Evaluating clinical diversity and plausibility of synthetic capsule endoscopic images

Vats, Anuja; Pedersen, Marius; Mohammed, Ahmed; Hovde, Øistein

doi:10.1038/s41598-023-36883-x

Cited by 4 publications

(1 citation statement)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the same time, more sophisticated approaches propose the use of generative adversarial networks in developing synthetic samples for a particular class. [ 55 56 57 ] These approaches are promising pursuits that can be explored further to improve the discrimination capability of the developed models while building on our initial findings.…”

Section: Discussionmentioning

confidence: 97%

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution

Jiang,

Dorosan,

Leong

et al. 2024

Singapore Medical Journal

View full text Add to dashboard Cite

Introduction: Deep learning models can assess the quality of images and discriminate among abnormalities in small bowel capsule endoscopy (CE), reducing fatigue and the time needed for diagnosis. They serve as a decision support system, partially automating the diagnosis process by providing probability predictions for abnormalities. Methods: We demonstrated the use of deep learning models in CE image analysis, specifically by piloting a bowel preparation model (BPM) and an abnormality detection model (ADM) to determine frame-level view quality and the presence of abnormal findings, respectively. We used convolutional neural network-based models pretrained on large-scale open-domain data to extract spatial features of CE images that were then used in a dense feed-forward neural network classifier. We then combined the open-source Kvasir-Capsule dataset (n = 43) and locally collected CE data (n = 29). Results: Model performance was compared using averaged five-fold and two-fold cross-validation for BPMs and ADMs, respectively. The best BPM model based on a pre-trained ResNet50 architecture had an area under the receiver operating characteristic and precision-recall curves of 0.969±0.008 and 0.843±0.041, respectively. The best ADM model, also based on ResNet50, had top-1 and top-2 accuracies of 84.03±0.051 and 94.78±0.028, respectively. The models could process approximately 200–250 images per second and showed good discrimination on time-critical abnormalities such as bleeding. Conclusion: Our pilot models showed the potential to improve time to diagnosis in CE workflows. To our knowledge, our approach is unique to the Singapore context. The value of our work can be further evaluated in a pragmatic manner that is sensitive to existing clinician workflow and resource constraints.

show abstract

Section: Discussionmentioning

confidence: 97%

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution

Jiang,

Dorosan,

Leong

et al. 2024

Singapore Medical Journal

View full text Add to dashboard Cite

show abstract

This Intestine Does Not Exist: Multiscale Residual Variational Autoencoder for Realistic Wireless Capsule Endoscopy Image Generation

Diamantis,

Gatoula,

Koulaouzidis

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Medical image synthesis has emerged as a promising solution to address the limited availability of annotated medical data needed for training machine learning algorithms in the context of image-based Clinical Decision Support (CDS) systems. To this end, Generative Adversarial Networks (GANs) have been mainly applied to support the algorithm training process by generating synthetic images for data augmentation. However, in the field of Wireless Capsule Endoscopy (WCE), the limited content diversity and size of existing publicly available annotated datasets adversely affect both the training stability and synthesis performance of GANs. In this paper a novel Variational Autoencoder (VAE) architecture is proposed for WCE image synthesis, namely 'This Intestine Does not Exist' (TIDE). This is the first VAE architecture comprising multiscale feature extraction convolutional blocks and residual connections. Its advantage is that it enables the generation of high-quality and diverse datasets even with a limited number of training images. Contrary to the current approaches, which are oriented towards the augmentation of the available datasets, this study demonstrates that using TIDE, real WCE datasets can be fully substituted by artificially generated ones, without compromising classification performance of CDS. It performs a spherical experimental evaluation study that covers both quantitative and qualitative aspects, including a user evaluation study performed by WCE specialists, which validate from a medical viewpoint that both the normal and abnormal WCE images synthesized by TIDE are sufficiently realistic. The quantitative results obtained by comparative experiments validate that the proposed architecture outperforms the state-of-the-art.

show abstract

Evaluating the plausibility of synthetic images for improving automated endoscopic stone recognition

Gonzalez-Perez,

Lopez-Tiro,

Reyes-Amezcua

et al. 2024

2024 IEEE 37th International Symposium on Computer-Based Medical Systems (CBMS)

View full text Add to dashboard Cite

Evaluating clinical diversity and plausibility of synthetic capsule endoscopic images

Cited by 4 publications

References 36 publications

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution

This Intestine Does Not Exist: Multiscale Residual Variational Autoencoder for Realistic Wireless Capsule Endoscopy Image Generation

Evaluating the plausibility of synthetic images for improving automated endoscopic stone recognition

Contact Info

Product

Resources

About