人工智能助力新型冠状病毒肺炎胸部CT诊断的应用

Chest CT with Artificial Intelligence in the Diagnosis of Coronavirus Disease 2019

新型冠状病毒肺炎(Coronavirus Disease 2019,COVID-19)自发病以来迅速蔓延,病症隐匿,但传播范围广。现阶段病毒病原学检测为临床确诊标准,但其阳性检出率低且结果易受多种因素影响。CT因检查快捷且阳性率高等原因已被推荐作为COVID-19临床诊断的主要依据。然而,面对全国COVID-19筛查人员众多,筛查医院集中,存在人传人感染风险,影像科医生观片工作量大且对于COVID-19肺部诊断缺乏客观定量评价指标等现状,使用人工智能的方法分析胸部CT影像,可有效提升工作效率,提高检出率,减少患者等待时间,优化诊断流程,在COVID-19早期发现、精准诊断方面意义重大。

Coronavirus disease 2019(COVID-19)has spread rapidly since its onset.At present,the diagnostic standard for COVID-19 is the coronavirus etiologic results that has low positive detection rate and affects by many factors.With fast and high positive detection rate,CT has been recommended as the main imaging method for the clinical diagnosis for COVID-19.However,the patients waiting for COVID-19 screening is increasing and the designated hospital is limited,the work load and the risk infection of radiologists are high.Furthermore,the objective quantitative evaluation standards are yet to be established.Thus,use of artificial intelligence with professional lung analysis may effectively improve the image review efficiency,increase the detection rate,reduce patient waiting time and standardize and optimize the diagnosis process,which is of great significance in early detection and accurate diagnosis of COVID-19.