Alberto Stefano Tagliafico1, Michele Cea2, Federica Rossi3, Francesca Valdora4, Bianca Bignotti5, Giulia Succio6, Stefano Gualco7, Alessio Conte8, Alida Dominietto9. 1. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy; Ospedale Policlinico San Martino, Largo R. Benzi 10, 16132 Genoa, Italy. Electronic address: alberto.tagliafico@unige.it. 2. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy; Ospedale Policlinico San Martino, Largo R. Benzi 10, 16132 Genoa, Italy. Electronic address: cea.michele@unige.it. 3. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy. Electronic address: federossi0590@gmail.com. 4. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy. Electronic address: valdorafrancesca@gmail.com. 5. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy; Ospedale Policlinico San Martino, Largo R. Benzi 10, 16132 Genoa, Italy. Electronic address: bignottibianca@gmail.com. 6. Ospedale Policlinico San Martino, Largo R. Benzi 10, 16132 Genoa, Italy. Electronic address: giulia.succio@hsanmartino.it. 7. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy. Electronic address: stefanogualco@gmail.com. 8. Department of Health Sciences, University of Genoa, Via A. Pastore 1, 16132 Genoa, Italy. Electronic address: alessioconte.1995@gmail.com. 9. Ospedale Policlinico San Martino, Largo R. Benzi 10, 16132 Genoa, Italy. Electronic address: alida.dominietto@hsanmartino.it.
Abstract
PURPOSE: Focal pattern in multiple myeloma (MM) seems to be related to poorer survival and differentiation from diffuse to focal pattern on computed tomography (CT) has inter-reader variability. We postulated that a Radiomic approach could help radiologists in differentiating diffuse from focal patterns on CT. METHODS: We retrospectively reviewed imaging data of 70 patients with MM with CT, PET-CT or MRI available before bone marrow transplant. Two general radiologist evaluated, in consensus, CT images to define a focal (at least one lytic lesion >5 mm in diameter) or a diffuse (lesions <5 mm, not osteoporosis) pattern. N = 104 Radiomics features were extracted and evaluated with an open source software. RESULTS: The pathological group included: 22 diffuse and 39 focal patterns. After feature reduction, 9 features were different (p < 0.05) in the diffuse and focal patterns (n = 2/9 features were Shape-based: MajorAxisLength and Sphericity; n = 7/9 were Gray Level Run Length Matrix (Glrlm)). AUC of the Radiologists versus Reference Standard was 0.64 (95 % CI: (0.49-0.78) p = 0.20. AUC of the best 4 features (MajorAxisLength, Median, SizeZoneNonUniformity, ZoneEntropy) were: 0.73 (95 % CI: 0.58-0.88); 0.71 (95 % CI: 0.54-0.88); 0.79 (95 % CI: 0.66-0.92); 0.68 (95 % CI: 0.53-0.83) respectively. CONCLUSION: A Radiomics approach improves radiological evaluation of focal and diffuse pattern of MM on CT.
PURPOSE: Focal pattern in multiple myeloma (MM) seems to be related to poorer survival and differentiation from diffuse to focal pattern on computed tomography (CT) has inter-reader variability. We postulated that a Radiomic approach could help radiologists in differentiating diffuse from focal patterns on CT. METHODS: We retrospectively reviewed imaging data of 70 patients with MM with CT, PET-CT or MRI available before bone marrow transplant. Two general radiologist evaluated, in consensus, CT images to define a focal (at least one lytic lesion >5 mm in diameter) or a diffuse (lesions <5 mm, not osteoporosis) pattern. N = 104 Radiomics features were extracted and evaluated with an open source software. RESULTS: The pathological group included: 22 diffuse and 39 focal patterns. After feature reduction, 9 features were different (p < 0.05) in the diffuse and focal patterns (n = 2/9 features were Shape-based: MajorAxisLength and Sphericity; n = 7/9 were Gray Level Run Length Matrix (Glrlm)). AUC of the Radiologists versus Reference Standard was 0.64 (95 % CI: (0.49-0.78) p = 0.20. AUC of the best 4 features (MajorAxisLength, Median, SizeZoneNonUniformity, ZoneEntropy) were: 0.73 (95 % CI: 0.58-0.88); 0.71 (95 % CI: 0.54-0.88); 0.79 (95 % CI: 0.66-0.92); 0.68 (95 % CI: 0.53-0.83) respectively. CONCLUSION: A Radiomics approach improves radiological evaluation of focal and diffuse pattern of MM on CT.
Authors: Yang Li; Yang Liu; Ping Yin; Chuanxi Hao; Chao Sun; Lei Chen; Sicong Wang; Nan Hong Journal: Front Oncol Date: 2021-12-01 Impact factor: 6.244