Expert Systems Help Identify Benign and Precancerous Colorectal Polyps on CT
During the past two decades, CT colonography emerged as a noninvasive screening method for colorectal cancer. A recent study suggests that the machine-learning-based image analysis enabled noninvasive detection and differentiation of benign and premalignant colorectal polyps with CT colonography. The study findings were published in the journal RADIOLOGY on February 23, 2021.
CT colonography is effective in visualizing portions of the colon not evaluated by optical colonoscopy (OC) in cases of complex anatomic conditions causing failed or incomplete OC and therefore permits robust polyp detection also in the right colon. But, CT colonography does not enable a definite differentiation between benign and premalignant polyps, crucial for individual risk stratification and therapy guidance. Therefore, researchers of the University Hospital, LMU Munich, Germany conducted a study to perform machine learning-based differentiation of benign and premalignant colorectal polyps detected with CT colonography in an average-risk asymptomatic colorectal cancer screening sample with external validation using radiomics.
It was a secondary analysis of a prospective trial. Researchers identified colorectal polyps of all size categories and morphologies and manually segmented them on CT colonographic images. They further classified them as benign (hyperplastic polyp or regular mucosa) or premalignant (adenoma) according to the histopathologic reference standard. They extracted Quantitative image features characterizing shape (n = 14), gray level histogram statistics (n = 18), and image texture (n = 68) from segmentations after applying 22 image filters, resulting in 1906 feature-filter combinations. They used a random forest classification algorithm to predict the individual polyp character.
Key findings of the study were:
• The machine-learning algorithm was trained on a set of more than 100 colorectal polyps in 63 patients and then tested on a set of 77polyps in 59 patients comprising 118 segmentations.
• In the test set, the researchers found that the machine learning approach enabled noninvasive differentiation of benign and premalignant CT colonography-detected colorectal polyps, with a sensitivity of 82%, and specificity of 85%.
• They also noted that the algorithm produced an AUC of 0.87 in polyps sized 6-9 mm and 0.90 in polyps sized 10 mm or larger.
• They further mentioned that with a relative importance of 3.7%, first-order gray-level histogram statistics was the most important radiomics feature in the model's decision-making.
The authors concluded, "In this proof-of-concept study with validation in an external multicenter test set, machine learning–assisted CT colonography analysis enabled the differentiation of benign and premalignant colorectal polyps. The present study provides a potential basis for future prospective studies with a larger number of patients to further examine the diagnostic performance of machine learning algorithms for the noninvasive analysis of CT colonography–detected polyps."
For further information:
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.