CPAISD: Core-penumbra acute ischemic stroke dataset
D. Umerenkov, S. Kudin, M. Peksheva, D. Pavlov
TL;DR
This paper tackles the challenge of early acute ischemic stroke detection on non-contrast CT by introducing CPAISD, a dataset with radiologist-annotated ischemic core and penumbra on NCCT, linked to CTA/CTP references. It documents a threefold data structure, 112 hyperacute AIS studies, and a baseline segmentation model based on a Feature Pyramid Network with EfficientNet-B0 that yields a Dice3D of 0.2183 on the test set. The key contributions are the open NCCT core-penumbra segmentation dataset, detailed data acquisition/format, and a replicable baseline model to spur further research and tool development. The dataset aims to enable rapid, accessible AI-driven stroke diagnostics, potentially improving triage, treatment decisions, and patient outcomes in resource-limited settings.
Abstract
We introduce the CPAISD: Core-Penumbra Acute Ischemic Stroke Dataset, aimed at enhancing the early detection and segmentation of ischemic stroke using Non-Contrast Computed Tomography (NCCT) scans. Addressing the challenges in diagnosing acute ischemic stroke during its early stages due to often non-revealing native CT findings, the dataset provides a collection of segmented NCCT images. These include annotations of ischemic core and penumbra regions, critical for developing machine learning models for rapid stroke identification and assessment. By offering a carefully collected and annotated dataset, we aim to facilitate the development of advanced diagnostic tools, contributing to improved patient care and outcomes in stroke management. Our dataset's uniqueness lies in its focus on the acute phase of ischemic stroke, with non-informative native CT scans, and includes a baseline model to demonstrate the dataset's application, encouraging further research and innovation in the field of medical imaging and stroke diagnosis.