GENERATING REVIEWS OF MEDICAL IMAGE REPORTS

摘要

Embodiments are directed towards generating reviews of exams. An exam of interest (EOI) may be anonymized and blinded. N−1 anonymized and blinded foil exams may be selected from an exam database, where N is an integer greater than 1. An anonymized and blinded exam set may be generated for review, where the exam set includes the EOI and the selected foil exams. The exam set may be provided to a reviewer. The reviewer may not be able to discriminate between the EOI and the foil exams. An assessment metric may be generated based on a review of each exam in the exam set. An assessment metric set may be generated, where the assessment metric set includes each generated assessment metric. A report may be generated based on the assessment metric set. The report may indicate a reviewer determination of an overall quality of the EOI. The report may indicate a consistency score for the reviewer, based at least on the reviews of the foil exams. The exam database may be updated based on at least one of the EOI, the assessment metric set, or the consistency score.

主权项

1. A method for reviewing exams with at least one computing device that performs actions, comprising:
modifying a plurality of exams to be anonymous, blinded and accessible as foil exams in a database; modifying an exam of interest (EOI) to be anonymous and blinded; selecting at least one foil exam based at least in part on the EOI; generating an exam set that includes the at least one foil exam and the modified EOI; providing the exam set to a reviewer, wherein based on the reviewer's review of the exam set, the reviewer provides assessment metrics that include a range of graded values for each exam in the exam set which correspond to whether a determination included in the at least one foil exam and another determination included in the modified EOI each have met a community standard which is based on a variability in a comparison of each range of other graded values that are provided by a plurality of different reviewers for each of the foil exams; and generating, by the at least one computing device, a report for the exam set based on at least a quality of the assessment metrics provided by reviewer in comparison to the community standard.