receiving a current report on a condition of a human or animal subject, the current report comprising text composed by a user based on examining one or more scans of the subject, each of the one or more scans comprising at least one image of at least part of the subject;

receiving the one or more scans;

inputting the current report and the one or more scans into a trained machine learning model, the machine learning model being pre-trained based on a plurality of past scans and corresponding reports composed based on the past scans, the past scans comprising at least some scans of other subjects;

based on the current report and the one or more scans of the subject, the machine learning model generating one or more suggestions for updating the text of the report; and

outputting data to cause a user interface to display to the user one or more suggestions for updating the text of the current report, with each respective suggestion visually linked in the user interface to a corresponding subregion within at least one image of at least one of the one or more scans based upon which the respective suggestion was generated.