摘要
一位新闻记者兼机器人与机器学习的工作人员新闻编辑每日新闻-人工智能的新研究是一篇报道的主题。根据NewsRx记者来自意大利罗马的消息，研究表明：“为缺血性中风患者制定可靠的预后仍然是一项具有挑战性的任务。我们的目标是开发一种人工智能模型，能够在中风后24小时内根据NIHSS制定独立的预后。”我们的新闻记者从Fondazione Policlin ico Universitario A. Gemelli IRCCS的研究中获得了一句话，“将74例急性缺血性脑卒中患者分为训练队列(597)和测试队列(197)。在前24小时收集临床和仪器数据。我们评估了四种机器学习模型(随机森林、-最近邻居、支持向量机。XGBoost)在预测出院时的NIHSS时，既根据出院和入院之间的变化（回归方法），也根据S级，即NIHSS 0-5、6-10、11-20，>20(验证者方法)。我们使用Shapley相加解释值来加权特征对预测的影响。XGBoost成为表现最好的模型。分类器和回归器方法在准确度(80%vs 75%)和F1得分(79%vs 77%)方面表现相似。然而，回归模型预测极重度脑卒中患者(NIHSS>20)的预后精度较高(85%vs 68%),入院时和24h时NIHSS、24h GCS、心率、CT急性缺血性病变和TICI评分是影响预后的最主要特征,本文采用人工智能工具,能够不断学习和提高预后。可以改善护理路径，支持中风医生与病人和CA注册者沟通。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Artificial Intelligenc e is the subject of a report. According to news originating from Rome, Italy, by NewsRx correspondents, research stated, “Formulating reliable prognosis for isc hemic stroke patients remains a challenging task. We aimed to develop an artific ial intelligence model able to formulate in the first 24 h after stroke an indiv idualized prognosis in terms of NIHSS.” Our news journalists obtained a quote from the research from Fondazione Policlin ico Universitario A. Gemelli IRCCS, “Seven hundred ninety four acute ischemic st roke patients were divided into a training (597) and testing (197) cohort. Clini cal and instrumental data were collected in the first 24 h. We evaluated the per formance of four machine-learning models (Random Forest, -Nearest Neighbors, Sup port Vector Machine, XGBoost) in predicting NIHSS at discharge both in terms of variation between discharge and admission (regressor approach) and in terms of s everity class namely NIHSS 0-5, 6-10, 11-20, >20 (classi fier approach). We used Shapley Additive exPlanations values to weight features impact on predictions. XGBoost emerged as the best performing model. The classif ier and regressor approaches perform similarly in terms of accuracy (80% vs 75%) and f1-score (79% vs 77%) respec tively. However, the regressor has higher precision (85% vs 68% ) in predicting prognosis of very severe stroke patients (NIHSS > 20). NIHSS at admission and 24 hours, GCS at 24 hours, heart rate, acute ischem ic lesion on CT-scan and TICI score were the most impacting features on the pred iction. Our approach, which employs an artificial intelligence based-tool, inher ently able to continuously learn and improve its performance, could improve care pathway and support stroke physicians in the communication with patients and ca regivers.”
NETL
NSTL