深度学习自然语言处理从影像报告中提取肿瘤结局的评估
该回顾性队列研究纳入2406例肺癌患者,利用深度学习模型从影像报告文本中提取癌症存在、好转/缓解和恶化/进展等结局。在109例测试子集中,模型区分上述结局的AUC均>0.90;机器与人工判读的无病生存期、无进展生存期和至好转/缓解时间相似。在1294例储备集中,算法检测的恶化/进展与总生存期下降相关(HR 4.04),好转/缓解与总生存期改善相关(HR 0.41)。
要点
- 深度学习模型从影像报告识别癌症存在、好转/缓解和恶化/进展的AUC均>0.90。
- 机器与人工判读的无病生存期、无进展生存期和至好转/缓解时间相似。
- 算法检测的恶化/进展与总生存期下降相关(HR 4.04),好转/缓解与总生存期改善相关(HR 0.41)。
核心结论
- 深度学习自然语言处理可加速从电子健康记录中提取相关癌症结局,促进快速学习。
研究设计与样本
- 回顾性队列研究,纳入2013年6月26日至2018年7月2日在Dana-Farber癌症研究所PROFILE研究中接受肿瘤基因分型的肺癌患者2406例。
主要发现
- 测试子集(n=109)中,深度学习模型识别癌症存在、好转/缓解和恶化/进展的AUC均>0.90。
- 机器与人工判读的无病生存期(HR 1.18;95%CI 0.71-1.95)、无进展生存期(HR 1.11;95%CI 0.71-1.71)和至好转/缓解时间(HR 1.03;95%CI 0.65-1.64)相似。
- reserve集中,算法检测的癌症恶化/进展与总生存期下降相关(HR 4.04;95%CI 2.78-5.85),好转/缓解与总生存期改善相关(HR 0.41;95%CI 0.22-0.77)。
临床怎么用
- 可利用深度学习模型从影像报告文本中自动提取癌症存在、好转/缓解和恶化/进展等结局,辅助临床研究终点判读。
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
IMPORTANCE: A rapid learning health care system for oncology will require scalable methods for extracting clinical end points from electronic health records (EHRs). Outside of clinical trials, end points such as cancer progression and response are not routinely encoded into structured data. OBJECTIVE: To determine whether deep natural language processing can extract relevant cancer outcomes from radiologic reports, a ubiquitous but unstructured EHR data source. DESIGN, SETTING, AND PARTICIPANTS: A retrospective cohort study evaluated 1112 patients who underwent tumor genotyping for a diagnosis of lung cancer and participated in the Dana-Farber Cancer Institute PROFILE study from June 26, 2013, to July 2, 2018. EXPOSURES: Patients were divided into curation and reserve sets. Human abstractors applied a structured framework to radiologic reports for the curation set to ascertain the presence of cancer and changes in cancer status over time (ie, worsening/progressing vs improving/responding). Deep learning models were then trained to capture these outcomes from report text and subsequently evaluated in a 10% held-out test subset of curation patients. Cox proportional hazards regression models compared human and machine curations of disease-free survival, progression-free survival, and time to improvement/response in the curation set, and measured associations between report classification and overall survival in the curation and reserve sets. MAIN OUTCOMES AND MEASURES: The primary outcome was area under the receiver operating characteristic curve (AUC) for deep learning models; secondary outcomes were time to improvement/response, disease-free survival, progression-free survival, and overall survival. RESULTS: A total of 2406 patients were included (mean [SD] age, 66.5 [10.8] years; 1428 female [59.7%]; 2170 [90.2%] white). Radiologic reports (n = 14 230) were manually reviewed for 1112 patients in the curation set. In the test subset (n = 109), deep learning models identified the presence of cancer, improvement/response, and worsening/progression with accurate discrimination (AUC >0.90). Machine and human curation yielded similar measurements of disease-free survival (hazard ratio [HR] for machine vs human curation, 1.18; 95% CI, 0.71-1.95); progression-free survival (HR, 1.11; 95% CI, 0.71-1.71); and time to improvement/response (HR, 1.03; 95% CI, 0.65-1.64). Among 15 000 additional reports for 1294 reserve set patients, algorithm-detected cancer worsening/progression was associated with decreased overall survival (HR for mortality, 4.04; 95% CI, 2.78-5.85), and improvement/response was associated with increased overall survival (HR, 0.41; 95% CI, 0.22-0.77). CONCLUSIONS AND RELEVANCE: Deep natural language processing appears to speed curation of relevant cancer outcomes and facilitate rapid learning from EHR data.
引用
- Kehl KL, Elmarakeby H, Nishino M, et al. Assessment of Deep Natural Language Processing in Ascertaining Oncologic Outcomes From Radiology Reports. JAMA Oncol. 2019. PMID: 31343664. doi:10.1001/jamaoncol.2019.1800
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。