Methods and apparatus for artificial intelligence informed radiological reporting and model refinement

A non-transitory computer readable medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

• receiving speech data from a dictation system;
• determining a clinical characterization based on the received speech data;
• receiving first image scan data for a first image;
• applying a feature extraction process to the received first image scan data to identify at least one feature of the first image;
• applying a trained machine learning model to the at least one feature of the first image, wherein the trained machine learning model is trained on historical anatomical associations generated for prior images;
• determining a first classification of the at least one feature in the first image based on the application of the trained machine learning model to the at least one feature of the first image;
• generating at least one anatomical association for the first image based on the first classification and the clinical characterization;
• and
• providing the at least one anatomical association for display.

A system comprising:

• a database;
• and
• at least one processor communicatively coupled to the database and configured to:
• receive speech data from a dictation system;
• determine a clinical characterization based on the received speech data;
• obtain first image scan data for a first image from the database;
• apply a feature extraction process to the received first image scan data to identify at least one feature of the first image;
• apply a trained machine learning model to the at least one feature of the first image, wherein the trained machine learning model is trained on historical anatomical associations generated for prior images;
• determine a first classification of the at least one feature in the first image based on the application of the trained machine learning model to the at least one feature of the first image;
• generate at least one anatomical association for the first image based on the first classification and the clinical characterization;
• and
• provide the at least one anatomical association for display.