MRI radiomics-based machine learning for classification of deep-seated lipoma and atypical lipomatous tumor of the extremities.

Autor: Gitto S; IRCCS Istituto Ortopedico Galeazzi, Milan, Italy.; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, Italy., Interlenghi M; DeepTrace Technologies, Milan, Italy., Cuocolo R; Department of Medicine, Surgery and Dentistry, University of Salerno, Baronissi, Italy.; Augmented Reality for Health Monitoring Laboratory (ARHeMLab), Department of Electrical Engineering and Information Technology, University of Naples 'Federico II', Naples, Italy., Salvatore C; DeepTrace Technologies, Milan, Italy.; Department of Science, Technology and Society, University School for Advanced Studies IUSS Pavia, Pavia, Italy., Giannetta V; Diagnostic and Interventional Radiology Department, IRCCS Ospedale San Raffaele-Turro, Università Vita-Salute San Raffaele, Milan, Italy., Badalyan J; Scuola di Specializzazione in Statistica Sanitaria e Biometria, Università Degli Studi Di Milano, Milan, Italy., Gallazzi E; UOC Patologia Vertebrale e Scoliosi, ASST Gaetano Pini - CTO, Milan, Italy., Spinelli MS; UOC Ortopedia Oncologica, ASST Gaetano Pini - CTO, Milan, Italy., Gallazzi M; UOC Radiodiagnostica, ASST Gaetano Pini - CTO, Milan, Italy., Serpi F; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, Italy., Messina C; IRCCS Istituto Ortopedico Galeazzi, Milan, Italy.; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, Italy., Albano D; IRCCS Istituto Ortopedico Galeazzi, Milan, Italy., Annovazzi A; Nuclear Medicine Unit, IRCCS Regina Elena National Cancer Institute, Rome, Italy., Anelli V; Radiology and Diagnostic Imaging Unit, IRCCS Regina Elena National Cancer Institute, Rome, Italy., Baldi J; Oncological Orthopaedics Unit, IRCCS Regina Elena National Cancer Institute, Rome, Italy., Aliprandi A; Istituti Clinici Zucchi, Monza, Italy., Armiraglio E; UOC Anatomia Patologica, ASST Gaetano Pini - CTO Milan, Milan, Italy., Parafioriti A; UOC Anatomia Patologica, ASST Gaetano Pini - CTO Milan, Milan, Italy., Daolio PA; UOC Ortopedia Oncologica, ASST Gaetano Pini - CTO, Milan, Italy., Luzzati A; IRCCS Istituto Ortopedico Galeazzi, Milan, Italy., Biagini R; Oncological Orthopaedics Unit, IRCCS Regina Elena National Cancer Institute, Rome, Italy., Castiglioni I; Department of Physics, Università degli Studi di Milano-Bicocca, Milan, Italy.; Institute of Biomedical Imaging and Physiology, Consiglio Nazionale Delle Ricerche, Segrate, Italy., Sconfienza LM; IRCCS Istituto Ortopedico Galeazzi, Milan, Italy. io@lucasconfienza.it.; Dipartimento di Scienze Biomediche per la Salute, Università degli Studi di Milano, Milan, Italy. io@lucasconfienza.it.
Jazyk: angličtina
Zdroj: La Radiologia medica [Radiol Med] 2023 Aug; Vol. 128 (8), pp. 989-998. Date of Electronic Publication: 2023 Jun 19.
DOI: 10.1007/s11547-023-01657-y
Abstrakt: Purpose: To determine diagnostic performance of MRI radiomics-based machine learning for classification of deep-seated lipoma and atypical lipomatous tumor (ALT) of the extremities.
Material and Methods: This retrospective study was performed at three tertiary sarcoma centers and included 150 patients with surgically treated and histology-proven lesions. The training-validation cohort consisted of 114 patients from centers 1 and 2 (n = 64 lipoma, n = 50 ALT). The external test cohort consisted of 36 patients from center 3 (n = 24 lipoma, n = 12 ALT). 3D segmentation was manually performed on T1- and T2-weighted MRI. After extraction and selection of radiomic features, three machine learning classifiers were trained and validated using nested fivefold cross-validation. The best-performing classifier according to previous analysis was evaluated and compared to an experienced musculoskeletal radiologist in the external test cohort.
Results: Eight features passed feature selection and were incorporated into the machine learning models. After training and validation (74% ROC-AUC), the best-performing classifier (Random Forest) showed 92% sensitivity and 33% specificity in the external test cohort with no statistical difference compared to the radiologist (p = 0.474).
Conclusion: MRI radiomics-based machine learning may classify deep-seated lipoma and ALT of the extremities with high sensitivity and negative predictive value, thus potentially serving as a non-invasive screening tool to reduce unnecessary referral to tertiary tumor centers.
(© 2023. The Author(s).)
Databáze: MEDLINE
Externí odkaz: