Deep learning techniques for detecting preneoplastic and neoplastic lesions in human colorectal histological images

Abstract

Trained pathologists base colorectal cancer identification on the visual interpretation of microscope images. However, image labeling is not always straightforward and this repetitive task is prone to mistakes due to human distraction. Significant efforts are underway to develop informative tools to assist pathologists and decrease the burden and frequency of errors. The present study proposes a deep learning approach to recognize four different stages of cancerous tissue development, including normal mucosa, early preneoplastic lesion, adenoma and cancer. A dataset of human colon tissue images collected and labeled over a 10-year period by a team of pathologists was partitioned into three sets. These were used to train, validate and test the neural network, comprising several convolutional and a few linear layers. The approach used in the present study is ‘direct’; it labels raw images and bypasses the segmentation step. An overall accuracy of >95% was achieved, with the majority of mislabeling referring to a near category. Tests on an external dataset with a different resolution yielded accuracies >80%. The present study demonstrated that the neural network, when properly trained, can provide fast, accurate and reproducible labeling for colon cancer images, with the potential to significantly improve the quality and speed of medical diagnoses.

The Histological Diagnosis of Colonic Adenocarcinoma by Applying Partial Self Supervised Learning

“…Another study performed on the image analysis of colorectal lesions by computer aided diagnostic system showed the sensitivity of 83.9 % and specificity of 82.6% (14). The accuracy of different deep learning models for the classification of colonic lesions varies from 80% to 99.2% (15–18). A study carried out by Chen M et al for the classification of benign and malignant tumors of liver by analysing the H&E images with the application of neural network revealed the accuracy of 96% (19).…”

Potential applications of artificial intelligence in colorectal polyps and cancer: Recent advances and prospects

“…Thus, an automatic image analysis AI system that can help pathologists to identify different types of colorectal polyps accurately is necessary. In recent years, many scholars have begun to probe into this area[ 49 - 51 ]. Korbar et al[ 50 ] proposed an AI system based on a deep neural network model to identify the types of colorectal polyps on whole slide, hematoxylin and eosin-stained images.…”

“…In another study, a deep learning model was proposed to recognize four different stages of cancerous tissue development, including normal mucosa, early preneoplastic lesion, adenoma, and cancer. An overall accuracy of > 95% was achieved[ 51 ].…”

Application of Massive Parallel Deep Learning Algorithm in the Prediction of Colorectal Carcinogenesis of Familial Polyposis

“…Auxiliary tasks can provide an inductive bias to the model and play a regularizing role, thereby reducing the risk of overfitting the model. Multi-task learning is more inclined to learn solutions that can explain multiple tasks at the same time so that the model performs well on multiple tasks, which help the model to improve the generalization ability of different tasks [36].…”