Acceso usuarios
Sebastian Florez, Santiago Gomez, Julian Garcia, Fabio Martínez
l accidente cerebrovascular (ACV), segunda causa de muerte en el mundo, requiere un diagnóstico temprano para un pronóstico favorable. Las imágenes de TC tienen limitaciones, especialmente en la identificación de lesiones agudas. Este trabajo introduce una novedosa representación profunda que utiliza datos multimodales TC y mapas paramétricos de perfusión para segmentar lesiones de ACV. La arquitectura sigue una representación autocodificadora que fuerza la atención sobre la geometría del ACV a través de módulos aditivos de atención cruzada. Además, se propone un entrenamiento en cascada para generar mapas de perfusión sintéticos que complementen las entradas multimodales, refinando la segmentación de las lesiones en cada etapa del procesamiento y apoyando el análisis observacional del experto. El enfoque propuesto fue validado en el conjunto de datos ISLES 2018 con 92 estudios; el método supera a las técnicas clásicas con una puntuación Dice de .66 y una precisión de .67.
Stroke is the second leading cause of mortality worldwide. Immediate attention and diagnosis play a crucial role in patient prognosis. Nowadays, computed tomography (CT) is the most utilized diagnostic imaging for early analysis and lesion stroke detection. Nonetheless, acute lesions are not visible on CT, and it is only possible to use this modality in screening analysis, discarding other neurological affectations. Expert radiologists can observe stroke lesions in advanced stages (subacute and chronic), as hypodense regions, but with limited sensibility and remarked subject variability. Computational strategies have been addressed to support lesion segmentation, following deep autoencoders and multimodal inputs. However, these strategies remain limited due to the high variability in the appearance and geometry of stroke lesions. This work introduces a novel deep representation that uses multimodal inputs from CT studies and parametric maps, computed from perfusion (CTP), to retrieve stroke lesions. The architecture follows an autoencoder deep representation, that forces attention on the geometry of stroke through additive cross-attention modules. Besides, a cascade train is herein proposed to generate synthetic perfusion maps that complement multimodal inputs and help with stroke lesion refinement at each stage of processing. The proposed approach brings saliency maps that support observational expert analysis, about lesion localization, but also lead with automatic shape estimation of the stroke. The proposed approach was validated on the ISLES 2018 public dataset with a total of 92 studies that include the annotation of an expert radiologist. The proposed approach achieves a Dice score of 0.66 and a precision of 0.67, outperforming classical autoencoder approximations.
