计算机断层扫描间质性肺异常的定量分析预测肺立体定向体部放疗后症状性放射性肺炎
本研究利用人工智能定量CT图像分析软件(AIQCT)对207例接受肺立体定向体部放疗(SBRT)的I期肺癌患者的高分辨率CT图像进行自动标注,分析间质性肺异常等定量指标与症状性放射性肺炎(SRP)的关联,并基于递归分割分析建立预测模型。结果显示,NL-Dmean ≥ 6.6 Gy为高危组(n=8),NL-Dmean < 6.6 Gy且Br-volume ≥ 2.5%为中危组(n=13),NL-Dmean < 6.6 Gy且Br-volume < 2.5%为低危组(n=133);训练队列SRP发生率分别为62.5%、38.4%和7.5%,测试队列分别为50.0%、27.3%和5.0%。
要点
- AIQCT可自动量化间质性肺异常等CT特征,并识别与SRP相关的指标。
- 基于NL-Dmean和Br-volume的RPA模型将患者分为高危、中危、低危三组。
- 训练队列三组SRP发生率分别为62.5%、38.4%和7.5%,测试队列分别为50.0%、27.3%和5.0%。
关键数值与推荐
- 高危组:NL-Dmean ≥ 6.6 Gy(n=8),训练队列SRP发生率62.5%,测试队列50.0%。
- 中危组:NL-Dmean < 6.6 Gy且Br-volume ≥ 2.5%(n=13),训练队列SRP发生率38.4%,测试队列27.3%。
- 低危组:NL-Dmean < 6.6 Gy且Br-volume < 2.5%(n=133),训练队列SRP发生率7.5%,测试队列5.0%。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND AND PURPOSE: Symptomatic radiation pneumonitis (SRP) is a complication of thoracic stereotactic body radiotherapy (SBRT). As visual assessments pose limitations, artificial intelligence-based quantitative computed tomography image analysis software (AIQCT) may help predict SRP risk. We aimed to evaluate high-resolution computed tomography (HRCT) images with AIQCT to develop a predictive model for SRP. MATERIALS AND METHODS: AIQCT automatically labelled HRCT images of patients treated with SBRT for stage I lung cancer according to lung parenchymal pattern. Quantitative data including the volume and mean dose (Dmean) were obtained for reticulation + honeycombing (Ret + HC), consolidation + ground-glass opacities, bronchi (Br), and normal lungs (NL). After associations between AIQCT's quantified metrics and SRP were investigated, we developed a predictive model using recursive partitioning analysis (RPA) for the training cohort and assessed its reproducibility with the testing cohort. RESULTS: Overall, 26 of 207 patients developed SRP. There were significant between-group differences in the Ret + HC, Br-volume, and NL-Dmean in patients with and without SRP. RPA identified the following risk groups: NL-Dmean ≥ 6.6 Gy (high-risk, n = 8), NL-Dmean < 6.6 Gy and Br-volume ≥ 2.5 % (intermediate-risk, n = 13), and NL-Dmean < 6.6 Gy and Br-volume < 2.5 % (low-risk, n = 133). The incidences of SRP in these groups within the training cohort were 62.5, 38.4, and 7.5 %; and in the testing cohort 50.0, 27.3, and 5.0 %, respectively. CONCLUSION: AIQCT identified CT features associated with SRP. A predictive model for SRP was proposed based on AI-detected Br-volume and the NL-Dmean.
引用
- Yoneyama M, Matsuo Y, Kishi N, et al. Quantitative analysis of interstitial lung abnormalities on computed tomography to predict symptomatic radiation pneumonitis after lung stereotactic body radiotherapy. Radiother Oncol. 2024. PMID: 38917885. doi:10.1016/j.radonc.2024.110408
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。