Rose Dawn Bharath1,2, Rajanikant Panda1,2,3, Jeetu Raj4, Sujas Bhardwaj1,2,5, Sanjib Sinha5, Ganne Chaitanya5,6, Kenchaiah Raghavendra5, Ravindranadh C Mundlamuri5, Arivazhagan Arimappamagan7, Malla Bhaskara Rao7, Jamuna Rajeshwaran8, Kandavel Thennarasu9, Kaushik K Majumdar10, Parthasarthy Satishchandra7, Tapan K Gandhi11. 1. Neuroimaging and Interventional Radiology, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 2. Advance Brain Imaging Facility, Cognitive Neuroscience Centre, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 3. Coma Science Group, GIGA-Consciousness, Universitè de Liège, Liège, Belgium. 4. Department of Computer Science, Indian Institute of Technology Delhi, New Delhi, Delhi, 110016, India. 5. Neurology, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 6. Department of Neurology, Thomas Jefferson University, Philadelphia, PA, USA. 7. Neurosurgery, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 8. Neuropsychology, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 9. Biostatistics, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, 560029, India. 10. Systems Science and Informatics Unit, Indian Statistical Institute, Bangalore, Karnataka, 560059, India. 11. Department of Electrical Engineering, Indian Institute of Technology Delhi, (IIT-D), New Delhi, Delhi, 110016, India. cns.researchers1@gmail.com.
Abstract
OBJECTIVES: Experimental models have provided compelling evidence for the existence of neural networks in temporal lobe epilepsy (TLE). To identify and validate the possible existence of resting-state "epilepsy networks," we used machine learning methods on resting-state functional magnetic resonance imaging (rsfMRI) data from 42 individuals with TLE. METHODS: Probabilistic independent component analysis (PICA) was applied to rsfMRI data from 132 subjects (42 TLE patients + 90 healthy controls) and 88 independent components (ICs) were obtained following standard procedures. Elastic net-selected features were used as inputs to support vector machine (SVM). The strengths of the top 10 networks were correlated with clinical features to obtain "rsfMRI epilepsy networks." RESULTS: SVM could classify individuals with epilepsy with 97.5% accuracy (sensitivity = 100%, specificity = 94.4%). Ten networks with the highest ranking were found in the frontal, perisylvian, cingulo-insular, posterior-quadrant, thalamic, cerebello-thalamic, and temporo-thalamic regions. The posterior-quadrant, cerebello-thalamic, thalamic, medial-visual, and perisylvian networks revealed significant correlation (r > 0.40) with age at onset of seizures, the frequency of seizures, duration of illness, and a number of anti-epileptic drugs. CONCLUSIONS: IC-derived rsfMRI networks contain epilepsy-related networks and machine learning methods are useful in identifying these networks in vivo. Increased network strength with disease progression in these "rsfMRI epilepsy networks" could reflect epileptogenesis in TLE. KEY POINTS: • ICA of resting-state fMRI carries disease-specific information about epilepsy. • Machine learning can classify these components with 97.5% accuracy. • "Subject-specific epilepsy networks" could quantify "epileptogenesis" in vivo.
OBJECTIVES: Experimental models have provided compelling evidence for the existence of neural networks in temporal lobe epilepsy (TLE). To identify and validate the possible existence of resting-state "epilepsy networks," we used machine learning methods on resting-state functional magnetic resonance imaging (rsfMRI) data from 42 individuals with TLE. METHODS: Probabilistic independent component analysis (PICA) was applied to rsfMRI data from 132 subjects (42 TLEpatients + 90 healthy controls) and 88 independent components (ICs) were obtained following standard procedures. Elastic net-selected features were used as inputs to support vector machine (SVM). The strengths of the top 10 networks were correlated with clinical features to obtain "rsfMRI epilepsy networks." RESULTS: SVM could classify individuals with epilepsy with 97.5% accuracy (sensitivity = 100%, specificity = 94.4%). Ten networks with the highest ranking were found in the frontal, perisylvian, cingulo-insular, posterior-quadrant, thalamic, cerebello-thalamic, and temporo-thalamic regions. The posterior-quadrant, cerebello-thalamic, thalamic, medial-visual, and perisylvian networks revealed significant correlation (r > 0.40) with age at onset of seizures, the frequency of seizures, duration of illness, and a number of anti-epileptic drugs. CONCLUSIONS: IC-derived rsfMRI networks contain epilepsy-related networks and machine learning methods are useful in identifying these networks in vivo. Increased network strength with disease progression in these "rsfMRI epilepsy networks" could reflect epileptogenesis in TLE. KEY POINTS: • ICA of resting-state fMRI carries disease-specific information about epilepsy. • Machine learning can classify these components with 97.5% accuracy. • "Subject-specific epilepsy networks" could quantify "epileptogenesis" in vivo.
Entities:
Keywords:
Magnetic resonance imaging; Seizures; Support vector machine; Temporal lobe epilepsy
Authors: Taha Gholipour; Xiaozhen You; Steven M Stufflebeam; Murray Loew; Mohamad Z Koubeissi; Victoria L Morgan; William D Gaillard Journal: Epilepsia Date: 2022-01-04 Impact factor: 6.740
Authors: Marianna La Rocca; Rachael Garner; Nicola Amoroso; Evan S Lutkenhoff; Martin M Monti; Paul Vespa; Arthur W Toga; Dominique Duncan Journal: Front Neurosci Date: 2020-11-30 Impact factor: 4.677