Department of Vascular Surgery, The Permanente Medical Group, South San Francisco, Calif; Division of Research, Kaiser Permanente, Oakland, Calif.
Division of Research, Kaiser Permanente, Oakland, Calif.
J Vasc Surg. 2021 Dec;74(6):1937-1947.e3. doi: 10.1016/j.jvs.2021.05.054. Epub 2021 Jun 25.
Investigation of asymptomatic carotid stenosis treatment is hindered by the lack of a contemporary population-based disease cohort. We describe the use of natural language processing (NLP) to identify stenosis in patients undergoing carotid imaging.
Adult patients with carotid imaging between 2008 and 2012 in a large integrated health care system were identified and followed through 2017. An NLP process was developed to characterize carotid stenosis according to the Society of Radiologists in Ultrasound (for ultrasounds) and North American Symptomatic Carotid Endarterectomy Trial (NASCET) (for axial imaging) guidelines. The resulting algorithm assessed text descriptors to categorize normal/non-hemodynamically significant stenosis, moderate or severe stenosis as well as occlusion in both carotid ultrasound (US) and axial imaging (computed tomography and magnetic resonance angiography [CTA/MRA]). For US reports, internal carotid artery systolic and diastolic velocities and velocity ratios were assessed and matched for laterality to supplement accuracy. To validate the NLP algorithm, positive predictive value (PPV or precision) and sensitivity (recall) were calculated from simple random samples from the population of all imaging studies. Lastly, all non-normal studies were manually reviewed for confirmation for prevalence estimates and disease cohort assembly.
A total of 95,896 qualifying index studies (76,276 US and 19,620 CTA/MRA) were identified among 94,822 patients including 1059 patients who underwent multiple studies on the same day. For studies of normal/non-hemodynamically significant stenosis arteries, the NLP algorithm showed excellent performance with a PPV of 99% for US and 96.5% for CTA/MRA. PPV/sensitivity to identify a non-normal artery with correct laterality in the CTA/MRA and US samples were 76.9% (95% confidence interval [CI], 74.1%-79.5%)/93.1% (95% CI, 91.1%-94.8%) and 74.7% (95% CI, 69.3%-79.5%)/94% (95% CI, 90.2%-96.7%), respectively. Regarding cohort assembly, 15,522 patients were identified with diseased carotid artery, including 2674 exhibiting equal bilateral disease. This resulted in a laterality-specific cohort with 12,828 moderate, 5283 severe, and 1895 occluded arteries and 326 diseased arteries with unknown stenosis. During follow-up, 30.1% of these patients underwent 61,107 additional studies.
Use of NLP to detect carotid stenosis or occlusion can result in accurate exclusion of normal/non-hemodynamically significant stenosis disease states with more moderate precision with lesion identification, which can substantially reduce the need for manual review. The resulting cohort allows for efficient research and holds promise for similar reporting in other vascular diseases.
无症状颈动脉狭窄治疗的研究受到缺乏当代基于人群的疾病队列的阻碍。我们描述了使用自然语言处理（NLP）来识别接受颈动脉成像的患者中的狭窄。
在一个大型综合医疗保健系统中，确定了 2008 年至 2012 年间进行颈动脉成像的成年患者，并随访至 2017 年。开发了一种 NLP 流程，根据放射学会超声（用于超声）和北美症状性颈动脉内膜切除术试验（NASCET）（用于轴向成像）指南来描述颈动脉狭窄。由此产生的算法评估了文本描述符，以将正常/非血流动力学显著狭窄、中度或重度狭窄以及颈动脉超声（US）和轴向成像（计算机断层扫描和磁共振血管造影 [CTA/MRA]）中的闭塞分类。对于 US 报告，评估了颈内动脉收缩期和舒张期速度以及速度比，并评估了与侧别匹配，以补充准确性。为了验证 NLP 算法，从所有成像研究的人群中随机抽取简单样本，计算阳性预测值（PPV 或精度）和敏感性（召回率）。最后，对所有非正常研究进行手动审查，以确认患病率估计和疾病队列组装。
在 94822 名患者中确定了 95896 项合格的索引研究（76276 项 US 和 19620 项 CTA/MRA），其中 1059 名患者在同一天接受了多次研究。对于正常/非血流动力学显著狭窄动脉的研究，NLP 算法表现出出色的性能，US 的 PPV 为 99%，CTA/MRA 的 PPV 为 96.5%。在 CTA/MRA 和 US 样本中，正确侧别的非正常动脉的 PPV/灵敏度分别为 76.9%（95%置信区间[CI]，74.1%-79.5%）/93.1%（95%CI，91.1%-94.8%）和 74.7%（95%CI，69.3%-79.5%）/94%（95%CI，90.2%-96.7%）。关于队列组装，确定了 15522 名患有颈动脉疾病的患者，其中 2674 名患者存在双侧相等疾病。这导致了一个具有特定侧别的队列，其中有 12828 例中度、5283 例重度和 1895 例闭塞以及 326 例狭窄程度未知的疾病。在随访期间，这些患者中有 30.1%接受了 61107 项额外研究。
使用 NLP 来检测颈动脉狭窄或闭塞可以准确排除正常/非血流动力学显著狭窄疾病状态，具有更高的中度精度和病变识别，这可以大大减少手动审查的需要。由此产生的队列允许进行有效的研究，并为其他血管疾病的类似报告提供了前景。
