ASML: Algorithm-Agnostic Architecture for Scalable Machine Learning

Diamantis, Dimitrios E.; Iakovidis, Dimitris K.

doi:10.1109/access.2021.3069857

Cited by 5 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alongside this study, a demonstration website 1 , has been created aiming to present the performance of the TIDE openly, and to become the first publicly available real-time intestine dataset generation platform. It is worth noting that the platform was developed using the Algorithm-agnostic architecture for Scalable Machine Learning (ASML), that we proposed in [25].…”

Section: Introductionmentioning

confidence: 99%

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

Diamantis,

Gatoula,

Koulaouzidis

et al. 2024

IEEE Access

Self Cite

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

Section: Introductionmentioning

confidence: 99%

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

Diamantis,

Gatoula,

Koulaouzidis

et al. 2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

EndoVAE: Generating Endoscopic Images with a Variational Autoencoder

Diamantis

Gatoula

Iakovidis

2022

2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)

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. Aiming to a viable solution for WCE image synthesis, a novel Variational Autoencoder architecture is proposed, namely 'This Intestine Does not Exist' (TIDE). The proposed architecture comprises multiscale feature extraction convolutional blocks and residual connections, which enable 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. Furthermore, qualitative and user evaluation studies by experienced WCE specialists, validate from a medical viewpoint that both the normal and abnormal WCE images synthesized by TIDE are sufficiently realistic.

show abstract

“…The strategy aims to minimize the space utilization at the cost of time and maximize the space utilization, providing guidance for the rational arrangement of existing resources. Other scholars have studied scalable machine learning algorithms and related applications [7][8]. Therefore, parallel and distributed optimization algorithms for scalable machine learning are studied in this paper.…”

Section: Introductionmentioning

confidence: 99%

Parallel and Distributed Optimization Algorithms for Scalable Machine Learning

2022

View full text Add to dashboard Cite

In recent years, the emergence of a large number of parallel data makes us pay more and more attention to how to improve the efficiency of decision-making. In this paper, scalable machine learning is the research object. First, the existing problems and development directions of existing related technologies are analyzed. Then, an extensible machine learning algorithm is introduced to solve these problems. Finally, MATLAB is used to realize the optimization simulation of n subsystems, and the regression equation between the final result and the optimal value as well as the mean square error is obtained. The corresponding conclusions are drawn. The parallel and distributed optimization algorithms of extensible machine learning run for a short time, The parallel capacity is about 3500k, and the global performance can be effectively improved by updating existing methods in the feasible region.

show abstract

ASML: Algorithm-Agnostic Architecture for Scalable Machine Learning

Cited by 5 publications

References 39 publications

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

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

EndoVAE: Generating Endoscopic Images with a Variational Autoencoder

Parallel and Distributed Optimization Algorithms for Scalable Machine Learning

Contact Info

Product

Resources

About