An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images
The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segment...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Biomolecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2218-273X/13/2/343 |
| _version_ | 1827758318995636224 |
|---|---|
| author | Lisa M. Duff Andrew F. Scarsbrook Nishant Ravikumar Russell Frood Gijs D. van Praagh Sarah L. Mackie Marc A. Bailey Jason M. Tarkin Justin C. Mason Kornelis S. M. van der Geest Riemer H. J. A. Slart Ann W. Morgan Charalampos Tsoumpas |
| author_facet | Lisa M. Duff Andrew F. Scarsbrook Nishant Ravikumar Russell Frood Gijs D. van Praagh Sarah L. Mackie Marc A. Bailey Jason M. Tarkin Justin C. Mason Kornelis S. M. van der Geest Riemer H. J. A. Slart Ann W. Morgan Charalampos Tsoumpas |
| author_sort | Lisa M. Duff |
| collection | DOAJ |
| description | The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segmented by convolutional neural network (CNN) on FDG PET-CT of aortitis and control patients. The FDG PET-CT dataset was split into training (43 aortitis:21 control), test (12 aortitis:5 control) and validation (24 aortitis:14 control) cohorts. Radiomic features (RF), including SUV metrics, were extracted from the segmented data and harmonized. Three radiomic fingerprints were constructed: A—RFs with high diagnostic utility removing highly correlated RFs; B used principal component analysis (PCA); C—Random Forest intrinsic feature selection. The diagnostic utility was evaluated with accuracy and area under the receiver operating characteristic curve (AUC). Several RFs and Fingerprints had high AUC values (AUC > 0.8), confirmed by balanced accuracy, across training, test and external validation datasets. Good diagnostic performance achieved across several multi-centre datasets suggests that a radiomic pipeline can be generalizable. These findings could be used to build an automated clinical decision tool to facilitate objective and standardized assessment regardless of observer experience. |
| first_indexed | 2024-03-11T09:06:01Z |
| format | Article |
| id | doaj.art-7d7a3b735b86439e87f00a13c1d8467f |
| institution | Directory Open Access Journal |
| issn | 2218-273X |
| language | English |
| last_indexed | 2024-03-11T09:06:01Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biomolecules |
| spelling | doaj.art-7d7a3b735b86439e87f00a13c1d8467f2023-11-16T19:23:51ZengMDPI AGBiomolecules2218-273X2023-02-0113234310.3390/biom13020343An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT ImagesLisa M. Duff0Andrew F. Scarsbrook1Nishant Ravikumar2Russell Frood3Gijs D. van Praagh4Sarah L. Mackie5Marc A. Bailey6Jason M. Tarkin7Justin C. Mason8Kornelis S. M. van der Geest9Riemer H. J. A. Slart10Ann W. Morgan11Charalampos Tsoumpas12School of Medicine, University of Leeds, Leeds LS2 9JT, UKSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKDepartment of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, 9713 GZ Groningen, The NetherlandsSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKDivision of Cardiovascular Medicine, University of Cambridge, Cambridge CB2 0QQ, UKNational Heart and Lung Institute, Imperial College London, London SW3 6LY, UKDepartment of Rheumatology and Clinical Immunology, University of Groningen, University Medical Center Groningen, 9713 GZ Groningen, The NetherlandsDepartment of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, 9713 GZ Groningen, The NetherlandsSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKSchool of Medicine, University of Leeds, Leeds LS2 9JT, UKThe aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segmented by convolutional neural network (CNN) on FDG PET-CT of aortitis and control patients. The FDG PET-CT dataset was split into training (43 aortitis:21 control), test (12 aortitis:5 control) and validation (24 aortitis:14 control) cohorts. Radiomic features (RF), including SUV metrics, were extracted from the segmented data and harmonized. Three radiomic fingerprints were constructed: A—RFs with high diagnostic utility removing highly correlated RFs; B used principal component analysis (PCA); C—Random Forest intrinsic feature selection. The diagnostic utility was evaluated with accuracy and area under the receiver operating characteristic curve (AUC). Several RFs and Fingerprints had high AUC values (AUC > 0.8), confirmed by balanced accuracy, across training, test and external validation datasets. Good diagnostic performance achieved across several multi-centre datasets suggests that a radiomic pipeline can be generalizable. These findings could be used to build an automated clinical decision tool to facilitate objective and standardized assessment regardless of observer experience.https://www.mdpi.com/2218-273X/13/2/343aortitisradiomicsmachine learningconvolutional neural networkpositron emission tomography/computed tomography |
| spellingShingle | Lisa M. Duff Andrew F. Scarsbrook Nishant Ravikumar Russell Frood Gijs D. van Praagh Sarah L. Mackie Marc A. Bailey Jason M. Tarkin Justin C. Mason Kornelis S. M. van der Geest Riemer H. J. A. Slart Ann W. Morgan Charalampos Tsoumpas An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images Biomolecules aortitis radiomics machine learning convolutional neural network positron emission tomography/computed tomography |
| title | An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images |
| title_full | An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images |
| title_fullStr | An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images |
| title_full_unstemmed | An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images |
| title_short | An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images |
| title_sort | automated method for artifical intelligence assisted diagnosis of active aortitis using radiomic analysis of fdg pet ct images |
| topic | aortitis radiomics machine learning convolutional neural network positron emission tomography/computed tomography |
| url | https://www.mdpi.com/2218-273X/13/2/343 |
| work_keys_str_mv | AT lisamduff anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT andrewfscarsbrook anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT nishantravikumar anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT russellfrood anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT gijsdvanpraagh anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT sarahlmackie anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT marcabailey anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT jasonmtarkin anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT justincmason anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT kornelissmvandergeest anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT riemerhjaslart anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT annwmorgan anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT charalampostsoumpas anautomatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT lisamduff automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT andrewfscarsbrook automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT nishantravikumar automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT russellfrood automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT gijsdvanpraagh automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT sarahlmackie automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT marcabailey automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT jasonmtarkin automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT justincmason automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT kornelissmvandergeest automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT riemerhjaslart automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT annwmorgan automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages AT charalampostsoumpas automatedmethodforartificalintelligenceassisteddiagnosisofactiveaortitisusingradiomicanalysisoffdgpetctimages |