Alterations of Iron Level in the Bilateral Basal Ganglia Region in Patients With Middle Cerebral Artery Occlusion

An exploration on the machine-learning-based stroke prediction model

Introduction

With the rapid development of artificial intelligence technology, machine learning algorithms have been widely applied at various stages of stroke diagnosis, treatment, and prognosis, demonstrating significant potential. A correlation between stroke and cytokine levels in the human body has recently been reported. Our study aimed to establish machine-learning models based on cytokine features to enhance the decision-making capabilities of clinical physicians.

Methods

This study recruited 2346 stroke patients and 2128 healthy control subjects from Chongqing University Central Hospital. A predictive model was established through clinical experiments and collection of clinical laboratory tests and demographic variables at admission. Three classification algorithms, namely Random Forest, Gradient Boosting, and Support Vector Machine, were employed. The models were evaluated using methods such as ROC curves, AUC values, and calibration curves.

Results

Through univariate feature selection, we selected 14 features and constructed three machine-learning models: Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting Machine (GBM). Our results indicated that in the training set, the RF model outperformed the GBM and SVM models in terms of both the AUC value and sensitivity. We ranked the features using the RF algorithm, and the results showed that IL-6, IL-5, IL-10, and IL-2 had high importance scores and ranked at the top. In the test set, the stroke model demonstrated a good generalization ability, as evidenced by the ROC curve, confusion matrix, and calibration curve, confirming its reliability as a predictive model for stroke.

Discussion

We focused on utilizing cytokines as features to establish stroke prediction models. Analyses of the ROC curve, confusion matrix, and calibration curve of the test set demonstrated that our models exhibited a strong generalization ability, which could be applied in stroke prediction.

Evaluation of brain iron deposition in different cerebral arteries of acute ischaemic stroke patients using quantitative susceptibility mapping

AIM

To investigate differences in iron deposition between infarct and normal cerebral arterial regions in acute ischaemic stroke (AIS) patients using quantitative susceptibility mapping (QSM).

MATERIALS AND METHODS

Forty healthy controls and 40 AIS patients were recruited, and their QSM images were obtained. There were seven regions of interest (ROIs) in AIS patients, including the infarct regions of responsible arteries (R1), the non-infarct regions of responsible arteries (R2), the contralateral symmetrical sites of lesions (R3), and the non-responsible cerebral arterial regions (R4, R5, R6, R7). For the healthy controls, the cerebral arterial regions corresponding to the AIS patient group were selected as ROIs. The differences in corresponding ROI susceptibilities between AIS patients and healthy controls and the differences in susceptibilities between infarcted and non-infarct regions in AIS patients were compared.

RESULTS

The susceptibilities of infarct regions in AIS patients were significantly higher than those in healthy controls (p<0.0001). There was no significant difference in non-infarct regions between the two groups (p>0.05). The susceptibility of the infarct regions in AIS patients was significantly higher than those of the non-infarct region of responsible artery and non-responsible cerebral arterial regions (p<0.01).

CONCLUSIONS

Abnormal iron deposition detected by QSM in the infarct regions of AIS patients may not affect iron levels in the non-infarct regions of responsible arteries and normal cerebral arteries, which may open the door for potential new diagnostic and treatment strategies.

A Preliminary Study of Alterations in Iron Disposal and Neural Activity in Ischemic Stroke

Purpose

The study aimed to evaluate the postrehabilitation changes in deep gray matter (DGM) nuclei, corticospinal tract (CST), and motor cortex area, involved in motor tasks in patients with ischemic stroke.

Methods

Three patients participated in this study, who had experienced an ischemic stroke on the left side of the brain. They underwent a standard rehabilitation program for four consecutive weeks, including transcranial direct current stimulation (tDCS), neuromuscular electrical stimulation (NMES), and occupational therapy. The patients' motor ability was evaluated by Fugl-Meyer assessment-upper extremity (FMA-UE) and Wolf motor function test (WMFT). Multimodal magnetic resonance imaging (MRI) was acquired from the patients by a 3 Tesla machine before and after the rehabilitation. The magnetic susceptibility changes were examined in DGM nuclei including the bilateral caudate (CA), putamen (PT), globus pallidus (GP), and thalamus (TH) using quantitative susceptibility mapping (QSM). Functional MRI (fMRI) in the motor cortex areas was acquired to evaluate the postrehab functional motor activity. The three-dimensional corticospinal tract (CST) was reconstructed using diffusion-weighted imaging (DWI) and diffusion tensor tractography (DTT), and the fractional anisotropy (FA) was measured along the tract. Ultimately, the relationship between the structural and functional changes was evaluated in CST and motor cortex.

Results

Postrehabilitation FMA-UE and WMFT scores increased for all patients compared to the prerehabilitation. QSM analysis revealed increasing in susceptibility values in GP and CA in all patients at the ipsilesional hemisphere. By fMRI analysis, the ipsilesional hemisphere demonstrated an increase in functional activity in motor areas for all 3 patients. In the ipsilesional hemisphere, the fractional anisotropy (FA) was increased in CST in two patients, while the mean diffusivity (MD) was decreased in CA in a patient, in PT and TH in another patient, and in PT in two patients.

Conclusion

This preliminary study demonstrates that the magnetic susceptibility may decrease at some ipsilesional DGM nuclei after tDCS, NMES, and occupational therapy for patients with ischemic stroke, suggesting a drop in the level of iron deposition, which may be associated with an increase in the level of activity in motor cortex after rehabilitation.

Iron Deposition in Gray Matter Nuclei of Patients With Intracranial Artery Stenosis: A Quantitative Susceptibility Mapping Study

Purpose: This study aimed to use quantitative susceptibility mapping (QSM) to systematically investigate the changes of iron content in gray matter (GM) nuclei in patients with long-term anterior circulation artery stenosis (ACAS) and posterior circulation artery stenosis (PCAS).

Methods: Twenty-five ACAS patients, 25 PCAS patients, and 25 age- and sex-matched healthy controls underwent QSM examination. Patients were scored using the National Institutes of Health Stroke Scale (NIHSS) and modified Rankin Scale (mRS) to assess the degree of neural function deficiency. On QSM images, iron related susceptibility of GM nuclei, including bilateral caudate nucleus, putamen (PU), globus pallidus (GP), thalamus (TH), substantia nigra (SN), red nucleus, and dentate nucleus (DN), were assessed. Susceptibility was compared between bilateral GM nuclei in healthy controls, ACAS patients, and PCAS patients. Partial correlation analysis, with age as a covariate, was separately performed to assess the relationships of susceptibility with NIHSS and mRS scores.

Results: There were no significant differences between the susceptibilities for left and right hemispheres in all seven GM nucleus subregions for healthy controls, ACAS patients, and PCAS patients. Compared with healthy controls, mean susceptibility of bilateral PU, GP, and SN in ACAS patients and of bilateral PU, GP, SN, and DN in PCAS patients were significantly increased (all P < 0.05). In addition, mean susceptibility of bilateral TH and SN in PCAS patients was significantly higher than in ACAS patients (both P < 0.05). With partial correlation analysis, mean susceptibility at bilateral PU of ACAS patients was significantly correlated with mRS score (r = 0.415, P < 0.05), and at bilateral PU in PCAS patients was correlated with NIHSS score (r = 0.424, P < 0.05).

Conclusion: Our findings indicated that abnormal iron metabolism may present in different subregions of GM nuclei after long-term ACAS and PCAS. In addition, iron content of PU in patients with ACAS and PCAS was correlated with neurological deficit scores. Therefore, iron quantification measured by QSM susceptibility may provide a new insight to understand the pathological mechanism of ischemic stroke caused by ACAS and PCAS.