Abstract
ObjectivesTo investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning.MethodsIn this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++.ResultsThe training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue.ConclusionsIn this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader.Key Points• Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans.• Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts.• Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6–9 mm size.
Highlights
Colorectal cancer is one of the three most frequent cancer-related causes of death among men and women [1]
Of 164 colorectal polyps detected in optical colonoscopy (OC), 57 were excluded due to retrospectively equivocal assignment to the histopathological reference standard and/or retrospectively uncertain localisation in computed tomography (CT) colonography, as described in detail previously [16]
107 colorectal polyps with histopathological reference were evaluated in 63 patients (23 female; mean age: 63 ± 8 years) comprising 169 polyp segmentations in CT colonography images (91 in supine position and 78 in prone position)
Summary
Colorectal cancer is one of the three most frequent cancer-related causes of death among men and women [1]. Its mortality and incidence can be significantly decreased by early detection of precancerous adenomatous polyps which grow over several years [2,3,4,5]. Screening methods such as immunochemical faecal occult blood test and optical colonoscopy (OC) are proven to reduce mortality from colorectal cancer, since clinical symptoms are often non-specific or absent [6, 7]. Computer-aided detection (CAD) algorithms can reduce the number of missed colorectal polyps at CT colonography when used as a second reader [11, 12]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have