable radiomics-based machine learning model for the differentiation between GCA and atherosclerosis in [F-18]FDG-PET scans.",

keywords = "Atherosclerosis, Explainable machine learning, Giant cell arteritis, Radiomics, [f-18]fdg-pet",

author = "Vries, {H. S.} and {Van Praagh}, {G. D.} and Nienhuis, {P. H.} and O. Bouhali and Slart, {R. H.J.A.} and L. Alic",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023 ; Conference date: 22-06-2023 Through 24-06-2023",

year = "2023",

doi = "10.1109/CBMS58004.2023.00334",

language = "English",

series = "Ieee International Symposium On Computer-based Medical Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "870--875",

editor = "JR Almeida and M Spiliopoulou and JAB Andrades and G Placidi and AR Gonzalez and R Sicilia and B Kane",

booktitle = "2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms",

address = "United States",

}

Vries, HS, Van Praagh, GD, Nienhuis, PH, Bouhali, O, Slart, RHJA & Alic, L 2023, Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach. in JR Almeida, M Spiliopoulou, JAB Andrades, G Placidi, AR Gonzalez, R Sicilia & B Kane (eds), 2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms. Ieee International Symposium On Computer-based Medical Systems, Institute of Electrical and Electronics Engineers Inc., pp. 870-875, 36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023, L�Aquila, Italy, 22/06/23. https://doi.org/10.1109/CBMS58004.2023.00334

Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach. / Vries, H. S.; Van Praagh, G. D.; Nienhuis, P. H. et al.
2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms. ed. / JR Almeida; M Spiliopoulou; JAB Andrades; G Placidi; AR Gonzalez; R Sicilia; B Kane. Institute of Electrical and Electronics Engineers Inc., 2023. p. 870-875 (Ieee International Symposium On Computer-based Medical Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET

T2 - 36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023

AU - Vries, H. S.

AU - Van Praagh, G. D.

AU - Nienhuis, P. H.

AU - Bouhali, O.

AU - Slart, R. H.J.A.

AU - Alic, L.

PY - 2023

Y1 - 2023

N2 - Background This work aims to investigate the feasibility of an explainable machine learning model based on radiomics features to differentiate between giant cell arteritis (GCA) and atherosclerosis in aortic [F-18]FDG-PET scans.Method Twenty [F-18]FDG-PET scans (ten of patients with GCA, ten with atherosclerosis) were retrospectively included. The aorta was delineated into four segments (ascending, arch, descending, and abdominal aorta). In total, 93 radiomic features and two quantitative features were extracted from each of the 80 segments. Four different feature selection methods and four classifiers were used to identify important features for the machine learning model and determine the probability. The model's performance was evaluated using accuracy and AUC. To enhance explainability of the model, feature importance was determined, and an occlusion sensitivity map of the aorta was created.Results The combination of the first-order skewness, GLDM dependence non-uniformity, and GLRLM run entropy features showed the highest accuracy and AUC of, 0.90 +/- 0.08 and 0.960 +/- 0.029, respectively.Conclusion This study demonstrated the potential of an explainable radiomics-based machine learning model for the differentiation between GCA and atherosclerosis in [F-18]FDG-PET scans.

AB - Background This work aims to investigate the feasibility of an explainable machine learning model based on radiomics features to differentiate between giant cell arteritis (GCA) and atherosclerosis in aortic [F-18]FDG-PET scans.Method Twenty [F-18]FDG-PET scans (ten of patients with GCA, ten with atherosclerosis) were retrospectively included. The aorta was delineated into four segments (ascending, arch, descending, and abdominal aorta). In total, 93 radiomic features and two quantitative features were extracted from each of the 80 segments. Four different feature selection methods and four classifiers were used to identify important features for the machine learning model and determine the probability. The model's performance was evaluated using accuracy and AUC. To enhance explainability of the model, feature importance was determined, and an occlusion sensitivity map of the aorta was created.Results The combination of the first-order skewness, GLDM dependence non-uniformity, and GLRLM run entropy features showed the highest accuracy and AUC of, 0.90 +/- 0.08 and 0.960 +/- 0.029, respectively.Conclusion This study demonstrated the potential of an explainable radiomics-based machine learning model for the differentiation between GCA and atherosclerosis in [F-18]FDG-PET scans.

KW - Atherosclerosis

KW - Explainable machine learning

KW - Giant cell arteritis

KW - Radiomics

KW - [f-18]fdg-pet

UR - http://www.scopus.com/inward/record.url?scp=85166476630&partnerID=8YFLogxK

U2 - 10.1109/CBMS58004.2023.00334

DO - 10.1109/CBMS58004.2023.00334

M3 - Conference contribution

AN - SCOPUS:85166476630

T3 - Ieee International Symposium On Computer-based Medical Systems

SP - 870

EP - 875

BT - 2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms

A2 - Almeida, JR

A2 - Spiliopoulou, M

A2 - Andrades, JAB

A2 - Placidi, G

A2 - Gonzalez, AR

A2 - Sicilia, R

A2 - Kane, B

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 22 June 2023 through 24 June 2023

ER -

Vries HS, Van Praagh GD, Nienhuis PH, Bouhali O, Slart RHJA, Alic L. Differentiating between giant cell arteritis and atherosclerosis on [18]FDG-PET: An explainable machine learning approach. In Almeida JR, Spiliopoulou M, Andrades JAB, Placidi G, Gonzalez AR, Sicilia R, Kane B, editors, 2023 Ieee 36th International Symposium On Computer-based Medical Systems, Cbms. Institute of Electrical and Electronics Engineers Inc. 2023. p. 870-875. (Ieee International Symposium On Computer-based Medical Systems). doi: 10.1109/CBMS58004.2023.00334

Differentiating between giant cell arteritis and atherosclerosis on [<SUP>18</SUP>]FDG-PET: An explainable machine learning approach