Artificial intelligence-based endoscopic diagnosis of colorectal polyps using residual networks

NICE polyp feature classification for colonoscopy screening

PURPOSE

Colorectal cancer is one of the most prevalent cancers worldwide, highlighting the critical need for early and accurate diagnosis to reduce patient risks. Inaccurate diagnoses not only compromise patient outcomes but also lead to increased costs and additional time burdens for clinicians. Enhancing diagnostic accuracy is essential, and this study focuses on improving the accuracy of polyp classification using the NICE classification, which evaluates three key features: colour, vessels, and surface pattern.

METHODS

A multiclass classifier was developed and trained to independently classify each of the three features in the NICE classification. The approach prioritizes clinically relevant features rather than relying on handcrafted or obscure deep learning features, ensuring transparency and reliability for clinical use. The classifier was trained on internal datasets and tested on both internal and public datasets.

RESULTS

The classifier successfully classified the three polyp features, achieving an accuracy of over 92% on internal datasets and exceeding 88% on a public dataset. The high classification accuracy demonstrates the system's effectiveness in identifying the key features from the NICE classification.

CONCLUSION

This study underscores the potential of using an independent classification approach for NICE features to enhance clinical decision-making in colorectal cancer diagnosis. The method shows promise in improving diagnostic accuracy, which could lead to better patient outcomes and more efficient clinical workflows.

Advancing equitable and personalized cancer care: Novel applications and priorities of artificial intelligence for fairness and inclusivity in the patient care workflow

Patient care workflows are highly multimodal and intertwined: the intersection of data outputs provided from different disciplines and in different formats remains one of the main challenges of modern oncology. Artificial Intelligence (AI) has the potential to revolutionize the current clinical practice of oncology owing to advancements in digitalization, database expansion, computational technologies, and algorithmic innovations that facilitate discernment of complex relationships in multimodal data. Within oncology, radiation therapy (RT) represents an increasingly complex working procedure, involving many labor-intensive and operator-dependent tasks. In this context, AI has gained momentum as a powerful tool to standardize treatment performance and reduce inter-observer variability in a time-efficient manner. This review explores the hurdles associated with the development, implementation, and maintenance of AI platforms and highlights current measures in place to address them. In examining AI's role in oncology workflows, we underscore that a thorough and critical consideration of these challenges is the only way to ensure equitable and unbiased care delivery, ultimately serving patients' survival and quality of life.

High-Speed and Accurate Diagnosis of Gastrointestinal Disease: Learning on Endoscopy Images Using Lightweight Transformer with Local Feature Attention

In response to the pressing need for robust disease diagnosis from gastrointestinal tract (GIT) endoscopic images, we proposed FLATer, a fast, lightweight, and highly accurate transformer-based model. FLATer consists of a residual block, a vision transformer module, and a spatial attention block, which concurrently focuses on local features and global attention. It can leverage the capabilities of both convolutional neural networks (CNNs) and vision transformers (ViT). We decomposed the classification of endoscopic images into two subtasks: a binary classification to discern between normal and pathological images and a further multi-class classification to categorize images into specific diseases, namely ulcerative colitis, polyps, and esophagitis. FLATer has exhibited exceptional prowess in these tasks, achieving 96.4% accuracy in binary classification and 99.7% accuracy in ternary classification, surpassing most existing models. Notably, FLATer could maintain impressive performance when trained from scratch, underscoring its robustness. In addition to the high precision, FLATer boasted remarkable efficiency, reaching a notable throughput of 16.4k images per second, which positions FLATer as a compelling candidate for rapid disease identification in clinical practice.

Self-supervised representation learning using feature pyramid siamese networks for colorectal polyp detection

Colorectal cancer is a leading cause of cancer-related deaths globally. In recent years, the use of convolutional neural networks in computer-aided diagnosis (CAD) has facilitated simpler detection of early lesions like polyps during real-time colonoscopy. However, the majority of existing techniques require a large training dataset annotated by experienced experts. To alleviate the laborious task of image annotation and utilize the vast amounts of readily available unlabeled colonoscopy data to further improve the polyp detection ability, this study proposed a novel self-supervised representation learning method called feature pyramid siamese networks (FPSiam). First, a feature pyramid encoder module was proposed to effectively extract and fuse both local and global feature representations among colonoscopic images, which is important for dense prediction tasks like polyp detection. Next, a self-supervised visual feature representation containing the general feature of colonoscopic images is learned by the siamese networks. Finally, the feature representation will be transferred to the downstream colorectal polyp detection task. A total of 103 videos (861,400 frames), 100 videos (24,789 frames), and 60 videos (15,397 frames) in the LDPolypVideo dataset are used to pre-train, train, and test the performance of the proposed FPSiam and its counterparts, respectively. The experimental results have illustrated that our FPSiam approach obtains the optimal capability, which is better than that of other state-of-the-art self-supervised learning methods and is also higher than the method based on transfer learning by 2.3 mAP and 3.6 mAP for two typical detectors. In conclusion, FPSiam provides a cost-efficient solution for developing colorectal polyp detection systems, especially in conditions where only a small fraction of the dataset is labeled while the majority remains unlabeled. Besides, it also brings fresh perspectives into other endoscopic image analysis tasks.

Transformer-based multi-task learning for classification and segmentation of gastrointestinal tract endoscopic images

Despite being widely utilized to help endoscopists identify gastrointestinal (GI) tract diseases using classification and segmentation, models based on convolutional neural network (CNN) have difficulties in distinguishing the similarities among some ambiguous types of lesions presented in endoscopic images, and in the training when lacking labeled datasets. Those will prevent CNN from further improving the accuracy of diagnosis. To address these challenges, we first proposed a Multi-task Network (TransMT-Net) capable of simultaneously learning two tasks (classification and segmentation), which has the transformer designed to learn global features and can combine the advantages of CNN in learning local features so that to achieve a more accurate prediction in identifying the lesion types and regions in GI tract endoscopic images. We further adopted the active learning in TransMT-Net to tackle the labeled image-hungry problem. A dataset was created from the CVC-ClinicDB dataset, Macau Kiang Wu Hospital, and Zhongshan Hospital to evaluate the model performance. Then, the experimental results show that our model not only achieved 96.94% accuracy in the classification task and 77.76% Dice Similarity Coefficient in the segmentation task but also outperformed those of other models on our test set. Meanwhile, active learning also produced positive results for the performance of our model with a small-scale initial training set, and even its performance with 30% of the initial training set was comparable to that of most comparable models with the full training set. Consequently, the proposed TransMT-Net has demonstrated its potential performance in GI tract endoscopic images and it through active learning can alleviate the shortage of labeled images.

Preparation of image databases for artificial intelligence algorithm development in gastrointestinal endoscopy

Over the past decade, technological advances in deep learning have led to the introduction of artificial intelligence (AI) in medical imaging. The most commonly used structure in image recognition is the convolutional neural network, which mimics the action of the human visual cortex. The applications of AI in gastrointestinal endoscopy are diverse. Computer-aided diagnosis has achieved remarkable outcomes with recent improvements in machine-learning techniques and advances in computer performance. Despite some hurdles, the implementation of AI-assisted clinical practice is expected to aid endoscopists in real-time decision-making. In this summary, we reviewed state-of-the-art AI in the field of gastrointestinal endoscopy and offered a practical guide for building a learning image dataset for algorithm development.