Short Training Significantly Improves Ganglion Cell Detection Using an Algorithm-Assisted Approach

• DOI: 10.5858/arpa.2021-0481-oa
• Publication Date: 2022-06-23T21:36:26Z
• Authors: Ariel Greenberg; Benzion Samueli; Ibrahim Fahoum; Shai Farkash; Orli Greenberg; Valentina Zemser-...; Edmond Sabo; Rami R. Hagege; Dov Hershkovitz
• Keywords: Presentation (obstetrics)

Medical education in pathology relies on the accumulation of experience gained through inspection numerous samples from each entity. Acquiring sufficient teaching material for rare diseases, such as Hirschsprung disease (HSCR), may be difficult, especially smaller institutes. The current study makes use a previously developed decision support system using algorithm meant to aid pathologists diagnosis HSCR.To assess effect short training session algorithm-assisted HSCR diagnosis.Five reviewed data set 568 image sets (1704 images total) selected 50 cases by and were tasked with scoring presence or absence ganglion cells. task was repeated total 3 times. Each pathologist had complete educational presentation between second third iterations.The resulted significantly increased rate correct diagnoses (true positive/negative) decreased need referrals expert consultation. No statistically significant changes false positives/negatives detected.A very (<10 minutes) can greatly improve pathologist's performance HSCR. same approach feasible other diseases.