首页|Heart Institute (InCor) Reports Findings in Artificial Intelligence (Explainable artificial intelligence in deep learning-based detection of aortic elongation o n chest X-ray images)
Heart Institute (InCor) Reports Findings in Artificial Intelligence (Explainable artificial intelligence in deep learning-based detection of aortic elongation o n chest X-ray images)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
New research on Artificial Intelligenc e is the subject of a report. According to news reporting originating in Sao Pau lo, Brazil, by NewsRx journalists, research stated, "Aortic elongation can resul t from age-related changes, congenital factors, aneurysms, or conditions affecti ng blood vessel elasticity. It is associated with cardiovascular diseases and se vere complications like aortic aneurysms and dissection." The news reporters obtained a quote from the research from Heart Institute (InCo r), "We assess qualitatively and quantitatively explainable methods to understan d the decisions of a deep learning model for detecting aortic elongation using c hest X-ray (CXR) images. In this study, we evaluated the performance of deep lea rning models (DenseNet and EfficientNet) for detecting aortic elongation using t ransfer learning and fine-tuning techniques with CXR images as input. EfficientN et achieved higher accuracy (86.7% 2.1), precision (82.7% 2.7), specificity (89.4% 1.7), F1 score (82.5% 2.9), and area under the receiver operating characteristic (92.7% 0.6) but lower sensitivity (82.3% 3.2) compared with DenseNet. To gain insights into the decision-making process of these models, we employed gradient- weighted class activation mapping and local interpretable model-agnostic explana tions explainability methods, which enabled us to identify the expected location of aortic elongation in CXR images. Additionally, we used the pixel-flipping me thod to quantitatively assess the model interpretations, providing valuable insi ghts into model behaviour. Our study presents a comprehensive strategy for analy sing CXR images by integrating aortic elongation detection models with explainab le artificial intelligence techniques."
NETL
NSTL
