Automated Identification of Myenteric Ganglia in Histopathology Images for the Study of Hirschsprung’s Disease

It appears your Web browser is not configured to display PDF files. Download adobe Acrobat or click here to download the PDF file.

Click here to download the PDF file.

Creator: 

Kurian, Jinu Achamma

Date: 

2021

Abstract: 

Automating digital pathology processing to identify ganglia in seromuscular biopsies from patients with Hirschsprung's disease may be able to provide objective measures and minimize evaluation time for expert pathologists. This adjunctive tool has the potential to improve surgical success for treatment. With thirty patient images, we proposed and evaluated an image processing pipeline that identifies colon tissue structures containing myenteric ganglia. This is an undertaking that is the first of its kind. From whole slide images of calretinin-stained colon sections, we initially segmented the muscularis propria using a convolutional neural network resulting in a mean inclusion of myenteric plexus of 95.96%. Then, colour thresholding identified myenteric plexus regions with a mean inclusion of ganglia of 99.2%. Finally, within these ideal search spaces, we segmented and classified ganglia candidates to achieve an overall mean precision and recall of 64.8% and 80.2%, respectively. Preliminary results encourage further development of these algorithms.

Subject: 

Engineering - Biomedical

Language: 

English

Publisher: 

Carleton University

Contributor: 

Journal Publisher: 
Copyright 2012 Elsevier Inc
Journal Publisher: 
Copyright 1996 Elsevier Inc
Journal Publisher: 
Copyright 2019 Georg Thieme Verlag KG Stuttgart New York

Thesis Degree Name: 

Master of Applied Science: 
M.App.Sc.

Thesis Degree Level: 

Master's

Thesis Degree Discipline: 

Engineering, Biomedical

Parent Collection: 

Theses and Dissertations

Items in CURVE are protected by copyright, with all rights reserved, unless otherwise indicated. They are made available with permission from the author(s).