人工智能软件检测胸部CT肋骨骨折的效能探讨

Effectiveness study of artificial intelligence software in detecting rib fractures on chest CT images

目的探讨评价人工智能(artificial intelligence,AI)软件辅助放射科医师在胸部CT图像检测肋骨骨折的应用效能。方法选取因胸部外伤行急诊胸部CT检查的患者301例。将6位放射科医师分为低年资组(≤5年)和高年资组(≥10年)对CT图像阅片;1个月后,医师+AI软件联合进行第2次阅片。采用卡方检验比较不同组间检测肋骨骨折敏感度的差异。配对样本t检验比较结合AI前后医师诊断时间是否有统计学意义。结果低年资组+AI、高年资组+AI结合阅片在全部肋骨骨折、错位型、隐匿型肋骨骨折的检测敏感度均优于医师单独阅片,差异有统计学意义(P<0.001)。应用AI阅片后,高低年资组医师的阅片时间均明显缩短,差异有统计学意义(P<0.001)。结论AI软件辅助诊断有助于放射科医师提高肋骨骨折检测效能,优化诊断流程并有效减少诊断时间。

Objective To evaluate the effectiveness of artificial intelligence(AI)software in assisting radiologists in detecting rib fractures on chest CT images.Methods 301 cases of emergency chest CT due to chest trauma were collected.Six imaging physicians were divided into a low seniority group(≤5 years)and a high seniority group(≥10 years)to review CT images;After one month,the physician and AI software jointly reviewed the film for the second time.Chi-square test was used to compare the sensitivity differences between different groups in detecting rib fractures.Paired sample t-tests were used to compare whether there was statistical significance in physician diagnostic time before and after AI.Results The detection sensitivity of low seniority group+AI and high seniority group+AI combined with film reading in all rib fractures,dislocation type,and concealed rib fractures was better than that of physicians alone,with a statistically significant difference(P<0.001).After using AI to read films,the reading time of senior and junior physicians was significantly shortened,with a statistically significant difference(P<0.001)Conclusion AI software-assisted diagnosis can help radiologists improve the detection efficiency of rib fractures,optimize the diagnostic process,and effectively reduce diagnostic time.