Abstract
Purpose: We are aiming to build a supervised machine learning-based classifier, in order to preoperatively distinguish focal cortical dysplasia (FCD) from glioneuronal tumors (GNTs) in patients with epilepsy.Methods: This retrospective study was comprised of 96 patients who underwent epilepsy surgery, with the final neuropathologic diagnosis of either an FCD or GNTs. Seven classical machine learning algorithms (i.e., Random Forest, SVM, Decision Tree, Logistic Regression, XGBoost, LightGBM, and CatBoost) were employed and trained by our dataset to get the classification model. Ten features [i.e., Gender, Past history, Age at seizure onset, Course of disease, Seizure type, Seizure frequency, Scalp EEG biomarkers, MRI features, Lesion location, Number of antiepileptic drug (AEDs)] were analyzed in our study.Results: We enrolled 56 patients with FCD and 40 patients with GNTs, which included 29 with gangliogliomas (GGs) and 11 with dysembryoplasic neuroepithelial tumors (DNTs). Our study demonstrated that the Random Forest-based machine learning model offered the best predictive performance on distinguishing the diagnosis of FCD from GNTs, with an F1-score of 0.9180 and AUC value of 0.9340. Furthermore, the most discriminative factor between FCD and GNTs was the feature “age at seizure onset” with the Chi-square value of 1,213.0, suggesting that patients who had a younger age at seizure onset were more likely to be diagnosed as FCD.Conclusion: The Random Forest-based machine learning classifier can accurately differentiate FCD from GNTs in patients with epilepsy before surgery. This might lead to improved clinician confidence in appropriate surgical planning and treatment outcomes.
Highlights
Focal cortical dysplasia (FCD) is a distinctive malformation of cortical development that is highly associated with refractory epilepsy
Inclusion criteria were as follows: [1] patients were diagnosed as epilepsy according to the guidelines for the Classification and Diagnosis of Epilepsy of International League Against Epilepsy (ILAE) [22]. [2] the neuropathologic diagnosis of either FCD or glioneuronal tumors (GNTs) was established by two senior neuropathologists [8, 23], discrepancies were discussed and resolved by verification from a third senior neuropathologist. [3] all the patients underwent a non-invasive pre-surgical evaluation, including long-term video-EEG monitoring, high-resolution MRI with epilepsy sequence and PET-CT for some of them; for patients whose surgical protocols were with difficulties, invasive evaluation with the stereo-electroencephalography was carried out
A total of 96 patients who underwent epilepsy surgery were analyzed in our study, including 56 patients with FCD (FCD I: n = 16; FCD II: n = 40) and 40 patients with GNTs (GG: n = 29; dysembryoplastic neuroepithelial tumors (DNTs) = 11)
Summary
Focal cortical dysplasia (FCD) is a distinctive malformation of cortical development that is highly associated with refractory epilepsy. Around 12–40% of patients with FCD were submitted to surgery for refractory epilepsy [1]. Previous studies have demonstrated that patients with FCD and GNTs have different postoperative seizure outcomes. Only 40–50% patients with FCD experienced no seizures after surgery [3, 4]. Recent studies on tumor-associated epilepsy have emphasized that total surgical resection of the tumor is sufficient and effective for seizure control in most patients with GNTs [6]. It is crucial to make the differential diagnosis of FCD and GNTs preoperatively. Their clinical manifestation and imaging findings could be similar, especially in cases of mass-like FCD [7]. What’ more, type III FCD was accompanied by an additional brain lesion as noted in the classification system by the International League Against Epilepsy (ILAE) [8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
