Deep Learning-Based Tract Classification of Preoperative DWI Tractography Advances the Prediction of Short-Term Postoperative Language Improvement in Children With Drug-Resistant Epilepsy

• Published in: IEEE transactions on biomedical engineering Vol. 72; no. 2; pp. 565 - 576
• Main Authors: Lee, Min-Hee, Banerjee, Soumyanil, Uda, Hiroshi, Carlson, Alanna, Dong, Ming, Rothermel, Robert, Juhasz, Csaba, Asano, Eishi, Jeong, Jeong-Won
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adolescent; Algorithms; Artificial neural networks; Brain - diagnostic imaging; Child; Child, Preschool; Classification; Connectome - methods; Deep convolutional neural network; Deep Learning; Diffusion Tensor Imaging - methods; Drug resistance; Drug Resistant Epilepsy - diagnostic imaging; Drug Resistant Epilepsy - surgery; Epilepsy; Female; Graph theory; Humans; Image Interpretation, Computer-Assisted - methods; Language; Learning algorithms; Machine learning; Machine learning algorithms; Magnetic resonance imaging; Male; network connectivity marker; Neural networks; Neuroimaging; pediatric epilepsy surgery; Pediatrics; Postoperative Period; prediction of language improvement; Predictions; psychometry-driven language modular network; Reproducibility of results; Statistical analysis; Surgery; Surgical instruments
• ISSN: 0018-9294; 1558-2531; 1558-2531
• DOI: 10.1109/TBME.2024.3463481

Objective: To develop an innovative deep convolutional neural network (DCNN)-based tract classification to enhance the prediction of short-term postoperative language improvement using axonal connectivity markers derived from specific language modular networks (LMNs) within the preoperative whole-brain diffusion-weighted imaging connectome (wDWIC). Methods: We employed a three-step approach. First, our previous DCNN-based tract classification to detect true-positive eloquent tracts was extended using an open-source database of high-quality wDWIC to facilitate the accurate classification of true-positive tracts within the preoperative backbone wDWIC of individual patients. Next, we applied psychometry-driven DWIC analysis to the resulting DCNN-based backbone wDWIC in order to create core, expressive, and receptive LMNs. Finally, graph and circuit theory-based connectivity markers were assessed within the three LMNs and compared using a series of machine learning algorithms to predict the presence of postoperative language improvement from a given LMN. Results: The results showed that the extended DCNN tract classification significantly improved the reproducibility of connectivity markers by up to 35.5<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> of F-statistics across different LMNs. The prediction accuracy increased by up to 40<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> across different machine learning algorithms. Notably, the best algorithm achieved the accuracy of 96<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula>/94<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula>/96<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> to predict the presence of language improvement about two months after surgery in core/expressive/receptive domain of an independent validation cohort. Conclusion: These domains hold great potential to assist physicians in identifying candidates whose language skills stand to benefit from early surgery. Significance: DCNN tract classification may be an effective tool to improve predicting short-term postoperative language improvement in pediatric epilepsy surgery.