Systems and methods for improved analysis and generation of medical imaging reports

A system for generating improved medical image reports comprising:

• a server system configured to receive medical images and textual data, wherein the server system comprises:
• an image processing unit configured to:
• train an artificial intelligence module comprising a convolutional neural network based on radiologist marked image data;
• and
• apply the trained artificial intelligence module to the received medical images to generate enhanced medical image data that identifies suspected abnormalities in the received medical images;
• a text processing unit configured to:
• train a natural language processing algorithm on radiology related text, wherein the natural language processing algorithm comprises at least one recurrent neural network and sentiment analysis;
• and
• apply the trained natural language processing to the received textual data to generate structured text data, wherein the structured text data comprises stratified, categorized and annotated statements regarding patient procedures and conditions;
• and
• a platform configured to:
• generate one or more missed findings by determining discrepancies between suspected abnormalities identified in the generated enhanced medical image data and generated structured text data that are not present in the received textual data;
• generate an alert responsive to generating one or more missed findings;
• and
• provide a proposed improved medical image report to a user or creator of the medical image report, wherein the proposed improved medical image report comprises the received medical image report supplemented with a color-coded presentation of the structured text corresponding to the missed finding.

A non-transitory computer-readable medium storing instructions that, when executed on one or more processors, cause the one or more processors to:

• train an artificial intelligence module comprising a convolutional neural network based on radiologist marked image data;
• train a natural language processing module on radiology related text, wherein the natural language processing module comprises at least one recurrent neural network and sentiment analysis;
• receive medical images;
• receive textual data comprising a medical image report corresponding to the received medical images;
• generate enhanced medical image data by applying the trained artificial intelligence module to the received medical images to identify suspected abnormalities in the received medical images;
• generate structured text data by applying the natural language processing module to the received textual data, wherein the structured text data comprises stratified, categorized and annotated statements regarding patient procedures and conditions;
• generate one or more missed findings by determining discrepancies between suspected abnormalities identified in the generated enhanced medical image data and generated structured text data that are not present in the received textual data;
• generate an alert responsive to generating one or more missed findings;
• and
• generate a proposed improved medical image report comprising the received medical image report supplemented with a color-coded presentation of the structured text corresponding to the missed findings.