Abstract：
Suboptimal chest radiographs (CXR) can limit interpretation of critical findings. Radiologist-trained AI models were evaluated for differentiating suboptimal(sCXR) and optimal(oCXR) chest radiographs.
Our IRB-approved study included 3278 CXRs from adult patients (mean age 55 ± 20 years) identified from a retrospective search of CXR in radiology reports from 5 sites. A chest radiologist reviewed all CXRs for the cause of suboptimality. The de-identified CXRs were uploaded into an AI server application for training and testing 5 AI models. The training set consisted of 2202 CXRs (n = 807 oCXR; n = 1395 sCXR) while 1076 CXRs (n = 729 sCXR; n = 347 oCXR) were used for testing. Data were analyzed with the Area under the curve (AUC) for the model\'s ability to classify oCXR and sCXR correctly.
For the two-class classification into sCXR or oCXR from all sites, for CXR with missing anatomy, AI had sensitivity, specificity, accuracy, and AUC of 78%, 95%, 91%, 0.87(95% CI 0.82-0.92), respectively. AI identified obscured thoracic anatomy with 91% sensitivity, 97% specificity, 95% accuracy, and 0.94 AUC (95% CI 0.90-0.97). Inadequate exposure with 90% sensitivity, 93% specificity, 92% accuracy, and AUC of 0.91 (95% CI 0.88-0.95). The presence of low lung volume was identified with 96% sensitivity, 92% specificity, 93% accuracy, and 0.94 AUC (95% CI 0.92-0.96). The sensitivity, specificity, accuracy, and AUC of AI in identifying patient rotation were 92%, 96%, 95%, and 0.94 (95% CI 0.91-0.98), respectively.
The radiologist-trained AI models can accurately classify optimal and suboptimal CXRs. Such AI models at the front end of radiographic equipment can enable radiographers to repeat sCXRs when necessary.
Our IRB-approved study included 3278 CXRs from adult patients (mean age 55 ± 20 years) identified from a retrospective search of CXR in radiology reports from 5 sites. A chest radiologist reviewed all CXRs for the cause of suboptimality. The de-identified CXRs were uploaded into an AI server application for training and testing 5 AI models. The training set consisted of 2202 CXRs (n = 807 oCXR; n = 1395 sCXR) while 1076 CXRs (n = 729 sCXR; n = 347 oCXR) were used for testing. Data were analyzed with the Area under the curve (AUC) for the model\'s ability to classify oCXR and sCXR correctly.
For the two-class classification into sCXR or oCXR from all sites, for CXR with missing anatomy, AI had sensitivity, specificity, accuracy, and AUC of 78%, 95%, 91%, 0.87(95% CI 0.82-0.92), respectively. AI identified obscured thoracic anatomy with 91% sensitivity, 97% specificity, 95% accuracy, and 0.94 AUC (95% CI 0.90-0.97). Inadequate exposure with 90% sensitivity, 93% specificity, 92% accuracy, and AUC of 0.91 (95% CI 0.88-0.95). The presence of low lung volume was identified with 96% sensitivity, 92% specificity, 93% accuracy, and 0.94 AUC (95% CI 0.92-0.96). The sensitivity, specificity, accuracy, and AUC of AI in identifying patient rotation were 92%, 96%, 95%, and 0.94 (95% CI 0.91-0.98), respectively.
The radiologist-trained AI models can accurately classify optimal and suboptimal CXRs. Such AI models at the front end of radiographic equipment can enable radiographers to repeat sCXRs when necessary.
摘要：
目的：不理想的胸部X光片（CXR）可能会限制对关键发现的解释。评估了放射科医生训练的AI模型，以区分次优（sCXR）和最佳（oCXR）胸片。
方法：我们的IRB批准的研究包括3278个来自成人患者（平均年龄55±20岁）的CXR，这些患者来自5个地点的放射学报告中的CXR回顾性检索。胸部放射科医生检查了所有CXR的次优原因。去识别的CXR被上传到AI服务器应用程序中，用于训练和测试5个AI模型。训练集由2202个CXR（n=807oCXR；n=1395sCXR）组成，而1076个CXR（n=729sCXR；n=347oCXR）用于测试。用曲线下面积(AUC)分析模型正确分类oCXR和sCXR的能力的数据。
结果：对于所有站点的sCXR或oCXR的两类分类，对于解剖缺失的CXR，AI有敏感性,特异性，准确度,AUC为78%，95%,91%,0.87（95%CI0.82-0.92），分别。AI识别出模糊的胸部解剖结构，灵敏度为91%，97%的特异性，95%的准确度,和0.94AUC（95%CI0.90-0.97）。曝光不足，灵敏度为90%，93%的特异性,92%的准确度，AUC为0.91（95%CI0.88-0.95）。以96%的灵敏度确定存在低肺容量，92%的特异性，93%的准确度,和0.94AUC（95%CI0.92-0.96）。敏感性,特异性，准确度,识别患者轮换的AIAUC为92%，96%,95%,和0.94（95%CI0.91-0.98），分别。
结论：放射科医师训练的AI模型可以准确地对最佳和次优CXR进行分类。射线照相设备前端的这种AI模型可以使射线照相师在必要时重复sCXR。
方法：我们的IRB批准的研究包括3278个来自成人患者（平均年龄55±20岁）的CXR，这些患者来自5个地点的放射学报告中的CXR回顾性检索。胸部放射科医生检查了所有CXR的次优原因。去识别的CXR被上传到AI服务器应用程序中，用于训练和测试5个AI模型。训练集由2202个CXR（n=807oCXR；n=1395sCXR）组成，而1076个CXR（n=729sCXR；n=347oCXR）用于测试。用曲线下面积(AUC)分析模型正确分类oCXR和sCXR的能力的数据。
结果：对于所有站点的sCXR或oCXR的两类分类，对于解剖缺失的CXR，AI有敏感性,特异性，准确度,AUC为78%，95%,91%,0.87（95%CI0.82-0.92），分别。AI识别出模糊的胸部解剖结构，灵敏度为91%，97%的特异性，95%的准确度,和0.94AUC（95%CI0.90-0.97）。曝光不足，灵敏度为90%，93%的特异性,92%的准确度，AUC为0.91（95%CI0.88-0.95）。以96%的灵敏度确定存在低肺容量，92%的特异性，93%的准确度,和0.94AUC（95%CI0.92-0.96）。敏感性,特异性，准确度,识别患者轮换的AIAUC为92%，96%,95%,和0.94（95%CI0.91-0.98），分别。
结论：放射科医师训练的AI模型可以准确地对最佳和次优CXR进行分类。射线照相设备前端的这种AI模型可以使射线照相师在必要时重复sCXR。
