Michaela Cellina1, Marta Pirovano2, Matteo Ciocca2, Daniele Gibelli3, Chiara Floridi4, Giancarlo Oliva5. 1. Radiology Department, ASST Fatebenefratelli Sacco, Piazza Principessa Clotilde 3, 20123, Milan, Italy. michaela.cellina@asst-fbf-sacco.it. 2. Neurology Department, ASST Fatebenefratelli Sacco, Piazza Principessa Clotilde 3, 20123, Milan, Italy. 3. Dipartimento Di Scienze Biomediche Per La Salute, Università Degli Studi Di Milano, Via Mangiagalli 31, 20121, Milan, Italy. 4. Department of Radiology - Division of Special and Pediatric Radiology, University Hospital "Umberto I - Lancisi - Salesi", Via Conca 71, 60126, Ancona, AN, Italy. 5. Radiology Department, ASST Fatebenefratelli Sacco, Piazza Principessa Clotilde 3, 20123, Milan, Italy.
Abstract
OBJECTIVE: Retinal nerve fiber layer thickness (RNFL) is a biomarker of neuroaxonal loss and index of visual function in multiple sclerosis (MS). We aimed to assess the correlation between radiomic features and RNFL, visual acuity (VA) at patients' presentation, visual outcome (VO), and clinical diagnosis. METHODS: We reviewed imaging and clinical data of 25 patients with a first episode of optic neuritis (ON) (14 females, 11 males; 5 bilateral ON; 7 left ON; 13 right ON). All patients underwent a complete ophthalmological assessment, including visual acuity and RNFL, neurological evaluation, orbits MRI. Segmentation of the optic nerves was performed through 3D slicer open software to get radiomics analysis. All patients underwent a complete neuro-ophthalmological follow-up at 6 months to assess the VO, classified as: complete recovery, partial recovery, deficit persistence/relapse, or visual worsening and were diagnosed as MS or clinically isolated syndrome. RESULTS: We observed significant correlations between radiomic features and RNFL and between radiomic features and VA. Regression model analysis identified 1 radiomic feature with significant association with VO (Gray Level non-uniformity Normalized, p = 0.004) and 6 radiomic features with significant correlation with diagnosis (High Gray Level Zone Emphasis, p < 0.001; Entropy, p < 0.001, for T1 segmentation; Mean Absolute Deviation, p < 0.001; Coarseness < 0.001; Small Area Low Gray Level Emphasis, p < 0.001; Contrast, p = 0.008, for STIR segmentation). CONCLUSION: Orbits MRI analysis at the first episode of ON has the potential to assess the visual function and VO in ON patients, and predict MS development.
OBJECTIVE: Retinal nerve fiber layer thickness (RNFL) is a biomarker of neuroaxonal loss and index of visual function in multiple sclerosis (MS). We aimed to assess the correlation between radiomic features and RNFL, visual acuity (VA) at patients' presentation, visual outcome (VO), and clinical diagnosis. METHODS: We reviewed imaging and clinical data of 25 patients with a first episode of optic neuritis (ON) (14 females, 11 males; 5 bilateral ON; 7 left ON; 13 right ON). All patients underwent a complete ophthalmological assessment, including visual acuity and RNFL, neurological evaluation, orbits MRI. Segmentation of the optic nerves was performed through 3D slicer open software to get radiomics analysis. All patients underwent a complete neuro-ophthalmological follow-up at 6 months to assess the VO, classified as: complete recovery, partial recovery, deficit persistence/relapse, or visual worsening and were diagnosed as MS or clinically isolated syndrome. RESULTS: We observed significant correlations between radiomic features and RNFL and between radiomic features and VA. Regression model analysis identified 1 radiomic feature with significant association with VO (Gray Level non-uniformity Normalized, p = 0.004) and 6 radiomic features with significant correlation with diagnosis (High Gray Level Zone Emphasis, p < 0.001; Entropy, p < 0.001, for T1 segmentation; Mean Absolute Deviation, p < 0.001; Coarseness < 0.001; Small Area Low Gray Level Emphasis, p < 0.001; Contrast, p = 0.008, for STIR segmentation). CONCLUSION: Orbits MRI analysis at the first episode of ON has the potential to assess the visual function and VO in ON patients, and predict MS development.
Entities:
Keywords:
Optic nerve; Optic neuritis; Orbit magnetic resonance; RNFL; Radiomics
Authors: Madhan Kolappan; Andrew P D Henderson; Thomas M Jenkins; Claudia A M Wheeler-Kingshott; Gordon T Plant; Alan J Thompson; David H Miller Journal: J Neurol Date: 2009-03-18 Impact factor: 4.849
Authors: Thomas M Jenkins; Ahmed T Toosy; Olga Ciccarelli; Katherine A Miszkiel; Claudia A Wheeler-Kingshott; Andrew P Henderson; Constantinos Kallis; Laura Mancini; Gordon T Plant; David H Miller; Alan J Thompson Journal: Ann Neurol Date: 2010-01 Impact factor: 10.422
Authors: Axel Petzold; Mike P Wattjes; Fiona Costello; Jose Flores-Rivera; Clare L Fraser; Kazuo Fujihara; Jacqueline Leavitt; Romain Marignier; Friedemann Paul; Sven Schippling; Christian Sindic; Pablo Villoslada; Brian Weinshenker; Gordon T Plant Journal: Nat Rev Neurol Date: 2014-07-08 Impact factor: 42.937
Authors: Sebastian Berg; Iris Kaschka; Kathrin S Utz; Konstantin Huhn; Alexandra Lämmer; Robert Lämmer; Anne Waschbisch; Stephan Kloska; De-Hyung Lee; Arnd Doerfler; Ralf A Linker Journal: PLoS One Date: 2015-01-30 Impact factor: 3.240
Authors: Simon J Hickman; Ahmed T Toosy; Katherine A Miszkiel; Stephen J Jones; Daniel R Altmann; David G MacManus; Gordon T Plant; Alan J Thompson; David H Miller Journal: J Neurol Date: 2004-08 Impact factor: 4.849
Authors: Michaela Cellina; Maurizio Cè; Giovanni Irmici; Carlo Martinenghi; Sara Marziali; Natallia Khenkina; Anna Maria Ierardi; Giancarlo Oliva; Gianpaolo Carrafiello Journal: Emerg Radiol Date: 2022-04-14
Authors: Vincenza Granata; Roberta Grassi; Roberta Fusco; Andrea Belli; Carmen Cutolo; Silvia Pradella; Giulia Grazzini; Michelearcangelo La Porta; Maria Chiara Brunese; Federica De Muzio; Alessandro Ottaiano; Antonio Avallone; Francesco Izzo; Antonella Petrillo Journal: Infect Agent Cancer Date: 2021-07-19 Impact factor: 2.965