■自然语言处理(NLP)可以从成像报告生成诊断代码。同时,国际疾病分类(ICD-10)代码是美国的计费/编码标准,能够追踪疾病负担和结果。这项横断面研究旨在测试NLP算法性能的可行性,并与放射科医师和医师手动编码进行比较。
■三名神经放射科医师和一名非放射科医师评审员手动编码随机选择的200份颅脑脊髓CT和MRI报告库,其中>10,000份。NLP算法(Radnosis,VEEV,Inc.,明尼阿波利斯,MN)将每个报告的印象细分为“短语”,每个短语有多个ICD-10匹配项。只看印象,医师评审人员为每个短语选择了最佳ICD-10编码.比较医生和算法选择的代码是否一致。
■该算法将报告\'印象提取为645个短语,每个人都有排名的ICD-10比赛。关于审阅者选择的代码,成对协议不可靠(Krippendorffα=0.39-0.63)。使用一致的审阅者协议作为“基本事实”,算法对前5个代码的敏感性/特异性/F2为0.88/0.80/0.83,对单个最佳代码的敏感性/特异性/F2为0.67/0.82/0.67.引擎将“相关否定”列表为所述发现的否定代码(例如“无颅内出血”)。引擎的匹配比全长ICD-10代码更具体(p=0.00582x10-3)。
■医师评审员的手动编码具有显着的可变性,并且耗时,而NLP算法的前5个诊断代码相对准确。这项初步工作证明了生成具有可靠性和一致性的代码的可行性和潜力。未来的工作可能包括将诊断代码与临床遭遇代码相关联,以评估成像对,和护理的相关性。
UNASSIGNED: Natural language processing (NLP) can generate diagnoses codes from imaging
reports. Meanwhile, the International Classification of Diseases (ICD-10) codes are the United States\' standard for billing/coding, which enable tracking disease burden and outcomes. This cross-sectional study aimed to test feasibility of an NLP algorithm\'s performance and comparison to radiologists\' and physicians\' manual coding.
UNASSIGNED: Three neuroradiologists and one non-radiologist physician reviewers manually coded a randomly-selected pool of 200 craniospinal CT and MRI
reports from a pool of >10,000. The NLP algorithm (Radnosis, VEEV, Inc., Minneapolis, MN) subdivided each report\'s Impression into \"phrases\", with multiple ICD-10 matches for each phrase. Only viewing the Impression, the physician reviewers selected the single best ICD-10 code for each phrase. Codes selected by the physicians and algorithm were compared for agreement.
UNASSIGNED: The algorithm extracted the
reports\' Impressions into 645 phrases, each having ranked ICD-10 matches. Regarding the reviewers\' selected codes, pairwise agreement was unreliable (Krippendorff α = 0.39-0.63). Using unanimous reviewer agreement as \"ground truth\", the algorithm\'s sensitivity/specificity/F2 for top 5 codes was 0.88/0.80/0.83, and for the single best code was 0.67/0.82/0.67. The engine tabulated \"pertinent negatives\" as negative codes for stated findings (e.g. \"no intracranial hemorrhage\"). The engine\'s matching was more specific for shorter than full-length ICD-10 codes (p = 0.00582x10-3).
UNASSIGNED: Manual coding by physician reviewers has significant variability and is time-consuming, while the NLP algorithm\'s top 5 diagnosis codes are relatively accurate. This preliminary work demonstrates the feasibility and potential for generating codes with reliability and consistency. Future works may include correlating diagnosis codes with clinical encounter codes to evaluate imaging\'s impact on, and relevance to care.