Abstract
Colorectal cancer is one of the most common cancers worldwide, accounting for an annual estimated 1.8 million incident cases. With the increasing number of colonoscopies being performed, colorectal biopsies make up a large proportion of any histopathology laboratory workload. We trained and validated a unique artificial intelligence (AI) deep learning model as an assistive tool to screen for colonic malignancies in colorectal specimens, in order to improve cancer detection and classification; enabling busy pathologists to focus on higher order decision-making tasks. The study cohort consists of Whole Slide Images (WSI) obtained from 294 colorectal specimens. Qritive’s unique composite algorithm comprises both a deep learning model based on a Faster Region Based Convolutional Neural Network (Faster-RCNN) architecture for instance segmentation with a ResNet-101 feature extraction backbone that provides glandular segmentation, and a classical machine learning classifier. The initial training used pathologists’ annotations on a cohort of 66,191 image tiles extracted from 39 WSIs. A subsequent application of a classical machine learning-based slide classifier sorted the WSIs into ‘low risk’ (benign, inflammation) and ‘high risk’ (dysplasia, malignancy) categories. We further trained the composite AI-model’s performance on a larger cohort of 105 resections WSIs and then validated our findings on a cohort of 150 biopsies WSIs against the classifications of two independently blinded pathologists. We evaluated the area under the receiver-operator characteristic curve (AUC) and other performance metrics. The AI model achieved an AUC of 0.917 in the validation cohort, with excellent sensitivity (97.4%) in detection of high risk features of dysplasia and malignancy. We demonstrate an unique composite AI-model incorporating both a glandular segmentation deep learning model and a classical machine learning classifier, with excellent sensitivity in picking up high risk colorectal features. As such, AI plays a role as a potential screening tool in assisting busy pathologists by outlining the dysplastic and malignant glands.
Highlights
The biopsy Whole Slide Images (WSI) were classified into high risk and low risk, similar to the resection slides used for training the slide classifier
After obtaining the output labels produced by the artificial intelligence (AI) algorithm from the validation set, we compared them to the ground truth labels produced by our expert pathologists
We demonstrated that our unique composite AI model incorporating a glandular segmentation deep learning model and a machine learning classifier has promising ability in picking up high risk colorectal features
Summary
Segmentation of colon glands Nuclei segmentation Deep Neural Network Visualization to Interpret WSI Analysis Outcomes for Colorectal Polyps Various studies Various studies Various studies Various studies Glandular segmentation deep learning model to detect high risk colorectal polypsDataset GLAS challenge (165 images) 537 images from Case Western Reserve UniversityF1 score 0.893–0.843 0.858–0.771176 WSIs from Dartmouth-Hitchcock Medical CenterVarious studies Various studies Various studies Various studies WSIs produced from 294 colorectal specimens from Singapore General Hospitaltasks[34]. Further studies on a larger dataset are underway. While the performance data from the validation had a high AUC of 91.7, it still contained 2 false negatives but 29 false positives. In a clinical institution with AI model being applied to patient care in the context of cancer diagnosis, the system was designed to favour sensitivity over specificity to ensure usability in assistive workflow. This allows the AI model to categorize all high grade dysplasia and adenocarcinoma as ‘high risk’. This carry-over effect, when applied onto the validation set, resulted in higher false positives as the prediction certainty of the various features were forced into a binary classification, leading to images with relatively low prediction certainty of 70% being highlighted as ‘high risk’
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
