X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context
Summary: Background: Atypical cartilaginous tumour (ACT) and high-grade chondrosarcoma (CS) of long bones are respectively managed with active surveillance or curettage and wide resection. Our aim was to determine diagnostic performance of X-rays radiomics-based machine learning for classification...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-03-01
|
| Series: | EBioMedicine |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352396424000537 |
| _version_ | 1797302499921625088 |
|---|---|
| author | Salvatore Gitto Alessio Annovazzi Kitija Nulle Matteo Interlenghi Christian Salvatore Vincenzo Anelli Jacopo Baldi Carmelo Messina Domenico Albano Filippo Di Luca Elisabetta Armiraglio Antonina Parafioriti Alessandro Luzzati Roberto Biagini Isabella Castiglioni Luca Maria Sconfienza |
| author_facet | Salvatore Gitto Alessio Annovazzi Kitija Nulle Matteo Interlenghi Christian Salvatore Vincenzo Anelli Jacopo Baldi Carmelo Messina Domenico Albano Filippo Di Luca Elisabetta Armiraglio Antonina Parafioriti Alessandro Luzzati Roberto Biagini Isabella Castiglioni Luca Maria Sconfienza |
| author_sort | Salvatore Gitto |
| collection | DOAJ |
| description | Summary: Background: Atypical cartilaginous tumour (ACT) and high-grade chondrosarcoma (CS) of long bones are respectively managed with active surveillance or curettage and wide resection. Our aim was to determine diagnostic performance of X-rays radiomics-based machine learning for classification of ACT and high-grade CS of long bones. Methods: This retrospective, IRB-approved study included 150 patients with surgically treated and histology-proven lesions at two tertiary bone sarcoma centres. At centre 1, the dataset was split into training (n = 71 ACT, n = 24 high-grade CS) and internal test (n = 19 ACT, n = 6 high-grade CS) cohorts, respectively, based on the date of surgery. At centre 2, the dataset constituted the external test cohort (n = 12 ACT, n = 18 high-grade CS). Manual segmentation was performed on frontal view X-rays, using MRI or CT for preliminary identification of lesion margins. After image pre-processing, radiomic features were extracted. Dimensionality reduction included stability, coefficient of variation, and mutual information analyses. In the training cohort, after class balancing, a machine learning classifier (Support Vector Machine) was automatically tuned using nested 10-fold cross-validation. Then, it was tested on both the test cohorts and compared to two musculoskeletal radiologists' performance using McNemar's test. Findings: Five radiomic features (3 morphology, 2 texture) passed dimensionality reduction. After tuning on the training cohort (AUC = 0.75), the classifier had 80%, 83%, 79% and 80%, 89%, 67% accuracy, sensitivity, and specificity in the internal (temporally independent) and external (geographically independent) test cohorts, respectively, with no difference compared to the radiologists (p ≥ 0.617). Interpretation: X-rays radiomics-based machine learning accurately differentiates between ACT and high-grade CS of long bones. Funding: AIRC Investigator Grant. |
| first_indexed | 2024-03-07T23:38:43Z |
| format | Article |
| id | doaj.art-d2142b4a810048c697d4e1b8ce36b919 |
| institution | Directory Open Access Journal |
| issn | 2352-3964 |
| language | English |
| last_indexed | 2024-03-07T23:38:43Z |
| publishDate | 2024-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | EBioMedicine |
| spelling | doaj.art-d2142b4a810048c697d4e1b8ce36b9192024-02-20T04:19:14ZengElsevierEBioMedicine2352-39642024-03-01101105018X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in contextSalvatore Gitto0Alessio Annovazzi1Kitija Nulle2Matteo Interlenghi3Christian Salvatore4Vincenzo Anelli5Jacopo Baldi6Carmelo Messina7Domenico Albano8Filippo Di Luca9Elisabetta Armiraglio10Antonina Parafioriti11Alessandro Luzzati12Roberto Biagini13Isabella Castiglioni14Luca Maria Sconfienza15IRCCS Istituto Ortopedico Galeazzi, Milan, Italy; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, ItalyNuclear Medicine Unit, IRCCS Regina Elena National Cancer Institute, Rome, ItalyRadiology Department, Riga East Clinical University Hospital, Riga, LatviaDeepTrace Technologies s.r.l., Milan, ItalyDeepTrace Technologies s.r.l., Milan, Italy; Department of Science, Technology and Society, University School for Advanced Studies IUSS Pavia, Pavia, ItalyRadiology and Diagnostic Imaging Unit, IRCCS Regina Elena National Cancer Institute, Rome, ItalyOncological Orthopaedics Unit, IRCCS Regina Elena National Cancer Institute, Rome, ItalyIRCCS Istituto Ortopedico Galeazzi, Milan, Italy; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, ItalyIRCCS Istituto Ortopedico Galeazzi, Milan, Italy; Dipartimento di Scienze Biomediche, Chirurgiche ed Odontoiatriche, Università degli Studi di Milano, Milan, ItalyScuola di Specializzazione in Radiodiagnostica, Università degli Studi di Milano, Milan, ItalyUOC Anatomia Patologica, ASST Gaetano Pini - CTO, Milan, ItalyUOC Anatomia Patologica, ASST Gaetano Pini - CTO, Milan, ItalyIRCCS Istituto Ortopedico Galeazzi, Milan, ItalyOncological Orthopaedics Unit, IRCCS Regina Elena National Cancer Institute, Rome, ItalyDepartment of Physics “G. Occhialini”, Università degli Studi di Milano-Bicocca, Milan, ItalyIRCCS Istituto Ortopedico Galeazzi, Milan, Italy; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, Italy; Corresponding author. Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano and IRCCS Istituto Ortopedico Galeazzi, via Cristina Belgioioso 173, 20157 Milan (MI), Italy.Summary: Background: Atypical cartilaginous tumour (ACT) and high-grade chondrosarcoma (CS) of long bones are respectively managed with active surveillance or curettage and wide resection. Our aim was to determine diagnostic performance of X-rays radiomics-based machine learning for classification of ACT and high-grade CS of long bones. Methods: This retrospective, IRB-approved study included 150 patients with surgically treated and histology-proven lesions at two tertiary bone sarcoma centres. At centre 1, the dataset was split into training (n = 71 ACT, n = 24 high-grade CS) and internal test (n = 19 ACT, n = 6 high-grade CS) cohorts, respectively, based on the date of surgery. At centre 2, the dataset constituted the external test cohort (n = 12 ACT, n = 18 high-grade CS). Manual segmentation was performed on frontal view X-rays, using MRI or CT for preliminary identification of lesion margins. After image pre-processing, radiomic features were extracted. Dimensionality reduction included stability, coefficient of variation, and mutual information analyses. In the training cohort, after class balancing, a machine learning classifier (Support Vector Machine) was automatically tuned using nested 10-fold cross-validation. Then, it was tested on both the test cohorts and compared to two musculoskeletal radiologists' performance using McNemar's test. Findings: Five radiomic features (3 morphology, 2 texture) passed dimensionality reduction. After tuning on the training cohort (AUC = 0.75), the classifier had 80%, 83%, 79% and 80%, 89%, 67% accuracy, sensitivity, and specificity in the internal (temporally independent) and external (geographically independent) test cohorts, respectively, with no difference compared to the radiologists (p ≥ 0.617). Interpretation: X-rays radiomics-based machine learning accurately differentiates between ACT and high-grade CS of long bones. Funding: AIRC Investigator Grant.http://www.sciencedirect.com/science/article/pii/S2352396424000537Artificial intelligenceAtypical cartilaginous tumourBone neoplasmChondrosarcomaRadiomics |
| spellingShingle | Salvatore Gitto Alessio Annovazzi Kitija Nulle Matteo Interlenghi Christian Salvatore Vincenzo Anelli Jacopo Baldi Carmelo Messina Domenico Albano Filippo Di Luca Elisabetta Armiraglio Antonina Parafioriti Alessandro Luzzati Roberto Biagini Isabella Castiglioni Luca Maria Sconfienza X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context EBioMedicine Artificial intelligence Atypical cartilaginous tumour Bone neoplasm Chondrosarcoma Radiomics |
| title | X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context |
| title_full | X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context |
| title_fullStr | X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context |
| title_full_unstemmed | X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context |
| title_short | X-rays radiomics-based machine learning classification of atypical cartilaginous tumour and high-grade chondrosarcoma of long bonesResearch in context |
| title_sort | x rays radiomics based machine learning classification of atypical cartilaginous tumour and high grade chondrosarcoma of long bonesresearch in context |
| topic | Artificial intelligence Atypical cartilaginous tumour Bone neoplasm Chondrosarcoma Radiomics |
| url | http://www.sciencedirect.com/science/article/pii/S2352396424000537 |
| work_keys_str_mv | AT salvatoregitto xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT alessioannovazzi xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT kitijanulle xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT matteointerlenghi xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT christiansalvatore xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT vincenzoanelli xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT jacopobaldi xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT carmelomessina xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT domenicoalbano xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT filippodiluca xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT elisabettaarmiraglio xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT antoninaparafioriti xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT alessandroluzzati xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT robertobiagini xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT isabellacastiglioni xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext AT lucamariasconfienza xraysradiomicsbasedmachinelearningclassificationofatypicalcartilaginoustumourandhighgradechondrosarcomaoflongbonesresearchincontext |