Function: The primary purpose of today research study was to test the expediency of predicting diagnostic mistakes in mammography by combining radiologists' stare behavior and image characteristics. Eye-tracking information and diagnostic decisions for 40 cases were obtained from 4 Radiology residents and 2 breast imaging experts as part of an IRB-approved pilot research study. Gaze habits features were drawn out from the eye-tracking data. Finally, artificial intelligence formulas were used to merge look and picture features for forecasting human mistake. Outcomes: Diagnostic error can be forecasted reliably by combining stare behavior characteristics from the radiologist and textural characteristics from the image under review. Leveraging data accumulated from numerous viewers created a practical group model. Personalized user modeling was more precise for the more experienced visitors than for the less experienced ones. The ideal doing group-based and personalized anticipating models included combinations of both gaze and photo features. Conclusions: Diagnostic mistakes in mammography can be forecasted reliably by leveraging the radiologists' stare behavior and photo material.
* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions
** If you believe that content on the Plex is summarised improperly, please, contact us, and we will get rid of it quickly; please, send an email with a brief explanation.