Recognition of Cognitive Impairment in Adult Moyamoya Disease: A Classifier Based on High-Order Resting-State Functional Connectivity Network

The new era of artificial intelligence in neuroradiology: current research and promising tools

Radiology has a number of characteristics that make it an especially suitable medical discipline for early artificial intelligence (AI) adoption. These include having a well-established digital workflow, standardized protocols for image storage, and numerous well-defined interpretive activities. The more than 200 commercial radiologic AI-based products recently approved by the Food and Drug Administration (FDA) to assist radiologists in a number of narrow image-analysis tasks such as image enhancement, workflow triage, and quantification, corroborate this observation. However, in order to leverage AI to boost efficacy and efficiency, and to overcome substantial obstacles to widespread successful clinical use of these products, radiologists should become familiarized with the emerging applications in their particular areas of expertise. In light of this, in this article we survey the existing literature on the application of AI-based techniques in neuroradiology, focusing on conditions such as vascular diseases, epilepsy, and demyelinating and neurodegenerative conditions. We also introduce some of the algorithms behind the applications, briefly discuss a few of the challenges of generalization in the use of AI models in neuroradiology, and skate over the most relevant commercially available solutions adopted in clinical practice. If well designed, AI algorithms have the potential to radically improve radiology, strengthening image analysis, enhancing the value of quantitative imaging techniques, and mitigating diagnostic errors.

Switching of brain networks across different cerebral perfusion states: insights from EEG dynamic microstate analyses

The alteration of neural interactions across different cerebral perfusion states remains unclear. This study aimed to fulfill this gap by examining the longitudinal brain dynamic information interactions before and after cerebral reperfusion. Electroencephalogram in eyes-closed state at baseline and postoperative 7-d and 3-month follow-ups (moyamoya disease: 20, health controls: 23) were recorded. Dynamic network analyses were focused on the features and networks of electroencephalogram microstates across different microstates and perfusion states. Considering the microstate features, the parameters were disturbed of microstate B, C, and D but preserved of microstate A. The transition probabilities of microstates A-B and B-D were increased to play a complementary role across different perfusion states. Moreover, the microstate variability was decreased, but was significantly improved after cerebral reperfusion. Regarding microstate networks, the functional connectivity strengths were declined, mainly within frontal, parietal, and occipital lobes and between parietal and occipital lobes in different perfusion states, but were ameliorated after cerebral reperfusion. This study elucidates how dynamic interaction patterns of brain neurons change after cerebral reperfusion, which allows for the observation of brain network transitions across various perfusion states in a live clinical setting through direct intervention.

Classify patients with Moyamoya disease according to their cognitive performance might be helpful in clinical and practical with support vector machine based on hypergraph

Moyamoya disease (MMD) patients were now classified according to their cerebrovascular manifestations, with cognition and emotion ignored, which attenuated the therapy. The present study tried to classify them based on their cognitive and emotional performance and explored the neural basis underlying this classification using resting-state fMRI (rs-fMRI). Thirty-nine MMD patients were recruited, assessed mental function and MRI scanned. We adopted hierarchical analysis of their mental performance for new subtypes. Next, a three-step analysis, with each step consisting of 10 random cross validation, was conducted for robust brain regions in classifying the three subtypes of patients in a support vector machine (SVM) model with hypergraph of rs-fMRI. We found three new subtypes including high depression-high anxiety-low cognition (HE-LC, 50%), low depression-low anxiety-high cognition (LE-HC, 14%), and low depression-low anxiety-low cognition (LE-LC, 36%), and no hemorrhagic MMD patients fell into the LE-HC group. The temporal and the bilateral superior frontal cortex, and so forth were included in all 10 randomized SVM modeling. The classification accuracy of the final three-way classification model was 67.5% in average of 10 random cross validation. In addition, the S value between the frontal cortex and the angular cortex was positively correlated with the anxiety score and backward digit span (p < .05). Our results might provide a new perspective for MMD classification concerning patients' mental status, guide timely surgery and suggest angular cortex, and so forth should be protected in surgery for cognitive consideration.

Preoperative brain connectome predicts postoperative changes in processing speed in moyamoya disease

Moyamoya disease is a rare cerebrovascular disorder associated with cognitive dysfunction. It is usually treated by surgical revascularization, but research on the neurocognitive outcomes of revascularization surgery is controversial. Given that neurocognitive impairment could affect the daily activities of patients with moyamoya disease, early detection of postoperative neurocognitive outcomes has the potential to improve patient management. In this study, we applied a well-established connectome-based predictive modelling approach to develop machine learning models that used preoperative resting-state functional connectivity to predict postoperative changes in processing speed in patients with moyamoya disease. Twelve adult patients with moyamoya disease (age range: 23–49 years; female/male: 9/3) were recruited prior to surgery and underwent follow-up at 1 and 6 months after surgery. Twenty healthy controls (age range: 24–54 years; female/male: 14/6) were recruited and completed the behavioural test at baseline, 1-month follow-up and 6-month follow-up. Behavioural results indicated that the behavioural changes in processing speed at 1 and 6 months after surgery compared with baseline were not significant. Importantly, we showed that preoperative resting-state functional connectivity significantly predicted postoperative changes in processing speed at 1 month after surgery (negative network: ρ = 0.63, P_corr = 0.017) and 6 months after surgery (positive network: ρ = 0.62, P_corr = 0.010; negative network: ρ = 0.55, P_corr = 0.010). We also identified cerebro-cerebellar and cortico-subcortical connectivities that were consistently associated with processing speed. The brain regions identified from our predictive models are not only consistent with previous studies but also extend previous findings by revealing their potential roles in postoperative neurocognitive functions in patients with moyamoya disease. Taken together, our findings provide preliminary evidence that preoperative resting-state functional connectivity might predict the post-surgical longitudinal neurocognitive changes in patients with moyamoya disease. Given that processing speed is a crucial cognitive ability supporting higher neurocognitive functions, this study’s findings offer important insight into the clinical management of patients with moyamoya disease.

A Review of Artificial Intelligence in Cerebrovascular Disease Imaging: Applications and Challenges

BACKGROUND

A variety of emerging medical imaging technologies based on artificial intelligence have been widely applied in many diseases, but they are still limitedly used in the cerebrovascular field even though the diseases can lead to catastrophic consequences.

OBJECTIVE

This work aims to discuss the current challenges and future directions of artificial intelligence technology in cerebrovascular diseases through reviewing the existing literature related to applications in terms of computer-aided detection, prediction and treatment of cerebrovascular diseases.

METHODS

Based on artificial intelligence applications in four representative cerebrovascular diseases including intracranial aneurysm, arteriovenous malformation, arteriosclerosis and moyamoya disease, this paper systematically reviews studies published between 2006 and 2021 in five databases: National Center for Biotechnology Information, Elsevier Science Direct, IEEE Xplore Digital Library, Web of Science and Springer Link. And three refinement steps were further conducted after identifying relevant literature from these databases.

RESULTS

For the popular research topic, most of the included publications involved computer-aided detection and prediction of aneurysms, while studies about arteriovenous malformation, arteriosclerosis and moyamoya disease showed an upward trend in recent years. Both conventional machine learning and deep learning algorithms were utilized in these publications, but machine learning techniques accounted for a larger proportion.

CONCLUSION

Algorithms related to artificial intelligence, especially deep learning, are promising tools for medical imaging analysis and will enhance the performance of computer-aided detection, prediction and treatment of cerebrovascular diseases.

Abnormal brain functional and structural connectivity between the left supplementary motor area and inferior frontal gyrus in moyamoya disease

BACKGROUND

Disruption of brain functional connectivity has been detected after stroke, but whether it also occurs in moyamoya disease (MMD) is unknown. Impaired functional connectivity is always correlated with abnormal white matter fibers. Herein, we used multimodal imaging techniques to explore the changes in brain functional and structural connectivity in MMD patients.

METHODS

We collected structural images, resting-state functional magnetic resonance imaging (rs-fMRI) and diffusion tensor imaging for each subject. Cognitive functions of MMD patients were evaluated using the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and Trail Making Test parts A and B (TMT-A/-B). We calculated the functional connectivity for every paired region using 90 regions of interest from the Anatomical Automatic Labeling Atlas and then determined the differences between MMD patients and HCs. We extracted the functional connectivity of paired brain regions with significant differences between the two groups. Correlation analyses were then performed between the functional connectivity and variable cognitive functions. To explore whether the impaired functional connectivity and cognitive performances were attributed to the destruction of white matter fibers, we further analyzed fiber integrity using tractography between paired regions that were correlated with cognition.

RESULTS

There was lower functional connectivity in MMD patients as compared to HCs between the bilateral inferior frontal gyrus, between the bilateral supramarginal gyrus, between the left supplementary motor area (SMA) and the left orbital part of the inferior frontal gyrus (IFGorb), and between the left SMA and the left middle temporal gyrus (P < 0.01, FDR corrected). The decreased functional connectivity between the left SMA and the left IFGorb was significantly correlated with the MMSE (r = 0.52, P = 0.024), MoCA (r = 0.60, P = 0.006), and TMT-B (r = -0.54, P = 0.048) in MMD patients. White matter fibers were also injured between the SMA and IFGorb in the left hemisphere and were positively correlated with reduced functional connectivity.

CONCLUSIONS

Brain functional and structural connectivity between the supplementary motor area and inferior frontal gyrus in the left hemisphere are damaged in MMD. These findings could be useful in the evaluation of disease progression and prognosis of MMD.

Multi-Order Brain Functional Connectivity Network-Based Machine Learning Method for Recognition of Delayed Neurocognitive Recovery in Older Adults Undergoing Non-cardiac Surgery

Objectives: Delayed neurocognitive recovery (DNR) seriously affects the post-operative recovery of elderly surgical patients, but there is still a lack of effective methods to recognize high-risk patients with DNR. This study proposed a machine learning method based on a multi-order brain functional connectivity (FC) network to recognize DNR.

Method: Seventy-four patients who completed assessments were included in this study, in which 16/74 (21.6%) had DNR following surgery. Based on resting-state functional magnetic resonance imaging (rs-fMRI), we first constructed low-order FC networks of 90 brain regions by calculating the correlation of brain region signal changing in the time dimension. Then, we established high-order FC networks by calculating correlations among each pair of brain regions. Afterward, we built sparse representation-based machine learning model to recognize DNR on the extracted multi-order FC network features. Finally, an independent testing was conducted to validate the established recognition model.

Results: Three hundred ninety features of FC networks were finally extracted to identify DNR. After performing the independent-sample T test between these features and the categories, 15 features showed statistical differences (P < 0.05) and 3 features had significant statistical differences (P < 0.01). By comparing DNR and non-DNR patients’ brain region connection matrices, it is found that there are more connections among brain regions in DNR patients than in non-DNR patients. For the machine learning recognition model based on multi-feature combination, the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the classifier reached 95.61, 92.00, 66.67, and 100.00%, respectively.

Conclusion: This study not only reveals the significance of preoperative rs-fMRI in recognizing post-operative DNR in elderly patients but also establishes a promising machine learning method to recognize DNR.

Progression in Moyamoya Disease: Clinical Feature, Neuroimaging Evaluation and Treatment

Moyamoya disease (MMD) is a chronic cerebrovascular disease characterized by progressive stenosis of the arteries of the circle of Willis, with the formation of collateral vascular network at the base of the brain. Its clinical manifestations are complicated. Numerous studies have attempted to clarify the clinical features of MMD, including its epidemiology, genetic characteristics, and pathophysiology. With the development of neuroimaging techniques, various neuroimaging modalities with different advantages have deepened the understanding of MMD in terms of structural, functional, spatial, and temporal dimensions. At present, the main treatment for MMD focuses on neurological protection, cerebral blood flow reconstruction, and neurological rehabilitation, such as pharmacological treatment, surgical revascularization, and cognitive rehabilitation. In this review, we discuss recent progress in understanding the clinical features, in the neuroimaging evaluation and treatment of MMD.