scholarly journals Radiological identification of temporal lobe epilepsy using artificial intelligence: a feasibility study

Author(s):  
Ezequiel Gleichgerrcht ◽  
Brent Munsell ◽  
Simon Keller ◽  
Daniel L Drane ◽  
Jens H Jensen ◽  
...  

Abstract Temporal lobe epilepsy is associated with magnetic resonance imaging (MRI) findings reflecting underlying mesial temporal sclerosis. Identifying these MRI features is critical for the diagnosis and management of temporal lobe epilepsy. To date, this process relies on visual assessment by highly trained human experts (e.g. neuroradiologists, epileptologists). Artificial intelligence is increasingly recognized as a promising aid in the radiological evaluation of neurological diseases, yet its applications in temporal lobe epilepsy have been limited. Here, we applied a convolutional neural networks to assess the classification accuracy of temporal lobe epilepsy based on structural MRI. We demonstrate that convoluted neural networks can achieve high accuracy in the identification of unilateral temporal lobe epilepsy cases even when the MRI had been originally interpreted as normal by experts. We show that accuracy can be potentiated by employing smoothed gray matter maps and a direct acyclic graphs approach. We further discuss the foundations for the development of computer-aided tools to assist with the diagnosis of epilepsy.

2006 ◽  
Vol 12 (3) ◽  
pp. 155-160 ◽  
Author(s):  
Márcio Andriani Rahal ◽  
Gerardo Maria de Araújo Filho ◽  
Luís Otávio Sales Ferreira Caboclo ◽  
Vivianne Pellegrino Rosa ◽  
Ricardo da Silva Centeno ◽  
...  

INTRODUCTION: Somatosensory auras (SSAs) are more usually described in patients with parietal lobe epilepsy (PLE), producing more commonly a localized cutaneous tingling sensation, involving hands and fingers followed by tonic or clonic focal seizures. These usually originate in the contralateral hemisphere. Etiology includes dysplasias, tumours, ischemic or postencephalitic gliosis. However, other focal epilepsies, such as frontal and temporal, may also originate SSAs. Although this type of aura is reported as rare in patients with mesial temporal lobe epilepsy (MTLE), this association has not been systematically studied. OBJECTIVES: The aim of this article was to describe the cases of four patients with refractory MTLE and SSAs, reporting their clinical characteristics and MRI findings. We discuss the localizing and lateralizing value of SSAs, particularly in the context of MTLE. METHODS AND RESULTS: Four patients with refractory MTLE and SSAs followed-up in the outpatient's clinic at the Epilepsy Section, Universidade Federal de São Paulo, were submitted to presurgical evaluation and corticoamygdalohippocampectomy. MRI in all cases showed unilateral mesial temporal sclerosis (MTS). Regarding seizure semiology, tingling sensation involving the upper extremity was the most prevalent symptom. Three of the four patients had SSAs contralateral to the MTS. Following the SSAs all patients most of the time presented other symptoms such as autonomic or psychic auras evolving to psychomotor seizures. After surgical treatment, two of the patients presented infrequent auras, and two were rendered seizure-free. CONCLUSION: Although rare, SSAs can be present in MTLE. The characteristics of autonomic or psychic auras, psychomotor seizures, neuropsychological deficits, and typical neurophysiologic and MRI findings may help differentiate patients with MTLE from those with PLE.


2013 ◽  
Vol 71 (3) ◽  
pp. 183-190 ◽  
Author(s):  
Kette D. R. Valente ◽  
Geraldo Busatto Filho

The relationship between depression and epilepsy has been known since ancient times, however, to date, it is not fully understood. The prevalence of psychiatric disorders in persons with epilepsy is high compared to general population. It is assumed that the rate of depression ranges from 20 to 55% in patients with refractory epilepsy, especially considering those with temporal lobe epilepsy caused by mesial temporal sclerosis. Temporal lobe epilepsy is a good biological model to understand the common structural basis between depression and epilepsy. Interestingly, mesial temporal lobe epilepsy and depression share a similar neurocircuitry involving: temporal lobes with hippocampus, amygdala and entorhinal and neocortical cortex; the frontal lobes with cingulate gyrus; subcortical structures, such as basal ganglia and thalamus; and the connecting pathways. We provide clinical and brain structural evidences that depression and epilepsy represent an epiphenomenon sharing similar neural networks.


2012 ◽  
Vol 117 (5) ◽  
pp. 962-969 ◽  
Author(s):  
Fernando L. Vale ◽  
Ali M. Bozorg ◽  
Mike R. Schoenberg ◽  
Kondi Wong ◽  
Thomas C. Witt

Epilepsy surgery is an effective treatment for medically resistant temporal lobe epilepsy (TLE). To minimize complication rates and potentially improve neuropsychology outcomes, stereotactic radiosurgery (SRS) has been explored as an alternative. Two pilot trials have demonstrated the effectiveness of SRS for the treatment of medically resistant TLE, with seizure-free outcomes for approximately 65% of patients at last follow-up. Despite encouraging results, no conclusive long-term outcomes are available for SRS. This article discusses a single patient who presented with recurrent seizures, worsening headaches, and persistent abnormal MRI findings 7 years and 8 months after SRS. This 29-year-old woman with a history of medically refractory complex partial seizures since childhood was referred for evaluation. Medical management had failed in this patient. The workup was compatible with left mesial temporal lobe onset, with MRI findings suggestive of mesial temporal sclerosis. In 2003, at the age of 23 years, she underwent Gamma Knife surgery (GKS) targeting the left temporal mesial area with a dose of 24 Gy at the 50% marginal isodose line. After GKS, the patient's seizures decreased in frequency over several months, but auras were persistent. Nine months after treatment, she developed worsening headaches. A follow-up MRI study demonstrated a thick, irregular, enhancing lesion in the medial part of the temporal lobe. She was placed on corticosteroids, with resolution of her headaches. Her seizures and headaches recurred in March 2010. An MRI study showed a 2.2-cm, ill-defined, enhancing cystic lesion in the left mesial temporal lobe with T2 and FLAIR hyperintensity, which was presumably radiation induced. At that time, the patient opted for left temporal lobe resection to control her seizures. Histological examination showed moderately severe, remote, longstanding sclerosis at the level of the hippocampus. A vascular lesion was identified, and it was most consistent with radiation-induced capillary hemangioma. The entorhinal region was severely damaged, with hemorrhage, necrosis, neuronal loss, astrogliosis, and hemosiderin deposition. There was evidence of radiation vasculopathy. Radiation-induced lesions after SRS for the treatment of epilepsy are not well documented. Although GKS is a promising technique for the treatment of medically resistant TLE, the ideal candidate is not yet well defined. The selection of the appropriate technical parameters to obtain a desirable functional effect without histological damage to the surrounding neural tissue remains a challenge. This case illustrates the need for long-term follow-up when radiosurgery is used for epilepsy.


2019 ◽  
Vol 26 (07) ◽  
pp. 1009-1013
Author(s):  
Abdul Raouf ◽  
Saba Jehangir ◽  
Asma Shoukat ◽  
Muhammad Ahsan

Mesial temporal sclerosis (MTS) is the most common pathology in patients undergoing anterior temporal lobectomy. Magnetic resonance imaging (MRI) is valuable in detecting MTS. Reduced hippocampal volume and elevated T2 signal are associated with MTS, and both quantitative T2 and volumetric measurementshave been associated with hippocampal cellular loss that characterizes this condition. Objectives: To determine the accuracy of hippocampal quantitative (T2 relaxometric) assessment in diagnosing hippocampal atrophy in patients with temporal lobe epilepsy by comparing it with qualitative (visual) assessment on MRI. Study Design: Cross sectional study. Setting: Radiology department of Allied Hospital Faisalabad. Period: 12 months from theapproval from Sep, 2016 to Dec, 2017. Subjects & Methods: After taking permission from hospital ethical committee, and written informed consent, patients with history of temporal lobe epilepsy and EEG findings consistent with temporal lobe epilepsy were examined on 1.5 Tesla Achieva philips scanner, visual assessment and T2 relaxometry. Section of the hippocampus head was defined as the first in which it was possible to see the temporal horn of the lateral ventricle and therefore to appropriately separate the hippocampal formation from the amygdala. The body of the hippocampus defined in the fourth coronal section after the region of interest of the hippocampus head, and the tail was defined in the third coronal section after the hippocampus body, in which it is also possible to visualize the quadrigeminal plate (section of 5mm).Visually the images were assessed and MRI examination was done. All the data was collected on a performa. Results: We concluded that the frequency of accuracy of hippocampal quantitative (t2 relaxometric) assessment in diagnosing hippocampal atrophy in patients with temporal lobe epilepsy by comparing it with qualitative (visual) assessment on MRIis high but needs validation through some-other studies. Conclusion: We concluded that the frequency of accuracy of hippocampal quantitative (t2 relaxometric) assessment in diagnosing hippocampal atrophy in patients with temporal lobe epilepsy by comparing it with qualitative (visual) assessment on MRI is high but needs validation through some-other studies. 


Brain ◽  
2001 ◽  
Vol 124 (1) ◽  
pp. 167-175 ◽  
Author(s):  
N. F. Moran ◽  
L. Lemieux ◽  
N. D. Kitchen ◽  
D. R. Fish ◽  
S. D. Shorvon

2015 ◽  
Vol 25 (6) ◽  
pp. 1030-1037 ◽  
Author(s):  
Michael Y. Xu ◽  
Erhan Ergene ◽  
Michael Zagardo ◽  
Patrick T. Tracy ◽  
Huaping Wang ◽  
...  

2015 ◽  
Vol 74 (1) ◽  
pp. 35-43 ◽  
Author(s):  
Fábio A. Nascimento ◽  
Luana Antunes Maranha Gatto ◽  
Carlos Silvado ◽  
Maria Joana Mäder-Joaquim ◽  
Marlus Sidney Moro ◽  
...  

ABSTRACT Objective To contribute our experience with surgical treatment of patients with mesial temporal lobe epilepsy (mTLE) undergoing anterior temporal lobectomy (ATL) or selective amygdalohippocampectomy (SelAH). Method This is a retrospective observational study. The sample included patients with medically refractory mTLE due to unilateral mesial temporal sclerosis who underwent either ATL or SelAH, at Hospital de Clinicas – UFPR, from 2005 to 2012. We report seizure outcomes, using Engel classification, cognitive outcomes, using measurements of verbal and visuospatial memories, as well as operative complications. Result Sixty-seven patients (33 ATL, 34 SelAH) were studied; median follow-up was 64 months. There was no statistically significant difference in seizure or neuropsychological outcomes, although verbal memory was more negatively affected in ATL operations on patients’ dominant hemispheres. Higher number of major complications was observed in the ATL group (p = 0.004). Conclusion Seizure and neuropsychological outcomes did not differ. ATL appeared to be associated with higher risk of complications.


Epilepsia ◽  
2010 ◽  
Vol 51 (6) ◽  
pp. 1024-1029 ◽  
Author(s):  
Michael Murphy ◽  
Paul D. Smith ◽  
Martin Wood ◽  
Stephen Bowden ◽  
Terence J. O’Brien ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document