Comparison between healthcare quality in primary stroke centers and comprehensive stroke centers for acute stroke patients: evidence from the Chinese Stroke Center Alliance

Deep Learning-Based Automatic Classification of Ischemic Stroke Subtype Using Diffusion-Weighted Images

BACKGROUND AND PURPOSE

Accurate classification of ischemic stroke subtype is important for effective secondary prevention of stroke. We used diffusion-weighted image (DWI) and atrial fibrillation (AF) data to train a deep learning algorithm to classify stroke subtype.

METHODS

Model development was done in 2,988 patients with ischemic stroke from three centers by using U-net for infarct segmentation and EfficientNetV2 for subtype classification. Experienced neurologists (n=5) determined subtypes for external test datasets, while establishing a consensus for clinical trial datasets. Automatically segmented infarcts were fed into the model (DWI-only algorithm). Subsequently, another model was trained, with AF included as a categorical variable (DWI+AF algorithm). These models were tested: (1) internally against the opinion of the labeling experts, (2) against fresh external DWI data, and (3) against clinical trial dataset.

RESULTS

In the training-and-validation datasets, the mean (±standard deviation) age was 68.0±12.5 (61.1% male). In internal testing, compared with the experts, the DWI-only and the DWI+AF algorithms respectively achieved moderate (65.3%) and near-strong (79.1%) agreement. In external testing, both algorithms again showed good agreements (59.3%-60.7% and 73.7%-74.0%, respectively). In the clinical trial dataset, compared with the expert consensus, percentage agreements and Cohen's kappa were respectively 58.1% and 0.34 for the DWI-only vs. 72.9% and 0.57 for the DWI+AF algorithms. The corresponding values between experts were comparable (76.0% and 0.61) to the DWI+AF algorithm.

CONCLUSION

Our model trained on a large dataset of DWI (both with or without AF information) was able to classify ischemic stroke subtypes comparable to a consensus of stroke experts.

Impact of Off-Hour Admission on In-Hospital Outcomes for Patients With Stroke Receiving Reperfusion Therapy in China

BACKGROUND

The structure and staffing of hospitals greatly impact patient outcomes, with frequent changes occurring during nights and weekends. This retrospective cohort study assessed the impact of admission timing on in-hospital management and outcomes for patients with stroke receiving reperfusion therapy in China using data from a nationwide registry.

METHODS

Data from patients receiving reperfusion therapy were extracted from the Chinese Stroke Center Alliance. Hospital admission time was categorized according to day/evening versus night and weekday versus weekend. Primary outcomes were in-hospital death or discharge against medical advice, hemorrhage transformation, early neurological deterioration, and major adverse cardiovascular events. Logistic regression was performed to compare in-hospital management performance and outcomes based on admission time categories.

RESULTS

Overall, 42 381 patients received recombinant tissue-type plasminogen activator (r-tPA) therapy, and 5224 underwent endovascular treatment (EVT). Patients admitted during nighttime had a higher probability of receiving r-tPA therapy within 4.5 hours from onset or undergoing EVT within 6 hours from onset compared with those admitted during day/evening hours (adjusted odds ratio, 1.04 [95% CI, 1.01-1.08]; P=0.021; adjusted odds ratio, 1.72 [95% CI, 1.59-1.86]; P<0.001, respectively). However, no significant difference was observed between weekend and weekday admissions for either treatment. No notable differences were noted between weekends and weekdays or nighttime and daytime periods in door-to-needle time for r-tPA or door-to-puncture time for EVT initiation. Furthermore, weekend or nighttime admission did not have a significant effect on the primary outcomes of r-tPA therapy or EVT. Nevertheless, in patients undergoing EVT, a higher incidence of pneumonia was observed among those admitted at night compared with those admitted during day/evening hours (adjusted odds ratio, 1.22 [95% CI, 1.05-1.42]; P=0.011).

CONCLUSIONS

Patients admitted at nighttime were more likely to receive r-tPA therapy or EVT within the time window recommended in the guidelines. However, patients receiving EVT admitted at night had an increased risk of pneumonia.