Medical image data augmentation: techniques, comparisons and interpretations

Abstract

Designing deep learning based methods with medical images has always been an attractive area of research to assist clinicians in rapid examination and accurate diagnosis. Those methods need a large number of datasets including all variations in their training stages. On the other hand, medical images are always scarce due to several reasons, such as not enough patients for some diseases, patients do not want to allow their images to be used, lack of medical equipment or equipment, inability to obtain images that meet the desired criteria. This issue leads to bias in datasets, overfitting, and inaccurate results. Data augmentation is a common solution to overcome this issue and various augmentation techniques have been applied to different types of images in the literature. However, it is not clear which data augmentation technique provides more efficient results for which image type since different diseases are handled, different network architectures are used, and these architectures are trained and tested with different numbers of data sets in the literature. Therefore, in this work, the augmentation techniques used to improve performances of deep learning based diagnosis of the diseases in different organs (brain, lung, breast, and eye) from different imaging modalities (MR, CT, mammography, and fundoscopy) have been examined. Also, the most commonly used augmentation methods have been implemented, and their effectiveness in classifications with a deep network has been discussed based on quantitative performance evaluations. Experiments indicated that augmentation techniques should be chosen carefully according to image types.

Transfer Learning for Effective Urolithiasis Detection

“…The data utilized in this study comprised urinary tract stones, which were expanded through various data augmentation techniques [ 12 ]. These methods have been employed to enhance the model’s performance in processing diverse urinary images in real-world settings.…”

Swin Routiformer: Moss Classification Algorithm Based on Swin Transformer With Bi-Level Routing Attention

“…In addressing the unique challenges posed by the small and amorphous morphological features of moss, traditional image augmentation methods often fall short, necessitating a cautious approach to the selection of augmentation algorithms. These algorithms must be carefully matched to the specific characteristics of the images and the classification task at hand, as highlighted in references [29] and [30]. When the original moss images are fed into Swin-R, they are resized to 224×224 pixels (as shown in Table 2).…”

Brain tumour segmentation framework with deep nuanced reasoning and Swin‐T

“…As a potential future endeavour, a comparison between the performance of the proposed method and segmentation methods based on capsule networks could be explored, delving into their respective advantages. Lastly, it is well recognized that deep networks require ample data and various augmentation techniques have been applied to address issues and enhance the reliability and robustness of deep networks [49–51]. Therefore, as an extension of this work, evaluating the proposed method with an increased amount of data could be considered.…”