Deep Relational Reasoning for the Prediction of Language Impairment and Postoperative Seizure Outcome Using Preoperative DWI Connectome Data of Children With Focal Epilepsy

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

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 of F-statistics across different LMNs. The prediction accuracy increased by up to 40 across different machine learning algorithms. Notably, the best algorithm achieved the accuracy of 96/94/96 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.

Artificial intelligence role in advancement of human brain connectome studies

Neurons are interactive cells that connect via ions to develop electromagnetic fields in the brain. This structure functions directly in the brain. Connectome is the data obtained from neuronal connections. Since neural circuits change in the brain in various diseases, studying connectome sheds light on the clinical changes in special diseases. The ability to explore this data and its relation to the disorders leads us to find new therapeutic methods. Artificial intelligence (AI) is a collection of powerful algorithms used for finding the relationship between input data and the outcome. AI is used for extraction of valuable features from connectome data and in turn uses them for development of prognostic and diagnostic models in neurological diseases. Studying the changes of brain circuits in neurodegenerative diseases and behavioral disorders makes it possible to provide early diagnosis and development of efficient treatment strategies. Considering the difficulties in studying brain diseases, the use of connectome data is one of the beneficial methods for improvement of knowledge of this organ. In the present study, we provide a systematic review on the studies published using connectome data and AI for studying various diseases and we focus on the strength and weaknesses of studies aiming to provide a viewpoint for the future studies. Throughout, AI is very useful for development of diagnostic and prognostic tools using neuroimaging data, while bias in data collection and decay in addition to using small datasets restricts applications of AI-based tools using connectome data which should be covered in the future studies.

Decreased GABA levels during development result in increased connectivity in the larval zebrafish tectum

γ-aminobutyric acid (GABA) is an abundant neurotransmitter that plays multiple roles in the vertebrate central nervous system (CNS). In the early developing CNS, GABAergic signaling acts to depolarize cells. It mediates several aspects of neural development, including cell proliferation, neuronal migration, neurite growth, and synapse formation, as well as the development of critical periods. Later in CNS development, GABAergic signaling acts in an inhibitory manner when it becomes the predominant inhibitory neurotransmitter in the brain. This behavior switch occurs due to changes in chloride/cation transporter expression. Abnormalities of GABAergic signaling appear to underlie several human neurological conditions, including seizure disorders. However, the impact of reduced GABAergic signaling on brain development has been challenging to study in mammals. Here we take advantage of zebrafish and light sheet imaging to assess the impact of reduced GABAergic signaling on the functional circuitry in the larval zebrafish optic tectum. Zebrafish have three gad genes: two gad1 paralogs known as gad1a and gad1b, and gad2. The gad1b and gad2 genes are expressed in the developing optic tectum. Null mutations in gad1b significantly reduce GABA levels in the brain and increase electrophysiological activity in the optic tectum. Fast light sheet imaging of genetically encoded calcium indicator (GCaMP)-expressing gab1b null larval zebrafish revealed patterns of neural activity that were different than either gad1b-normal larvae or gad1b-normal larvae acutely exposed to pentylenetetrazole (PTZ). These results demonstrate that reduced GABAergic signaling during development increases functional connectivity and concomitantly hyper-synchronization of neuronal networks.

Connectome-based approaches in pediatric epilepsy surgery: "State-of-the art" and future perspectives

Modern epilepsy science has overcome the traditional interpretation of a strict region-specific origin of epilepsy, highlighting the involvement of wider patterns of altered neuronal circuits. In selected cases, surgery may constitute a valuable option to achieve both seizure freedom and neurocognitive improvement. Although epilepsy is now considered as a brain network disease, the most relevant literature concerning the "connectome-based" epilepsy surgery mainly refers to adults, with a limited number of studies dedicated to the pediatric population. In this review, the Authors summarized the main current available knowledge on the relevance of WM surgical anatomy in epilepsy surgery, the post-surgical modifications of brain structural connectivity and the related clinical impact of such modifications within the pediatric context. In the last part, possible implications and future perspectives of this approach have been discussed, especially concerning the optimization of surgical strategies and the predictive value of the epilepsy network analysis for planning tailored approaches, with the final aim of improving case selection, presurgical planning, intraoperative management, and postoperative results.

Deep learning in neuroimaging of epilepsy

In recent years, artificial intelligence, particularly deep learning (DL), has demonstrated utility in diverse areas of medicine. DL uses neural networks to automatically learn features from the raw data while this is not possible with conventional machine learning. It is helpful for the assessment of patients with epilepsy and whilst most published studies have been aimed at the automatic detection and prediction of seizures from electroencephalographic records, there is a growing number of investigations that use neuroimaging modalities (structural and functional magnetic resonance imaging, diffusion-weighted imaging and positron emission tomography) as input data. We review the application of DL to neuroimaging (sMRI, fMRI, DWI and PET) of focal epilepsy, specifically presurgical evaluation of drug-refractory epilepsy. First, a brief theoretical overview of artificial neural networks and deep learning is presented. Next, we review applications of deep learning to neuroimaging of epilepsy: diagnosis and lateralization, automated detection of lesion, presurgical evaluation and prediction of postsurgical outcome. Finally, the limitations, challenges and possible future directions in the application of these methods in the study of epilepsies are discussed. This approach could become an essential tool in clinical practice, particularly in the evaluation of images considered negative by visual inspection, in individualized treatments, and in the approach to epilepsy as a network disorder. However, greater multicenter collaboration is required to achieve the collection of sufficient data with the required quality together with the open access availability of the developed codes and tools.

Recognizing Pediatric Tuberous Sclerosis Complex Based on Multi-Contrast MRI and Deep Weighted Fusion Network

Multi-contrast magnetic resonance imaging (MRI) is wildly applied to identify tuberous sclerosis complex (TSC) children in a clinic. In this work, a deep convolutional neural network with multi-contrast MRI is proposed to diagnose pediatric TSC. Firstly, by combining T2W and FLAIR images, a new synthesis modality named FLAIR3 was created to enhance the contrast between TSC lesions and normal brain tissues. After that, a deep weighted fusion network (DWF-net) using a late fusion strategy is proposed to diagnose TSC children. In experiments, a total of 680 children were enrolled, including 331 healthy children and 349 TSC children. The experimental results indicate that FLAIR3 successfully enhances the visibility of TSC lesions and improves the classification performance. Additionally, the proposed DWF-net delivers a superior classification performance compared to previous methods, achieving an AUC of 0.998 and an accuracy of 0.985. The proposed method has the potential to be a reliable computer-aided diagnostic tool for assisting radiologists in diagnosing TSC children.

Spatial-Temporal Synchronous Graph Transformer network (STSGT) for COVID-19 forecasting

COVID-19 has become a matter of serious concern over the last few years. It has adversely affected numerous people around the globe and has led to the loss of billions of dollars of business capital. In this paper, we propose a novel Spatial-Temporal Synchronous Graph Transformer network (STSGT) to capture the complex spatial and temporal dependency of the COVID-19 time series data and forecast the future status of an evolving pandemic. The layers of STSGT combine the graph convolution network (GCN) with the self-attention mechanism of transformers on a synchronous spatial-temporal graph to capture the dynamically changing pattern of the COVID time series. The spatial-temporal synchronous graph simultaneously captures the spatial and temporal dependencies between the vertices of the graph at a given and subsequent time-steps, which helps capture the heterogeneity in the time series and improve the forecasting accuracy. Our extensive experiments on two publicly available real-world COVID-19 time series datasets demonstrate that STSGT significantly outperforms state-of-the-art algorithms that were designed for spatial-temporal forecasting tasks. Specifically, on average over a 12-day horizon, we observe a potential improvement of 12.19% and 3.42% in Mean Absolute Error (MAE) over the next best algorithm while forecasting the daily infected and death cases respectively for the 50 states of US and Washington, D.C. Additionally, STSGT also outperformed others when forecasting the daily infected cases at the state level, e.g., for all the counties in the State of Michigan. The code and models are publicly available at https://github.com/soumbane/STSGT.

Deep Learning for Outcome Prediction in Neurosurgery: A Systematic Review of Design, Reporting, and Reproducibility

Deep learning (DL) is a powerful machine learning technique that has increasingly been used to predict surgical outcomes. However, the large quantity of data required and lack of model interpretability represent substantial barriers to the validity and reproducibility of DL models. The objective of this study was to systematically review the characteristics of DL studies involving neurosurgical outcome prediction and to assess their bias and reporting quality. Literature search using the PubMed, Scopus, and Embase databases identified 1949 records of which 35 studies were included. Of these, 32 (91%) developed and validated a DL model while 3 (9%) validated a pre-existing model. The most commonly represented subspecialty areas were oncology (16 of 35, 46%), spine (8 of 35, 23%), and vascular (6 of 35, 17%). Risk of bias was low in 18 studies (51%), unclear in 5 (14%), and high in 12 (34%), most commonly because of data quality deficiencies. Adherence to transparent reporting of a multivariable prediction model for individual prognosis or diagnosis reporting standards was low, with a median of 12 transparent reporting of a multivariable prediction model for individual prognosis or diagnosis items (39%) per study not reported. Model transparency was severely limited because code was provided in only 3 studies (9%) and final models in 2 (6%). With the exception of public databases, no study data sets were readily available. No studies described DL models as ready for clinical use. The use of DL for neurosurgical outcome prediction remains nascent. Lack of appropriate data sets poses a major concern for bias. Although studies have demonstrated promising results, greater transparency in model development and reporting is needed to facilitate reproducibility and validation.

Application value of molecular imaging technology in epilepsy

Epilepsy is a common neurological disease with various seizure types, complicated etiologies, and unclear mechanisms. Its diagnosis mainly relies on clinical history, but an electroencephalogram is also a crucial auxiliary examination. Recently, brain imaging technology has gained increasing attention in the diagnosis of epilepsy, and conventional magnetic resonance imaging can detect epileptic foci in some patients with epilepsy. However, the results of brain magnetic resonance imaging are normal in some patients. New molecular imaging has gradually developed in recent years and has been applied in the diagnosis of epilepsy, leading to enhanced lesion detection rates. However, the application of these technologies in epilepsy patients with negative brain magnetic resonance must be clarified. Thus, we reviewed the relevant literature and summarized the information to improve the understanding of the molecular imaging application value of epilepsy.

Deep reasoning neural network analysis to predict language deficits from psychometry-driven DWI connectome of young children with persistent language concerns

This study investigated whether current state-of-the-art deep reasoning network analysis on psychometry-driven diffusion tractography connectome can accurately predict expressive and receptive language scores in a cohort of young children with persistent language concerns (n = 31, age: 4.25 ± 2.38 years). A dilated convolutional neural network combined with a relational network (dilated CNN + RN) was trained to reason the nonlinear relationship between "dilated CNN features of language network" and "clinically acquired language score". Three-fold cross-validation was then used to compare the Pearson correlation and mean absolute error (MAE) between dilated CNN + RN-predicted and actual language scores. The dilated CNN + RN outperformed other methods providing the most significant correlation between predicted and actual scores (i.e., Pearson's R/p-value: 1.00/<.001 and .99/<.001 for expressive and receptive language scores, respectively) and yielding MAE: 0.28 and 0.28 for the same scores. The strength of the relationship suggests elevated probability in the prediction of both expressive and receptive language scores (i.e., 1.00 and 1.00, respectively). Specifically, sparse connectivity not only within the right precentral gyrus but also involving the right caudate had the strongest relationship between deficit in both the expressive and receptive language domains. Subsequent subgroup analyses inferred that the effectiveness of the dilated CNN + RN-based prediction of language score(s) was independent of time interval (between MRI and language assessment) and age of MRI, suggesting that the dilated CNN + RN using psychometry-driven diffusion tractography connectome may be useful for prediction of the presence of language disorder, and possibly provide a better understanding of the neurological mechanisms of language deficits in young children.

Prediction of baseline expressive and receptive language function in children with focal epilepsy using diffusion tractography-based deep learning network

PURPOSE

Focal epilepsy is a risk factor for language impairment in children. We investigated whether the current state-of-the-art deep learning network on diffusion tractography connectome can accurately predict expressive and receptive language scores of children with epilepsy.

METHODS

We studied 37 children with a diagnosis of drug-resistant focal epilepsy (age: 11.8 ± 3.1 years) using 3 T MRI and diffusion tractography connectome: G = (S, Ω), where S is an adjacency matrix of edges representing the connectivity strength (number of white-matter tract streamlines) between each pair of brain regions, and Ω reflects a set of brain regions. A convolutional neural network (CNN) was trained to learn the nonlinear relationship between 'S (input)' and 'language score (output)'. Repeated hold-out validation was then employed to measure the Pearson correlation and mean absolute error (MAE) between CNN-predicted and actual language scores.

RESULTS

We found that CNN-predicted and actual scores were significantly correlated (i.e., Pearson's R/p-value: 0.82/<0.001 and 0.75/<0.001), yielding MAE: 7.77 and 7.40 for expressive and receptive scores, respectively. Specifically, sparse connectivity not only within the left cortico-cortical network but also involving the right subcortical structures was predictive of language impairment of expressive or receptive domain. Subsequent subgroup analyses inferred that the effectiveness of diffusion tractography-based prediction of language outcome was independent of clinical variables. Intrinsic diffusion tractography connectome properties may be useful for predicting the severity of baseline language dysfunction and possibly provide a better understanding of the biological mechanisms of epilepsy-related language impairment in children.