What question did this study set out to answer?

To review and categorize data augmentation techniques utilized to train deep learning models on medical imaging data.

June 19, 2021Journal of Medical Imaging and Radiation Oncology

A review of medical image data augmentation techniques for deep learning applications

View Full Paper

Ask AI

Bookmark

Authors

PCPhillip ChlapLiverpool Hospital HMHang MinCommonwealth Scientific and Industrial Research Organisation NVNym VandenbergThe University of Sydney

Discussion

Member takes

Implication

Systematic review categorizes data augmentation methods for CT and MRI deep learning models, highlighting strategies to improve clinical artificial intelligence tools.

Key Points

To review and categorize data augmentation techniques utilized to train deep learning models on medical imaging data.
Conducted a systematic review of literature utilizing data augmentation to train deep learning models on medical images.
Restricted imaging modalities evaluated strictly to computed tomography (CT) and magnetic resonance imaging (MRI).
Categorized augmentation strategies into basic, deformable, deep learning-based, and other methods.
Identified data augmentation as an effective solution to overcome the scarcity of large medical imaging datasets in radiology and radiotherapy.
Demonstrated that training models with augmented data consistently improves performance on independent, unseen validation datasets.
Classified the landscape of clinical imaging augmentation into basic transformations, deformable models, and deep generative techniques.

Cite This Study

Chlap et al. (2021) studied this question.

synapsesocial.com/papers/69d8d7a7ade63f05b9bedaca https://doi.org/10.1111/1754-9485.13261

View Full Paper

Ask AI

Bookmark

Implication

Systematic review categorizes data augmentation methods for CT and MRI deep learning models, highlighting strategies to improve clinical artificial intelligence tools.

A review of medical image data augmentation techniques for deep learning applications

Discussion

Member takes

Implication

Key Points

Cite This Study

A review of medical image data augmentation techniques for deep learning applications

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider