A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography
The present study was conducted to investigate the potential of radiomics to develop an explainable AI-based system to be applied to ultra-widefield fundus retinographies (UWF-FRTs) with the objective of predicting the presence of the early signs of Age-related Macular Degeneration (AMD) and stratif...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/18/2965 |
| _version_ | 1797580596602470400 |
|---|---|
| author | Matteo Interlenghi Giancarlo Sborgia Alessandro Venturi Rodolfo Sardone Valentina Pastore Giacomo Boscia Luca Landini Giacomo Scotti Alfredo Niro Federico Moscara Luca Bandi Christian Salvatore Isabella Castiglioni |
| author_facet | Matteo Interlenghi Giancarlo Sborgia Alessandro Venturi Rodolfo Sardone Valentina Pastore Giacomo Boscia Luca Landini Giacomo Scotti Alfredo Niro Federico Moscara Luca Bandi Christian Salvatore Isabella Castiglioni |
| author_sort | Matteo Interlenghi |
| collection | DOAJ |
| description | The present study was conducted to investigate the potential of radiomics to develop an explainable AI-based system to be applied to ultra-widefield fundus retinographies (UWF-FRTs) with the objective of predicting the presence of the early signs of Age-related Macular Degeneration (AMD) and stratifying subjects with low- versus high-risk of AMD. The ultimate aim was to provide clinicians with an automatic classifier and a signature of objective quantitative image biomarkers of AMD. The use of Machine Learning (ML) and radiomics was based on intensity and texture analysis in the macular region, detected by a Deep Learning (DL)-based macular detector. Two-hundred and twenty six UWF-FRTs were retrospectively collected from two centres and manually annotated to train and test the algorithms. Notably, the combination of the ML-based radiomics model and the DL-based macular detector reported 93% sensitivity and 74% specificity when applied to the data of the centre used for external testing, capturing explainable features associated with drusen or pigmentary abnormalities. In comparison to the human operator’s annotations, the system yielded a 0.79 Cohen <i>κ</i>, demonstrating substantial concordance. To our knowledge, these results are the first provided by a radiomic approach for AMD supporting the suitability of an explainable feature extraction method combined with ML for UWF-FRT. |
| first_indexed | 2024-03-10T22:51:59Z |
| format | Article |
| id | doaj.art-3c6837e3e4ca476ab4b2c7614e39e962 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T22:51:59Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-3c6837e3e4ca476ab4b2c7614e39e9622023-11-19T10:14:03ZengMDPI AGDiagnostics2075-44182023-09-011318296510.3390/diagnostics13182965A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus RetinographyMatteo Interlenghi0Giancarlo Sborgia1Alessandro Venturi2Rodolfo Sardone3Valentina Pastore4Giacomo Boscia5Luca Landini6Giacomo Scotti7Alfredo Niro8Federico Moscara9Luca Bandi10Christian Salvatore11Isabella Castiglioni12DeepTrace Technologies S.R.L., 20122 Milan, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyDeepTrace Technologies S.R.L., 20122 Milan, ItalyNational Institute of Gastroenterology—IRCCS “Saverio de Bellis”, 70013 Castellana Grotte, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyEye Clinic, Hospital “SS. Annunziata”, ASL Taranto, 74121 Taranto, ItalyDepartment of Medical Science, Neuroscience and Sense Organs, Eye Clinic, University of Bari Aldo Moro, 70121 Bari, ItalyDeepTrace Technologies S.R.L., 20122 Milan, ItalyDeepTrace Technologies S.R.L., 20122 Milan, ItalyDepartment of Physics “Giuseppe Occhialini”, University of Milan-Bicocca, 20126 Milan, ItalyThe present study was conducted to investigate the potential of radiomics to develop an explainable AI-based system to be applied to ultra-widefield fundus retinographies (UWF-FRTs) with the objective of predicting the presence of the early signs of Age-related Macular Degeneration (AMD) and stratifying subjects with low- versus high-risk of AMD. The ultimate aim was to provide clinicians with an automatic classifier and a signature of objective quantitative image biomarkers of AMD. The use of Machine Learning (ML) and radiomics was based on intensity and texture analysis in the macular region, detected by a Deep Learning (DL)-based macular detector. Two-hundred and twenty six UWF-FRTs were retrospectively collected from two centres and manually annotated to train and test the algorithms. Notably, the combination of the ML-based radiomics model and the DL-based macular detector reported 93% sensitivity and 74% specificity when applied to the data of the centre used for external testing, capturing explainable features associated with drusen or pigmentary abnormalities. In comparison to the human operator’s annotations, the system yielded a 0.79 Cohen <i>κ</i>, demonstrating substantial concordance. To our knowledge, these results are the first provided by a radiomic approach for AMD supporting the suitability of an explainable feature extraction method combined with ML for UWF-FRT.https://www.mdpi.com/2075-4418/13/18/2965age-related macular degeneration (AMD)ultra-widefield (UWF)fundus retinography (FRT)artificial intelligence (AI)machine learning (ML)radiomics |
| spellingShingle | Matteo Interlenghi Giancarlo Sborgia Alessandro Venturi Rodolfo Sardone Valentina Pastore Giacomo Boscia Luca Landini Giacomo Scotti Alfredo Niro Federico Moscara Luca Bandi Christian Salvatore Isabella Castiglioni A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography Diagnostics age-related macular degeneration (AMD) ultra-widefield (UWF) fundus retinography (FRT) artificial intelligence (AI) machine learning (ML) radiomics |
| title | A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography |
| title_full | A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography |
| title_fullStr | A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography |
| title_full_unstemmed | A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography |
| title_short | A Radiomic-Based Machine Learning System to Diagnose Age-Related Macular Degeneration from Ultra-Widefield Fundus Retinography |
| title_sort | radiomic based machine learning system to diagnose age related macular degeneration from ultra widefield fundus retinography |
| topic | age-related macular degeneration (AMD) ultra-widefield (UWF) fundus retinography (FRT) artificial intelligence (AI) machine learning (ML) radiomics |
| url | https://www.mdpi.com/2075-4418/13/18/2965 |
| work_keys_str_mv | AT matteointerlenghi aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giancarlosborgia aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT alessandroventuri aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT rodolfosardone aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT valentinapastore aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giacomoboscia aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT lucalandini aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giacomoscotti aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT alfredoniro aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT federicomoscara aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT lucabandi aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT christiansalvatore aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT isabellacastiglioni aradiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT matteointerlenghi radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giancarlosborgia radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT alessandroventuri radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT rodolfosardone radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT valentinapastore radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giacomoboscia radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT lucalandini radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT giacomoscotti radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT alfredoniro radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT federicomoscara radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT lucabandi radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT christiansalvatore radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography AT isabellacastiglioni radiomicbasedmachinelearningsystemtodiagnoseagerelatedmaculardegenerationfromultrawidefieldfundusretinography |