Christian Rubbert1, Christian Mathys1,2, Christiane Jockwitz3,4, Christian J Hartmann5,6, Simon B Eickhoff7,8, Felix Hoffstaedter7,8, Svenja Caspers3,9,10, Claudia R Eickhoff5,8, Benjamin Sigl1, Nikolas A Teichert1, Martin Südmeyer11, Bernd Turowski1, Alfons Schnitzler5,6, Julian Caspers1. 1. 1University Dusseldorf, Medical Faculty, Department of Diagnostic and Interventional Radiology, D-40225 Dusseldorf, Germany. 2. 2Institute of Radiology and Neuroradiology, Evangelisches Krankenhaus, University of Oldenburg, Germany. 3. 3Institute of Neuroscience and Medicine (INM-1) Research Centre Jülich, Jülich, Germany. 4. 4Department of Psychiatry, Psychotherapy and Psychosomatics, RWTH Aachen University, Medical Faculty, Aachen, Germany. 5. 5Institute of Clinical Neuroscience and Medical Psychology Medical Faculty Heinrich-Heine-University, Düsseldorf, Germany. 6. 6Department of Neurology, Center for Movement Disorders and Neuromodulation, Medical Faculty, Heinrich Heine University, Düsseldorf, Germany. 7. 7Institute for Systems Neuroscience, Medical Faculty, Heinrich-Heine University, Dusseldorf, Germany. 8. 8Institute of Neuroscience and Medicine (INM-7), Research Centre Jülich, Germany. 9. 9JARA-BRAIN Jülich-Aachen Research Alliance, Jülich, Germany. 10. 10Institute for Anatomy I, University Dusseldorf, Medical Faculty, Düsseldorf, Germany. 11. 11Department of Neurology, Ernst-von-Bergmann Klinikum, Potsdam, Germany.
Abstract
OBJECTIVE: Evaluation of a data-driven, model-based classification approach to discriminate idiopathic Parkinson's disease (PD) patients from healthy controls (HC) based on between-network connectivity in whole-brain resting-state functional MRI (rs-fMRI). METHODS: Whole-brain rs-fMRI (EPI, TR = 2.2 s, TE = 30 ms, flip angle = 90°. resolution = 3.1 × 3.1 × 3.1 mm, acquisition time ≈ 11 min) was assessed in 42 PD patients (medical OFF) and 47 HC matched for age and gender. Between-network connectivity based on full and L2-regularized partial correlation measures were computed for each subject based on canonical functional network architectures of two cohorts at different levels of granularity (Human Connectome Project: 15/25/50/100/200 networks; 1000BRAINS: 15/25/50/70 networks). A Boosted Logistic Regression model was trained on the correlation matrices using a nested cross-validation (CV) with 10 outer and 10 inner folds for an unbiased performance estimate, treating the canonical functional network architecture and the type of correlation as hyperparameters. The number of boosting iterations was fixed at 100. The model with the highest mean accuracy over the inner folds was trained using an non-nested 10-fold 20-repeats CV over the whole dataset to determine feature importance. RESULTS: Over the outer folds the mean accuracy was found to be 76.2% (median 77.8%, SD 18.2, IQR 69.4 - 87.1%). Mean sensitivity was 81% (median 80%, SD 21.1, IQR 75 - 100%) and mean specificity was 72.7% (median 75%, SD 20.4, IQR 66.7 - 80%). The 1000BRAINS 50-network-parcellation, using full correlations, performed best over the inner folds. The top features predominantly included sensorimotor as well as sensory networks. CONCLUSION: A rs-fMRI whole-brain-connectivity, data-driven, model-based approach to discriminate PD patients from healthy controls shows a very good accuracy and a high sensitivity. Given the high sensitivity of the approach, it may be of use in a screening setting. ADVANCES IN KNOWLEDGE: Resting-state functional MRI could prove to be a valuable, non-invasive neuroimaging biomarker for neurodegenerative diseases. The current model-based, data-driven approach on whole-brain between-network connectivity to discriminate Parkinson's disease patients from healthy controls shows promising results with a very good accuracy and a very high sensitivity.
OBJECTIVE: Evaluation of a data-driven, model-based classification approach to discriminate idiopathic Parkinson's disease (PD) patients from healthy controls (HC) based on between-network connectivity in whole-brain resting-state functional MRI (rs-fMRI). METHODS: Whole-brain rs-fMRI (EPI, TR = 2.2 s, TE = 30 ms, flip angle = 90°. resolution = 3.1 × 3.1 × 3.1 mm, acquisition time ≈ 11 min) was assessed in 42 PD patients (medical OFF) and 47 HC matched for age and gender. Between-network connectivity based on full and L2-regularized partial correlation measures were computed for each subject based on canonical functional network architectures of two cohorts at different levels of granularity (Human Connectome Project: 15/25/50/100/200 networks; 1000BRAINS: 15/25/50/70 networks). A Boosted Logistic Regression model was trained on the correlation matrices using a nested cross-validation (CV) with 10 outer and 10 inner folds for an unbiased performance estimate, treating the canonical functional network architecture and the type of correlation as hyperparameters. The number of boosting iterations was fixed at 100. The model with the highest mean accuracy over the inner folds was trained using an non-nested 10-fold 20-repeats CV over the whole dataset to determine feature importance. RESULTS: Over the outer folds the mean accuracy was found to be 76.2% (median 77.8%, SD 18.2, IQR 69.4 - 87.1%). Mean sensitivity was 81% (median 80%, SD 21.1, IQR 75 - 100%) and mean specificity was 72.7% (median 75%, SD 20.4, IQR 66.7 - 80%). The 1000BRAINS 50-network-parcellation, using full correlations, performed best over the inner folds. The top features predominantly included sensorimotor as well as sensory networks. CONCLUSION: A rs-fMRI whole-brain-connectivity, data-driven, model-based approach to discriminate PD patients from healthy controls shows a very good accuracy and a high sensitivity. Given the high sensitivity of the approach, it may be of use in a screening setting. ADVANCES IN KNOWLEDGE: Resting-state functional MRI could prove to be a valuable, non-invasive neuroimaging biomarker for neurodegenerative diseases. The current model-based, data-driven approach on whole-brain between-network connectivity to discriminate Parkinson's disease patients from healthy controls shows promising results with a very good accuracy and a very high sensitivity.
Authors: Andreas Horn; Martin Reich; Johannes Vorwerk; Ningfei Li; Gregor Wenzel; Qianqian Fang; Tanja Schmitz-Hübsch; Robert Nickl; Andreas Kupsch; Jens Volkmann; Andrea A Kühn; Michael D Fox Journal: Ann Neurol Date: 2017-07 Impact factor: 10.422
Authors: C Salvatore; A Cerasa; I Castiglioni; F Gallivanone; A Augimeri; M Lopez; G Arabia; M Morelli; M C Gilardi; A Quattrone Journal: J Neurosci Methods Date: 2013-11-26 Impact factor: 2.390
Authors: Julian Caspers; Christian Mathys; Felix Hoffstaedter; Martin Südmeyer; Edna C Cieslik; Christian Rubbert; Christian J Hartmann; Claudia R Eickhoff; Kathrin Reetz; Christian Grefkes; Jochen Michely; Bernd Turowski; Alfons Schnitzler; Simon B Eickhoff Journal: Front Hum Neurosci Date: 2017-05-30 Impact factor: 3.169
Authors: Alexandra Abós; Hugo C Baggio; Bàrbara Segura; Anna I García-Díaz; Yaroslau Compta; Maria José Martí; Francesc Valldeoriola; Carme Junqué Journal: Sci Rep Date: 2017-03-28 Impact factor: 4.379
Authors: Gemma C Monte-Rubio; Barbara Segura; Antonio P Strafella; Thilo van Eimeren; Naroa Ibarretxe-Bilbao; Maria Diez-Cirarda; Carsten Eggers; Olaia Lucas-Jiménez; Natalia Ojeda; Javier Peña; Marina C Ruppert; Roser Sala-Llonch; Hendrik Theis; Carme Uribe; Carme Junque Journal: Hum Brain Mapp Date: 2022-03-19 Impact factor: 5.399
Authors: Julián Benito-León; Elan D Louis; Virginia Mato-Abad; Alvaro Sánchez-Ferro; Juan P Romero; Michele Matarazzo; J Ignacio Serrano Journal: Ann Clin Transl Neurol Date: 2019-11-26 Impact factor: 4.511
Authors: Julian Caspers; Christian Rubbert; Simon B Eickhoff; Felix Hoffstaedter; Martin Südmeyer; Christian J Hartmann; Benjamin Sigl; Nikolas Teichert; Joel Aissa; Bernd Turowski; Alfons Schnitzler; Christian Mathys Journal: Neuroradiology Date: 2021-05-21 Impact factor: 2.804