Systems and methods for identifying errors and/or critical results in medical reports

Abstract

Systems and methods for analyzing a medical report to determine whether the medical report includes at least one instance of at least one category selected from a group consisting of: gender error, laterality error, and critical finding. In some embodiments, one or more portions of text are identified from the medical report. Contextual information associated with the medical report is used to determine whether the identified one or more portions of text comprise at least one instance of at least one category selected from the group.

What is claimed is: 
     
         1 .- 20 . (canceled) 
     
     
         21 . A system comprising:
 at least one processor; and   at least one storage medium storing executable instructions that, when executed by the at least one processor, cause the at least one processor to carry out a method comprising:
 analyzing a medical report to determine whether the medical report includes at least one laterality error, wherein a laterality error indicates a determination that an entity is associated with an incorrect side of a body, wherein analyzing the medical report comprises acts of:
 identifying, in text of the medical report, a first mention of a first entity; 
 identifying, in the text of the medical report, a first side of a body, the first side of the body coupled with the first mention of the first entity; 
 analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body; and 
 based on the analyzing, determining whether the medical report includes the at least one laterality error; and 
 
 in response to determining that the medical report includes the at least one laterality error, triggering a notification for a medical professional, wherein the notification indicates that the medical report includes the at least one laterality error. 
   
     
     
         22 . The system of  claim 21 , wherein:
 identifying the first mention of the first entity in the text of the medical report comprises identifying, in the text, a mention of a first anatomical feature;   the first anatomical feature is an anatomical feature that is associated with an accurate side of the body;   analyzing whether there is a mismatch between the first entity and the first side of the body comprises determining whether there is a mismatch between the accurate side of the body for the first anatomical feature and the first side of the body that is coupled with the first entity in the first mention in the text.   
     
     
         23 . The system of  claim 22 , wherein the analyzing whether there is a mismatch using the text of the medical report and/or the contextual information comprises identifying, from the text and/or the contextual information, the accurate side of the body. 
     
     
         24 . The system of  claim 23 , wherein identifying the accurate side of the body using the text of the medical report and/or the contextual information comprises:
 identifying a second mention of the first entity in the text of the medical report and/or in the contextual information, wherein the second mention of the first entity is coupled in the text and/or in the contextual information with a side of the body; and   identifying that the side of the body coupled with the second mention is the accurate side of the body.   
     
     
         25 . The system of  claim 21 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one statistical model, trained to recognize in medical reports text indicative of a laterality error, to the text of the medical report and/or to contextual information associated with the medical report. 
     
     
         26 . The system of  claim 25 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body further comprises:
 tokenizing the text of the medical report;   determining, with the at least one statistical model, a concept related to one or more words and/or phrases of the text of the medical report;   identifying, with the at least one statistical model, a candidate laterality error from the text of the medical report, wherein the candidate laterality error comprises the first side of the body coupled with the first mention of the first entity;   determining, with the at least one statistical model, a confidence value representing a likelihood that the candidate laterality error is a laterality error; and   evaluating the confidence value determined for the candidate laterality error using the at least one statistical model to determine whether the candidate laterality error is a laterality error.   
     
     
         27 . The system of  claim 21 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one rule to the text of the medical report and/or to contextual information associated with the medical report. 
     
     
         28 . The system of  claim 21 , wherein analyzing the medical report to determine whether the medical report includes the at least one laterality error further comprises:
 identifying, in the medical report, a second mention of the first entity;   determining whether the first mention of the first entity and the second mention of the first entity mention refer to a same entity; and   in response to determining that the first mention of the first entity and the second mention of the first entity mention refer to the same entity, linking the first mention of the first entity with the second mention of the first entity;   wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises analyzing the second mention of the first entity linked with the first mention of the first entity.   
     
     
         29 . A method comprising:
 analyzing a medical report to determine whether the medical report includes at least one laterality error, wherein a laterality error indicates a determination that an entity is associated with an incorrect side of a body, wherein analyzing the medical report comprises acts of:
 identifying, in text of the medical report, a first mention of a first entity; 
 identifying, in the text of the medical report, a first side of a body, the first side of the body coupled with the first mention of the first entity; 
 analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body; and 
 based on the analyzing, determining whether the medical report includes the at least one laterality error; and 
   in response to determining that the medical report includes the at least one laterality error, triggering a notification for a medical professional, wherein the notification indicates that the medical report includes the at least one laterality error.   
     
     
         30 . The method of  claim 29 , wherein:
 identifying the first mention of the first entity in the text of the medical report comprises identifying, in the text, a mention of a first anatomical feature;   the first anatomical feature is an anatomical feature that is associated with an accurate side of the body;   analyzing whether there is a mismatch between the first entity and the first side of the body comprises determining whether there is a mismatch between the accurate side of the body for the first anatomical feature and the first side of the body that is coupled with the first entity in the first mention in the text.   
     
     
         31 . The method of  claim 30 , wherein the analyzing whether there is a mismatch using the text of the medical report and/or the contextual information comprises identifying, from the text and/or the contextual information, the accurate side of the body. 
     
     
         32 . The system of  claim 31 , wherein identifying the accurate side of the body using the text of the medical report and/or the contextual information comprises:
 identifying a second mention of the first entity in the text of the medical report and/or in the contextual information, wherein the second mention of the first entity is coupled in the text and/or in the contextual information with a side of the body; and   identifying that the side of the body coupled with the second mention is the accurate side of the body.   
     
     
         33 . The method of  claim 29 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one statistical model, trained to recognize in medical reports text indicative of a laterality error, to the text of the medical report and/or to contextual information associated with the medical report. 
     
     
         34 . The method of  claim 33 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body further comprises:
 tokenizing the text of the medical report;   determining, with the at least one statistical model, a concept related to one or more words and/or phrases of the text of the medical report;   identifying, with the at least one statistical model, a candidate laterality error from the text of the medical report, wherein the candidate laterality error comprises the first side of the body coupled with the first mention of the first entity;   determining, with the at least one statistical model, a confidence value representing a likelihood that the candidate laterality error is a laterality error; and   evaluating the confidence value determined for the candidate laterality error using the at least one statistical model to determine whether the candidate laterality error is a laterality error.   
     
     
         35 . The method of  claim 29 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one rule to the text of the medical report and/or to contextual information associated with the medical report. 
     
     
         36 . At least one non-transitory computer-readable storage medium having stored thereon instructions that, when executed by at least one processor, perform a method comprising:
 analyzing a medical report to determine whether the medical report includes at least one laterality error, wherein a laterality error indicates a determination that an entity is associated with an incorrect side of a body, wherein analyzing the medical report comprises acts of:
 identifying, in text of the medical report, a first mention of a first entity; 
 identifying, in the text of the medical report, a first side of a body, the first side of the body coupled with the first mention of the first entity; 
 analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body; and 
 based on the analyzing, determining whether the medical report includes the at least one laterality error; and 
   in response to determining that the medical report includes the at least one laterality error, triggering a notification for a medical professional, wherein the notification indicates that the medical report includes the at least one laterality error.   
     
     
         37 . The at least one non-transitory computer-readable storage medium of  claim 36 , wherein:
 identifying the first mention of the first entity in the text of the medical report comprises identifying, in the text, a mention of a first anatomical feature;   the first anatomical feature is an anatomical feature that is associated with an accurate side of the body;   analyzing whether there is a mismatch between the first entity and the first side of the body comprises determining whether there is a mismatch between the accurate side of the body for the first anatomical feature and the first side of the body that is coupled with the first entity in the first mention in the text.   
     
     
         38 . The at least one non-transitory computer-readable storage medium of  claim 36 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one statistical model, trained to recognize in medical reports text indicative of a laterality error, to the text of the medical report and/or to contextual information associated with the medical report. 
     
     
         39 . The at least one non-transitory computer-readable storage medium of  claim 38 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body further comprises:
 tokenizing the text of the medical report;   determining, with the at least one statistical model, a concept related to one or more words and/or phrases of the text of the medical report;   identifying, with the at least one statistical model, a candidate laterality error from the text of the medical report, wherein the candidate laterality error comprises the first side of the body coupled with the first mention of the first entity;   determining, with the at least one statistical model, a confidence value representing a likelihood that the candidate laterality error is a laterality error; and   evaluating the confidence value determined for the candidate laterality error using the at least one statistical model to determine whether the candidate laterality error is a laterality error.   
     
     
         40 . The at least one non-transitory computer-readable storage medium of  claim 36 , wherein analyzing, using the text of the medical report and/or contextual information associated with the medical report, whether there is a mismatch between the first entity and the first side of the body comprises applying at least one rule to the text of the medical report and/or to contextual information associated with the medical report.