We present StrokeNeXt, a model for stroke classification in 2D Computed Tomography (CT) images. StrokeNeXt employs a dual-branch design with two ConvNeXt encoders, whose features are fused through a lightweight convolutional decoder based on stacked 1D operations, including a bottleneck projection and transformation layers, and a compact classification head. The model is evaluated on a curated dataset of 6,774 CT images, addressing both stroke detection and subtype classification between ischemic and hemorrhage cases. StrokeNeXt consistently outperforms convolutional and Transformer-based baselines, reaching accuracies and F1-scores of up to 0.988. Paired statistical tests confirm that the performance gains are statistically significant, while class-wise sensitivity and specificity demonstrate robust behavior across diagnostic categories. Calibration analysis shows reduced prediction error compared to competing methods, and confusion matrix results indicate low misclassification rates. In addition, the model exhibits low inference time and fast convergence.
Brain stroke is a life-threatening medical condition and a leading cause of adult mortality and long-term disability worldwide, affecting millions of individuals each year [30]. It occurs when cerebral blood flow is disrupted due to vascular blockage or rupture [13,25], depriving brain tissue of oxygen and nutrients and resulting in cellular injury or death [13]. Owing to its abrupt onset and severe clinical consequences, stroke remains a major challenge in emergency medicine and neurology [11].
Clinically, strokes are broadly categorized into ischemic and hemorrhagic types, depending on whether they are caused by vascular blockage or rupture [13]. Typically, initial assessment relies on neuroimaging techniques such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) [1,5], which enable visualization of affected brain regions and discrimination between stroke sub-types, thereby supporting timely therapeutic decisions [1].
However, image interpretation remains complex and highly time-sensitive [7]. Manual analysis demands expert radiologists and often involves labor-intensive procedures to identify subtle lesions or distinguish pathological patterns [1], making it susceptible to human error and interobserver variability [3]. Also, high operational costs and the need for specialized personnel can restrict timely diagnosis in resource-limited environments [6,11]. These challenges have driven increasing interest in computational approaches, particularly Artificial Intelligence (AI), to support and enhance clinical decision-making in stroke assessment.
AI methods, particularly those based on computer vision, enable automated analysis of medical images by learning discriminative visual patterns from large annotated datasets [9,22]. In stroke diagnosis, such models can identify subtle abnormalities in MRI or CT scans with minimal human intervention, supporting more consistent and efficient assessment. Consequently, AI-based solutions have gained increasing adoption in clinical research and practice [10].
Given the critical nature of stroke assessment, effective methods must combine high diagnostic accuracy with computational efficiency [29]. Misclassification of stroke subtypes can lead to inappropriate treatment decisions, increasing the risk of irreversible brain damage or death [6], while excessive computational cost may limit deployment in timecritical clinical settings [2]. Accordingly, balancing reliable predictions with fast inference is essential.
Based on the aforementioned, we present StrokeNeXt (Fig. 1), a Deep Learning (DL) approach for brain stroke classification based on a dual-branch feature extraction architecture. The two branches process the same input image independently, enabling the capture of complementary representations and mitigating information loss typically associated with single-path pipelines. The resulting features are fused through a lightweight decoding module designed to enhance feature integration and generate the final classification. StrokeNeXt is evaluated on a real-world CT dataset, where it demonstrates competitive performance. The main contributions of this work are summarized as follows:
• We present StrokeNeXt, an approach for brain stroke
Stroke identification has been widely explored using diverse paradigms. Early approaches relied on traditional classifiers such as Random Forest [4] and ElasticNet [8], which achieved moderate success in stroke detection but were limited by their dependence on handcrafted features and lack of spatial context. Subsequent efforts shifted toward DL, where CNN-based models such as MobileNetV2 [29] improved automation and spatial feature extraction but still exhibited limited representational capacity and robustness.
Other architectures have been proposed to enhance feature representation. Models such as P-CNN [11], D-UNet [29], and OzNet [20] improve lesion localization through multi-scale encoding or task-specific optimizations. In parallel, 3D approaches, including 3D-CNN [19] and CAD systems [28], exploit volumetric information to capture richer spatial context. However, these methods typically rely on computationally expensive 3D pipelines, increasing complexity and overfitting risk, limiting their suitability for real-time clinical deployment.
Recent work has explored hybrid architectures to model long-range dependencies. For example, StrokeViT [21] combines convolutional networks with Vision Transformers to enhance feature representation and prediction accuracy. However, Transformer-based models generally demand large datasets, high computational cost, and long training times, limiting their practicality in resource-and time-constrained clinical settings.
Overall, existing methods exhibit a trade-off between accuracy, efficiency, and reliability. While some achieve strong classification performance, they often incur high computational cost, whereas others lack proper calibration for time-sensitive clinical deplo
This content is AI-processed based on open access ArXiv data.