Actual Impact of GAN Augmentation on CNN Classification Performance

Pinetz, Thomas; Ruisz, Johannes; Soukup, Daniel

doi:10.5220/0007244600150023

Cited by 5 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the introduction of a conversion of the 3D head geometry into a 2D image, image-based convolutional neural network (CNN)–based classification ( 17 ) can be applied on low-resolution images. Generative adversarial networks (GANs) ( 21 ) have been suggested as a data augmentation tool ( 15 ) and have been able to increase classification performance for small datasets ( 22 ).…”

Section: Introductionmentioning

confidence: 99%

Impact of data synthesis strategies for the classification of craniosynostosis

Schaufelberger,

Kühle,

Wachter

et al. 2023

Front. Med. Technol.

View full text Add to dashboard Cite

IntroductionPhotogrammetric surface scans provide a radiation-free option to assess and classify craniosynostosis. Due to the low prevalence of craniosynostosis and high patient restrictions, clinical data are rare. Synthetic data could support or even replace clinical data for the classification of craniosynostosis, but this has never been studied systematically.MethodsWe tested the combinations of three different synthetic data sources: a statistical shape model (SSM), a generative adversarial network (GAN), and image-based principal component analysis for a convolutional neural network (CNN)–based classification of craniosynostosis. The CNN is trained only on synthetic data but is validated and tested on clinical data.ResultsThe combination of an SSM and a GAN achieved an accuracy of 0.960 and an F1 score of 0.928 on the unseen test set. The difference to training on clinical data was smaller than 0.01. Including a second image modality improved classification performance for all data sources.ConclusionsWithout a single clinical training sample, a CNN was able to classify head deformities with similar accuracy as if it was trained on clinical data. Using multiple data sources was key for a good classification based on synthetic data alone. Synthetic data might play an important future role in the assessment of craniosynostosis.

show abstract

Section: Introductionmentioning

confidence: 99%

Impact of data synthesis strategies for the classification of craniosynostosis

Schaufelberger,

Kühle,

Wachter

et al. 2023

Front. Med. Technol.

View full text Add to dashboard Cite

show abstract

“…The GAN algorithm of deep learning (DL) techniques has been used successfully in many applications, such as style transfer, image synthesising, and the famous DeepFake synthetic media creator. The power of the GAN algorithm comes from learning directly from data without human knowledge [7]. That means that GAN does not require a human to select features to predict; it extracts from the data itself.…”

Section: Introductionmentioning

confidence: 99%

Simultaneous wound border segmentation and tissue classification using a conditional generative adversarial network

Sarp

Kuzlu

Pipattanasomporn

et al. 2021

J. eng.

View full text Add to dashboard Cite

Generative adversarial network (GAN) applications on medical image synthesis have the potential to assist caregivers in deciding a proper chronic wound treatment plan by understanding the border segmentation and the wound tissue classification visually. This study proposes a hybrid wound border segmentation and tissue classification method utilising conditional GAN, which can mimic real data without expert knowledge. We trained the network on chronic wound datasets with different sizes. The performance of the GAN algorithm is evaluated through the mean squared error, Dice coefficient metrics and visual inspection of generated images. This study also analyses the optimum number of training images as well as the number of epochs using GAN for wound border segmentation and tissue classification. The results show that the proposed GAN model performs efficiently for wound border segmentation and tissue classification tasks with a set of 2000 images at 200 epochs.

show abstract

“…Another natural approach to perform "Data Synthesis" is through Generative Image Models, as done by several existing works, mostly with GANs (Goodfellow et al, 2014). Many works use generative models to synthesize samples to expand Image Classification datasets (Antoniou et al, 2017;Mariani et al, 2018;Wang and Perez, 2017;Yamaguchi et al, 2019;Zhang et al, 2019), and this constitutes a promising research direction, specially on domains like medical imaging (Frid-Adar et al, 2018), or situations with underrepresented classes (Lim et al, 2018;Pinetz et al, 2019;Zhu et al, 2018).…”

Section: Data Synthesismentioning

confidence: 99%

An analysis of sample synthesis for deep learning based object detection

Blanger¹

View full text Add to dashboard Cite

show abstract

Actual Impact of GAN Augmentation on CNN Classification Performance

Cited by 5 publications

References 0 publications

Impact of data synthesis strategies for the classification of craniosynostosis

Impact of data synthesis strategies for the classification of craniosynostosis

Simultaneous wound border segmentation and tissue classification using a conditional generative adversarial network

An analysis of sample synthesis for deep learning based object detection

Contact Info

Product

Resources

About