| Title:
|
FCD Type I: Unveiling the Invisible—Bridging Diagnostic Gaps with Advanced Neuroimaging and AI Innovation
|
| Author(s):
|
1
|
| Subject:
|
Brain Mapping Applications in Epilepsy – Across Pediatric and Adult Populations
|
|
|
|
Presenter Author:
|
|
| Name:
|
Arshia Alizadeh
|
|
|
|
Author:
|
|
| Name 1:
|
Arshia Alizadeh - Presenter
|
| Affiliation 1:
|
Mashhad University of Medical Sciences
|
|
|
| Name 2:
|
Hoda zare - Corresponding
|
| Affiliation 2:
|
Mashhad University of Medical Sciences
|
|
|
| Name 3:
|
Zohre Ganji
|
| Affiliation 3:
|
Mashhad University of Medical Sciences
|
|
|
| Name 4:
|
Amirhossein Mahmoudi
|
| Affiliation 4:
|
Mashhad University of Medical Sciences
|
|
|
| Name 5:
|
|
| Affiliation 5:
|
|
|
|
| Name 6:
|
|
| Affiliation 6:
|
|
|
|
| Name 7:
|
|
| Affiliation 7:
|
|
|
|
| Name 8:
|
|
| Affiliation 8:
|
|
|
|
| Name 9:
|
|
| Affiliation 9:
|
|
|
|
| Name 10:
|
|
| Affiliation 10:
|
|
|
|
Abstract
|
Original Research
|
|
introduction :
|
Focal Cortical Dysplasia (FCD) is a common cause of drug-resistant epilepsy, especially in children and young adults. Type I FCD is particularly difficult to detect on conventional MRI due to its subtle microstructural abnormalities, making it "MRI-negative" in up to 40% of cases. Advanced MRI techniques such as FLAWS, 3D-EDGE, and DTI have improved detection, but challenges like protocol variability and subjective interpretation remain. Artificial intelligence (AI), including machine learning and deep learning models, shows promise in automating lesion detection, though limited and heterogeneous datasets—especially for Type I—hinder widespread adoption. Future progress depends on standardized imaging protocols, multicenter data sharing, and integration of multimodal imaging for more precise diagnosis.
|
|
Methods:
|
The study used the PICO framework focusing on patients with histopathologically or clinically confirmed FCD Type I. Inclusion criteria were original English-language articles from 2015 to 2025 using MRI for lesion identification, excluding reviews and studies without quantitative data. A comprehensive search was performed in PubMed, Embase, Scopus, Web of Science, and ScienceDirect using keywords related to FCD, MRI, and artificial intelligence. Language filters and database-specific search optimizations were applied to improve relevance.
|
|
Results:
|
This systematic review analyzed five studies with 271 participants from countries including China, the US, and Canada. Three studies were retrospective, analyzing existing clinical and imaging data, while two focused on advanced MRI techniques and AI applications. Most studies emphasized novel MRI sequences like FLAWS and 3D-EDGE to improve lesion detection, with only one using machine learning (SVM) for FCD subtype classification. The FLAWS sequence showed about 72% sensitivity and specificity, and 3D-EDGE enhanced gray-white matter junction visualization. Despite these advances, limitations such as small sample sizes, variable MRI protocols, and lack of standardization remain. Future research should integrate AI with optimized MRI sequences to enhance detection and subtype differentiation.
|
|
Conclusion:
|
FCD Type I remains difficult to diagnose due to subtle abnormalities missed by conventional MRI. Advanced sequences like FLAWS and 3D-EDGE improve detection by enhancing gray matter contrast, while DTI helps differentiate lesion subtypes. AI methods, especially SVM, show promise with high accuracy but face challenges such as small sample sizes and lack of interpretability. Variability in MRI protocols and limited histopathological validation hinder consistency across studies. Future progress depends on standardized imaging, larger datasets, multimodal approaches, and explainable AI to improve diagnosis and guide treatment.
|
| Attachment:
|
|
| Keywords:
|
FCD I ,Advanced Neuroimaging, Artificial Intelligence, Diagnostic Challenges, Quantitative Analyses
|
| Key message:
|
|
| Do you want your abstract to be indexed, if this is selected as one of the top 100
articles?
|
