RESUMEN
OBJECTIVE: This study aimed to develop a deep learning (DL) model to improve the diagnostic performance of EIC and ASPECTS in acute ischemic stroke (AIS). METHODS: Acute ischemic stroke patients were retrospectively enrolled from 5 hospitals. We proposed a deep learning model to simultaneously segment the infarct and estimate ASPECTS automatically using baseline CT. The model performance of segmentation and ASPECTS scoring was evaluated using dice similarity coefficient (DSC) and ROC, respectively. Four raters participated in the multi-reader and multicenter (MRMC) experiment to fulfill the region-based ASPECTS reading under the assistance of the model or not. At last, sensitivity, specificity, interpretation time and interrater agreement were used to evaluate the raters' reading performance. RESULTS: In total, 1391 patients were enrolled for model development and 85 patients for external validation with onset to CT scanning time of 176.4 ± 93.6 min and NIHSS of 5 (IQR 2-10). The model achieved a DSC of 0.600 and 0.762 and an AUC of 0.876 (CI 0.846-0.907) and 0.729 (CI 0.679-0.779), in the internal and external validation set, respectively. The assistance of the DL model improved the raters' average sensitivities and specificities from 0.254 (CI 0.22-0.26) and 0.896 (CI 0.884-0.907), to 0.333 (CI 0.301-0.345) and 0.915 (CI 0.904-0.926), respectively. The average interpretation time of the raters was reduced from 219.0 to 175.7 s (p = 0.035). Meanwhile, the interrater agreement increased from 0.741 to 0.980. CONCLUSIONS: With the assistance of our proposed DL model, radiologists got better performance in the detection of AIS lesions on NCCT.
RESUMEN
Previous studies reported discrepant white matter diffusivity in mild traumatic brain injury (mTBI) on the base of Glasgow Coma Scale, which are unreliable for some TBI severity indicators and the frequency of missing documentation in the medical record. In the present study, we adopted the Mayo classification system for TBI severity. In this system, the mTBI is also divided into two groups as "probable and symptomatic" TBI. We aimed to investigate altered microstructural integrity in symptomatic acute TBI (<1 week) by using tract-based spatial statics (TBSS) approach. A total of 12 patients and 13 healthy volunteers were involved and underwent MRI scans including conventional scan, and SWI and DTI. All the patients had no visible lesions by using conventional and SWI neuroimaging techniques, while showing widespread declines in the fractional anisotropy (FA) of gray matter and white matter throughout the TBSS skeleton, particularly in the limbic-subcortical structures. By contrast, symptomatic TBI patients showed no significant enhanced changes in FA compared to the healthy controls. A better understanding of the acute changes occurring following symptomatic TBI may increase our understanding of neuroplasticity and continuing degenerative change, which, in turn, may facilitate advances in management and intervention.