Helping stroke survivors regain function has long been a challenge in neurorehabilitation, and the field is changing from traditional, experience-based clinical judgments towards smarter and more interactive human-machine models. In this review, we follow a simple thread: start with clinical needs, then look at technology integration, then decision support, and finally system reshaping. We survey recent advances in intelligent sensing and digital therapeutics for stroke rehabilitation, and three stand out. First, flexible smart sensors plus multimodal data fusion now enable us to measure dysfunction precisely and quantitatively. Second, brain-computer interfaces with closed-loop feedback systems can drive active and adaptive neural remodeling. Third, a "Meta-consultation" model helps build a coordinated rehab network which connects hospital, community, and home-based training. Together, these changes push stroke rehabilitation toward more precise, more personalized, and more affordable care. We also discuss the major challenges that remain before these technologies can be used widely in clinics, and suggest directions for future interdisciplinary research. Our goal is to provide both theoretical frameworks and practical pathways for building a human-centered smart rehab ecosystem.

Kennedy's disease (KD) is a rare X-linked recessive neurodegenerative disease. Although KD progresses slowly, it leads to weakness and atrophy of the limb and bulbar-innervated muscles, ultimately resulting in gradual loss of walking ability, as well as speech, swallowing, and respiratory function impairments. The primary causes of death in patients are pneumonia caused by aspiration, respiratory failure, or bedsore caused by long-term bed rest. Early comprehensive intervention may help delay disease progression and improve patients' quality of life. To standardize and promote the diagnosis and management of KD in China, this consensus summarizes recent progress in the understanding of its clinical manifestations, neurophysiological findings, muscle biopsy characteristics, genetic testing, treatment strategies, and genetic interventions for clinical reference.

To investigate the relationship between oral frailty (OF) in patients with cerebral small vessel disease (CSVD) and body composition, as well as neuroimaging biomarkers.

The study included patients with CSVD hospitalized in the Department of Neurology at the Seventh Medical Center of the Chinese People's Liberation Army General Hospital from January 2022 to June 2023. OF was assessed by the oral frailty index-8 (OFI-8), with patients stratified into a low-risk group (score ≤3) and a high-risk group (score ≥4). Data collected encompassed body fat percentage (BF), appendicular skeletal muscle mass index (ASMI), and basal metabolic rate (BMR). Cognitive function was also evaluated, alongside neuroimaging markers of CSVD, including white matter hyperintensities (WMH), lacunes, cerebral microbleeds (CMBs), and CSVD composite score. Variables conforming to a normal distribution were compared using the Student's t-test, while those not following a normal distribution were analyzed with the Mann-Whitney U test. Categorical data were compared using the Chi-square test. Binary Logistic regression was employed to identify independent risk factors associated with OF.

This study enrolled a total of 95 patients, comprising 48 males (31 in the low-risk group and 17 in the high-risk group) and 47 females (33 in the low-risk group and 14 in the high-risk group). Among male CSVD patients, individuals in the high-risk group exhibited a significant reduction in ASMI (OR=0.058, P=0.002) and a marked increase in BF (OR=1.331, P=0.002). Even after adjusting for confounding factors, both ASMI and BF remained independently associated with oral frailty (ASMI: OR=0.038, P=0.003; BF: OR=1.359, P=0.005). Gender differences were observed, with female CSVD patients showing a significant increase in BF exclusively within the high-risk group (OR=1.274, P=0.023). After adjusting for confounding variables, BF remained an independent risk factor for oral frailty (OR=1.308, P=0.028). Moreover, the high-risk group exhibited a significantly higher CSVD composite score (OR=6.532, P=0.004). After adjusting for potential confounders, the CSVD composite score remained an independent predictor of oral frailty risk (OR=5.898, P=0.008).

Among patients with CSVD, body composition is significantly associated with oral frailty, with this relationship exhibiting notable gender differences. Additionally, the overall burden of CSVD plays a crucial role in the onset and progression of oral frailty, with a higher CSVD composite score serving as a key risk factor for its development.

To investigate the effects of muscle health on ischemic stroke and the glymphatic system (GS), and to analyze the mediating role of the GS between the two.

Based on data from the UK Biobank, patients diagnosed with ischemic stroke [International Classification of Diseases, Tenth Revision (ICD-10) codes G45 and/or I63] at the time of first imaging acquisition were included in the analysis. Muscle health measures (walking pace and grip strength) were collected for all participants. The diffusion tensor imaging along the perivascular space (DTI-ALPS) index was derived from diffusion-weighted magnetic resonance imaging, including anterior (aALPS), middle (mALPS), posterior (pALPS), and total (tALPS) indices. Walking pace was self-reported and categorized into slow, steady, and fast pace groups. Grip strength was divided into tertiles based on absolute and relative values, resulting in low, medium, and high grip strength groups. Logistic regression was used to analyze the association of walking pace and grip strength with ischemic stroke risk, while multivariate linear regression assessed their association with the DTI-ALPS index. A causal mediation analysis framework was applied to examine the mediating effect of the DTI-ALPS index.

This study included a total of 37 370 ischemic stroke patients with data on grip strength, walking pace, and DTI-ALPS index. Slow walking pace was a risk factor for ischemic stroke (OR=1.577, 95%CI: 1.186 – 2.070, P=0.001), while high absolute grip strength was a protective factor (OR=0.774, 95%CI: 0.628 – 0.952, P=0.016). In contrast, relative grip strength showed no significant effect on the risk of ischemic stroke (OR=0.819 – 0.871, both P＞0.05). Compared with the steady pace group, the slow pace group showed a decrease in all DTI-ALPS indices (β: -0.037 – -0.025, all P＜0.001), while the fast pace group showed an increase in all DTI-ALPS indices (β: 0.009 – 0.012, all P＜0.001). Relative to the low absolute grip strength group, the medium absolute grip strength group exhibited increased DTI-ALPS indices (β: 0.010 – 0.013, all P＜0.001), and the high absolute grip strength group demonstrated even greater increases (β: 0.015 – 0.020, all P＜0.001). Logistic regression analysis revealed that higher DTI-ALPS indices were associated with significantly lower risk of ischemic stroke (aALPS: OR=0.577, P=0.011; mALPS: OR=0.631, P=0.015; pALPS: OR=0.574, P=0.016; tALPS: OR=0.526, P=0.006), with the tALPS index being the most significant. Mediation analysis revealed that all DTI-ALPS indices partially mediated the association of ischemic stroke with slow walking pace (mediation proportion: 2.19% – 3.11%) and high absolute grip strength (mediation proportion: 3.60% – 4.80%) with ischemic stroke.

Individuals with better muscle function, characterized by faster walking pace and stronger grip strength, exhibited improved GS function and a lower risk of ischemic stroke. Improving muscle health in patients with ischemic stroke may help regulate GS function, thereby reducing the risk of stroke occurrence.

To assess whether there are causal associations between four air pollutants [nitrogen dioxide (NO₂), nitrogen oxides (NO_x), fine particulate matter (PM_2.5), and inhalable particulate matter (PM₁₀)] and the risk of ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage.

A two-sample Mendelian randomization (MR) approach was employed. Single nucleotide polymorphisms associated with air pollutants such as NO₂, NO_x, PM_2.5, and PM₁₀ were used as instrumental variables, with summary data sourced from the UK Biobank database. Data for ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage were obtained from the MEGASTROKE project, the International Stroke Genetics Consortium, and the dataset of aneurysmal subarachnoid hemorrhage with European populations in the 2020 genome-wide association study, respectively. The primary analysis method was the inverse variance weighting (IVW) method, with sensitivity analyses conducted using the simple median, weighted median, MR-Egger regression, leave-one-out and MR-PRESSO methods.

The random-effects IVW analysis showed a nominal association between PM_2.5 and subarachnoid hemorrhage, but the confidence interval was wide (OR=5.75, 95%CI: 1.13 – 29.20, P=0.035). After Bonferroni correction, this nominal association was no longer statistically significant. Further sensitivity analyses using MR-PRESSO and leave-one-out methods identified an outlier SNP; after correction for this outlier, the effect estimate substantially decreased and lost statistical significance (OR=2.05, 95%CI: 0.98 – 4.29, P=0.069). No associations were observed between subarachnoid hemorrhage and other air pollutants (NO₂, NO_x, and PM₁₀), and no significant associations were found between any air pollutants and intracerebral hemorrhage or ischemic stroke.

This study found no robust causal association between genetically predicted exposure to common air pollutants (NO₂, NO_x, PM_2.5, and PM₁₀) and the risk of ischemic stroke, intracerebral hemorrhage, or subarachnoid hemorrhage. The nominally significant association between PM_2.5 and subarachnoid hemorrhage was largely driven by an outlier SNP, the signal was mainly driven by an outlier SNP, suggesting hat this genetic variant may act as an effect modifier in gene-environment interactions.

To investigate the association between early blood pressure variability (BPV) and 90-day functional outcome in patients with acute ischemic stroke with large vessel occlusion (AIS-LVO) after mechanical thrombectomy, and to evaluate the predictive value of BPV for 90-day favorable outcome.

This retrospective study analyzed 100 AIS-LVO patients who underwent mechanical thrombectomy at Beijing Chaoyang Hospital, Capital Medical University from October 2023 to June 2024. Patients were divided into a low-BPV group (n=72) and a high-BPV group (n=28) according to the median BPV within the first 24 hours postoperatively. Demographic and clinical characteristics were compared between groups. The correlations between BPV and the 90-day modified Rankin scale (mRS) score were assessed by Pearson correlation coefficient (r) and Spearman rank correlation coefficient (ρ). Multivariable Logistic regression was used to determine the independent predictive value of BPV for a 90-day favorable outcome (90-day mRS score ≤2). Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminative ability of BPV.

BPV showed a significant positive correlation with the 90-day mRS score (r=0.782, ρ=0.719, P＜0.001). Multivariable Logistic regression identified that BPV (OR=0.504, 95%CI: 0.360 – 0.694, P＜0.001) and preoperative National Institutes of Health stroke scale (NIHSS) score (OR=0.950, 95%CI: 0.910 – 0.994, P=0.035) were independent predictors of a favorable outcome at 90 days, with higher BPV showing a significant negative correlation with good prognosis. ROC curve analysis demonstrated that BPV predicted 90-day favorable outcome with an area under the curve (AUC) of 0.865, a sensitivity of 85.7%, and a specificity of 75.0%.

Elevated early BPV after mechanical thrombectomy is closely associated with poor 90-day functional outcome in AIS-LVO patients. BPV can serve as an independent predictor of a favorable prognosis, and early postoperative BPV management may improve long-term neurological recovery.

To investigate the efficacy and safety of endovascular treatment (EVT) guided by CT perfusion imaging (CTP) in patients with minor stroke caused by emergent large vessel occlusion (ELVO) in the anterior circulation.

A retrospective, consecutive cohort of patients with anterior circulation ELVO-related minor stroke admitted to 8 hospitals between January 2021 and December 2023 was enrolled. Based on the initial treatment strategy, patients were divided into the EVT group (n=74) and the best medical therapy (BMT) group (n=110). Treatment outcomes [24-hour National Institutes of Health stroke scale (NIHSS) score, 72-hour NIHSS score, discharge NIHSS score, incidence of early neurological deterioration (END)], safety [incidence of symptomatic intracranial hemorrhage (sICH)], and follow-up outcomes [90-day modified Rankin scale (mRS) score, 1-year ischemic stroke recurrence rate, and mortality] were compared between the two groups. Procedural data for the EVT group [door-to-puncture time (DPT), post-procedural modified Thrombolysis in Cerebral Infarction (mTICI) grade] were collected. All patients underwent cranial CTP assessment before treatment, within 3 days after treatment, and at the 90-day follow-up. The volumes of the infarct core (region with relative cerebral blood flow ＜30%) and the hypoperfusion region (region with Tmax ＞6 s) were compared between the two groups. Independent samples t test or Mann-Whitney U test was used for comparison of continuous data between groups. The χ² test was used for comparison of categorical data between groups.

The immediate successful recanalization rate (mTICI grade ≥2b) in the EVT group was 95.95%. Compared with the BMT group, the EVT group demonstrated significantly lower NIHSS scores at 24 hours [2 (0, 8) vs 2 (1, 10)], 72 hours [1 (0, 10) vs 1 (0, 14)], and at discharge [0 (0, 7) vs 1 (0, 12)], along with a significantly lower incidence of END (13.51% vs 30.00%). These differences were statistically significant (Z=-4.007, -4.677, -4.563, χ²=6.714; P＜0.001, ＜0.001, ＜0.001, =0.010). At the 90-day follow-up, the EVT group had significantly higher proportions of patients with favorable functional outcomes (86.49% vs 72.73%, χ²=4.923, P=0.027) and excellent functional outcomes (68.92% vs 47.27%, χ²=8.412, P=0.004), as well as a significantly lower 1-year ischemic stroke recurrence rate (5.41% vs 16.36%, χ²=4.059, P=0.044). The incidence of sICH (2.70% vs 1.82%) and the 1-year mortality rate (4.05% vs 1.82%) were slightly higher in the EVT group than in the BMT group, but the differences were not statistically significant (χ²=0.000, 0.205; P＞0.999, =0.651). Compared with BMT, EVT significantly reduced the volumes of the infarct core and the hypoperfusion region in the early phase (all P＜0.05). At the 90-day follow-up, the hypoperfusion volume decreased significantly from baseline in both the EVT and BMT groups (Z=-10.504, P＜0.001; Z=-12.819, P＜0.001, respectively). Furthermore, the EVT group exhibited a significantly smaller hypoperfusion volume than the BMT group (Z=-9.897, P＜0.001).

EVT is safe and effective for minor strokes caused by anterior circulation ELVO. Early administration of EVT in these patients can significantly reduce the volumes of the infarct core and hypoperfusion region, thereby preventing END and ischemic stroke recurrence.

To explore the relationship between thrombus migration before mechanical thrombectomy and intravenous thrombolysis in patients with acute ischemic stroke (AIS), as well as its impact on the surgical process and clinical prognosis.

A total of 378 patients with anterior circulation AIS who underwent mechanical thrombectomy in the Department of Neurology, Huaian First Hospital Affiliated to Nanjing Medical University from January 2020 to December 2024 were enrolled in the study. According to the status of thrombus migration [changes in thrombus location between admission cranial CT angiography (CTA) and pre-thrombectomy digital subtraction angiography (DSA)], the patients were divided into the thrombus migration group (n=63) and the thrombus stability group (n=315). T-test, Mann-Whitney U test, and Chi-square test were used to compare the differences in baseline clinical characteristics, surgical process, and 90-day prognosis between the two groups. Multivariate Logistic regression analysis was performed to identify the independent influencing factors of thrombus migration.

There were no statistically significant differences between the two groups in age, gender, past medical history, baseline National Institutes of Health stroke scale (NIHSS) score, baseline Alberta stroke program early CT score (ASPECTS), CTA thrombus location, time intervals, number of thrombectomy attempts, vascular recanalization status, symptomatic intracerebral hemorrhage, and other aspects (all P＞0.05). Compared with the thrombus stability group, the thrombus migration group had a lower proportion of cardioembolic etiology (31.75% vs 46.67%), while a higher proportion of patients received recombinant tissue plasminogen activator (rt-PA) intravenous thrombolysis (49.21% vs 28.57%), underwent stent thrombectomy (88.89% vs 75.24%), and achieved good 90-day prognosis (68.25% vs 52.70%), with statistically significant differences (χ²=4.739, 10.271, 5.613, 5.139; P=0.030, 0.001, 0.018, 0.023). Multivariate Logistic regression analysis showed that cardioembolic etiology was an independent protective factor for thrombus migration (OR=0.543, 95%CI: 0.302 – 0.974, P=0.041), and rt-PA intravenous thrombolysis was an independent risk factor for thrombus migration (OR=2.424, 95%CI: 1.382 – 4.252, P=0.002).

Thrombus migration before mechanical thrombectomy in patients with acute ischemic stroke does not affect vascular recanalization outcomes, but is associated with good 90-day prognosis. Cardioembolic etiology can reduce the risk of thrombus migration, while rt-PA intravenous thrombolysis can increase this risk.

To explore the application effect of 5G⁺ intelligent emergency system in the treatment of patients with acute large vessel occlusion ischemic stroke.

The patients who received endovascular treatment in Zhangzhou Affiliated Hospital of Fujian Medical University from January 1, 2022 to March 1, 2024 were selected and divided into 5G transport group and non-5G transport group according to whether the 5G⁺ intelligent emergency system was used. The treatment time (onset-to-vascular puncture time, arrival-to-vascular puncture time, arrival-to-vascular recanalization time, puncture-to-vascular recanalization time) and clinical outcomes after 90 days were compared between the two groups. The Wilcoxon rank sum test was used for comparison of treatment time between groups. Pearson χ² test was used for comparison of clinical outcomes between groups.

A total of 76 patients were enrolled, including 25 patients in the 5G transport group and 51 patients in the non-5G transport group. Compared with the non-5G transport group, the time from onset to puncture in the 5G transport group was significantly shorter [215.0 (142.0, 355.0) min vs 510.0 (348.0, 789.0) min, Z=4.505, P<0.001], the time from arrival to vascular puncture was significantly shorter [24.0 (19.5, 26.0) min vs 91.0 (79.0, 112.5) min, Z=6.890, P<0.001], and the time from arrival to recanalization was significantly shorter [77.0 (61.0, 101.5) min vs 146.0 (114.0, 169.5) min, Z=4.732, P<0.001]. There was no significant difference in the time from puncture to vascular recanalization between the two groups (P>0.05). Functional independence was better in the 5G transport group than in the non-5G transport group after 90 days (modified Rankin scale score 0 - 3: 68.00% vs 35.29%, χ²=7.223, P=0.007), while the symptomatic intracranial hemorrhage rate (0 vs 17.65%, χ²=5.004, P=0.025) and 90-day mortality rate (0 vs 33.33%, χ²=10.734, P=0.001) were significantly lower.

The 5G⁺ intelligent emergency system can effectively shorten the treatment time and significantly improve the prognosis of patients with acute large vessel occlusive ischemic stroke by optimizing the treatment process of stroke.

To explore the effect of an internet plus (internet+) transitional care strategy led by brain-heart health managers on self-management behaviors of stroke patients.

Using convenience sampling, stroke patients hospitalized at the First People's Hospital of Changzhou from August to December 2022 were included as the study subjects. Patients were randomly divided into a control group and an intervention group (35 cases each) using a random number table. The control group received routine health education and follow-up strategies. The intervention group received individualized, full-process in-hospital health management and an internet+ transitional care strategy led by dedicated brain-heart health managers; the intervention lasted until 12 months after discharge. Univariate analysis (independent samples t-test, non-parametric rank-sum test, or χ² test) was used to compare differences between the two groups in: (1) self-management behavior, blood lipid target achievement rate, and National Institutes of Health stroke scale (NIHSS) scores within 24 hours of admission, on the day before discharge, and at 12 months after discharge; and (2) blood pressure/blood glucose target achievement rates and modified Rankin scale (mRS) scores within 24 hours of admission, on day 1 before discharge, and at 3, 6, and 12 months after discharge.

Finally, 34 patients in each group completed the study. On the day before discharge and at 12 months after discharge, the self-management behavior scale for stroke scores in the intervention group were significantly higher than those in the control group (221.38±15.34 vs 204.88±19.07; 241.97±10.77 vs 188.56±26.04; t=3.931, 11.054, both P<0.001). At 12 months after discharge, the blood pressure and blood lipid target achievement rates in the intervention group were higher than those in the control group (88.24% vs 55.88%, χ²=8.838, P=0.003; 79.41% vs 44.12%, χ²=8.967, P=0.006). The blood glucose target achievement rates in the intervention group were higher than those in the control group at 3, 6, and 12 months after discharge (91.18% vs 70.59%, χ²=4.660, P=0.031; 85.29% vs 64.71%, χ²=3.843, P=0.049; 85.29% vs 41.18%, χ²=14.233, P<0.001). No statistically significant differences were found in mRS scores between the intervention and control groups on the day before discharge and at 3, 6, and 12 months after discharge (all P>0.05). No statistically significant differences were found in NIHSS scores between the intervention and control groups on the day before discharge and at 12 months after discharge (both P>0.05).

The internet+ transitional care strategy led by brain-heart health managers can significantly improve the self-management behavior of stroke patients and enhance their blood pressure, blood glucose, and blood lipid target achievement rates.

To systematically evaluate and compare the diagnostic performance of different artificial intelligence (AI) models in predicting poor functional outcomes in patients with acute ischemic stroke, with a particular focus on the trade-off between sensitivity and specificity across models.

PubMed, Web of Science, Embase, The Cochrane Library, China National Knowledge Infrastructure, and Wanfang databases were systematically searched to identify studies applying AI-based models for functional outcome prediction after stroke. According to modeling principles, the models were categorized into regression models, single decision trees, random forest (RF) models, boosting ensemble models, support vector machines (SVM), deep learning (DL) models, and other machine learning approaches. The 2×2 contingency table data were extracted or calculated. Diagnostic test accuracy Meta-analysis was performed to pool sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the area under the receiver operating characteristic curve (AUROC). Summary receiver operating characteristic (SROC) curves were used to compare the overall diagnostic performance of different model categories.

A total of 32 studies involving 56 458 patients were included. The pooled AUROC values across different model categories ranged from 0.79 to 0.85, indicating moderate to high discriminative ability. RF models achieved the highest AUROC (0.85, 95%CI: 0.82 – 0.88) with relatively high specificity. Boosting ensemble models demonstrated stable and well-balanced diagnostic performance. SVM and DL models showed an advantage in sensitivity. SROC curve comparisons indicated that RF and boosting ensemble models performed better in balancing sensitivity and specificity.

The diagnostic performance of different AI models for predicting poor functional outcomes after stroke varies across modeling approaches. Ensemble learning models, including RF and boosting-based models, showed overall advantages in specificity and in balancing sensitivity and specificity. In clinical practice, model selection should consider specific application scenarios by jointly evaluating sensitivity and specificity, and further studies are warranted to assess the clinical generalizability of AI-based prediction models.

To analyze the research achievements in the field of stroke fall care, and explore the research focus and future trends.

CiteSpace 6.3.R1 software was used to conduct a bibliometric analysis of the authors, institutions, and keywords in literature related to stroke fall care included in CNKI and Web of Science from January 1, 2014 to December 31, 2024.

A total of 199 Chinese articles and 179 English articles were included in this study. Through the analysis of authors and institutions, it was found that no fixed research team had been formed; the institutions with the highest number of publications in the Chinese and English literature were respectively the First Hospital of Shanxi Medical University and the Hong Kong Polytechnic University; the research hotspots and trends were mainly focused on the interaction between psychological and cognitive functions, risk assessment tools, and comprehensive care support systems.

At present, the research field of stroke fall care is developing rapidly. Paying attention to the psychological state of stroke patients and improving their quality of life are expected to become future research trends in this field. In the future, cooperation between authors and institutions should be further strengthened to promote the innovative development of the field of stroke fall care.

Cerebral small vessel disease (CSVD) is the primary cause of ischemic stroke and vascular cognitive impairment, and its occurrence and development are closely related to multi-scale hemodynamic dysfunction from intracranial large arteries to distal microcirculation. However, it is difficult for traditional imaging assessment methods to achieve precise, non-invasive, and integrated quantitative analysis of this complete pathway. In recent years, the rapid development of artificial intelligence (AI) technology has provided innovative tools to solve this problem. This article systematically reviews the research progress and clinical translation status of AI-empowered hemodynamic assessment of CSVD. First, we elaborate on the technical support role of AI in the pre-processing steps such as automated segmentation, noise reduction, and registration of multimodal imaging data. Then, we focus on the application progress of AI in the precise quantification of key parameters such as intracranial large artery blood flow velocity and wall shear stress, cerebral microcirculation perfusion heterogeneity, cerebral autoregulatory function, three-dimensional morphology of perforating arteries, and blood-brain barrier permeability. Finally, we summarize the predictive value of AI through multi-scale information fusion in the subtype differentiation of CSVD, cognitive decline, and risk stratification of disease progression. Based on this, we further explore the current core bottlenecks in clinical translation, including: the lack of structured human-factor usability verification for explainable AI technology, the potential for automated biases triggered by post hoc explanations, the mismatch between the traditional evidence-based medical framework and the dynamic evolving AI system paradigm, and the discontinuity of the evidence chain of health technology assessment. Future breakthroughs need to focus on constructing a "surrogate endpoint - intermediate indicator - hard endpoint" hierarchical verification path, placing human-factor verification prior to the AI model development process, and establishing a dynamic regulatory assessment system covering the entire life cycle.

Cardiovascular and cerebrovascular diseases are a serious threat to human health. Endovascular interventional therapy is an important means for treating these diseases. In recent years, the emergence and application of artificial intelligence (AI) technology in the medical field, particularly the rapid development of the clinical application of deep learning, have provided significant opportunities for the innovation and optimization of cardiovascular and cerebrovascular interventional diagnosis and treatment pathways. AI innovation brings a technological revolution and innovative paradigms, reshaping the current landscape and future development paths of cardiovascular and cerebrovascular disease diagnosis and treatment. This article provides a comprehensive overview of the current state of AI technology in the medical field and its latest applications in cardiovascular and cerebrovascular interventional diagnosis and treatment.

Sepsis is one of the leading causes of death among critically ill patients in intensive care units (ICUs). Sepsis-associated encephalopathy (SAE) can lead to long-term cognitive dysfunction and diagnosis and treatment, thereby increasing the economic burden on families and society. Therefore, cerebral function monitoring in septic patients is of great clinical significance. In the ICU setting, the neurological assessment of septic patients is often hindered by multiple confounding factors. Electroencephalogram (EEG) plays a critical role in the assessment of SAE due to its non-invasive nature and high sensitivity. Therefore, this article focuses on the specific advances in the application of EEG in SAE, aiming to provide a methodological basis and evidence-based support for the early identification and prediction of SAE.