Image pre-processing: enhance the performance of medical image classification using various data augmentation technique

Rama, J; Nalini, C.; Kumaravel, A.

doi:10.19101/tipcv.413001

Cited by 10 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We expected that the reflection over the vertical axis would have worked very well due the bilateral symmetry of the human body, but it performed quite poorly in comparison of the random rotation of 15 degrees or less. In research by Rama et al [22], augmentation based on reflections also led the least accuracy in classification of lung X-rays of TB patients (different data set than here). However, the reflection over the vertical axis worked the best on the HNC MRI and PET data sets which contain highly heterogeneous images of the human head and neck area and are therefore difficult to classify correctly.…”

Section: Discussionmentioning

confidence: 69%

“…Furthermore, one simple augmentation transformation is to add blur to the images, which might produce very different results for an imaging method such as positron emission tomography (PET) that produces already blurry images without sharp borders between regions than it would for magnetic resonance imaging (MRI). In existing literature [4,9,10,13,22], these augmentations have typically been compared by using only one data set rather than analyzing more systemically the differences caused by the different imaging modalities.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of simple augmentation transformations for a convolutional neural network classifying medical images

Rainio,

Klén

2024

SIViP

View full text Add to dashboard Cite

Simple image augmentation techniques, such as reflection, rotation, or translation, might work differently for medical images than they do for regular photographs due to the fundamental properties of medical imaging techniques and the bilateral symmetry of the human body. Here, we compare the predictions of a convolutional neural network (CNN) trained for binary classification by using either no augmentation or one of seven usual types augmentation. We have 11 different medical data sets, mostly related to lung infections or cancer, with X-rays, ultrasound (US) images, and images from positron emission tomography (PET) and magnetic resonance imaging (MRI). According to our results, the augmentation types do not produce statistically significant differences for US and PET data sets, but, for X-rays and MRI images, the best augmentation technique is adding Gaussian blur to images.

show abstract

Section: Discussionmentioning

confidence: 69%

Section: Introductionmentioning

confidence: 99%

Comparison of simple augmentation transformations for a convolutional neural network classifying medical images

Rainio,

Klén

2024

SIViP

View full text Add to dashboard Cite

show abstract

“…Flipping, as a geometric augmentation technique, often appears as a convenient tactic for natural images, and numerous research studies have been conducted in this field ( 16 ). However, in medical imaging study, flipping including both vertical and horizontal are adopted widely across several modalities of mammogram ( 32 , 33 ), dermoscopy images ( 34 , 35 ), chest CT scan ( 36 , 37 ), chest X-ray ( 15 , 38 ), brain tumor MRI ( 21 , 39 ), tympanic membrane ( 40 , 41 ), breast cancer histopathology image ( 42 , 43 ), and breast cancer ultrasound images ( 44 , 45 ) which might be an obvious reason acquiring poor performance as the alteration may not result in clinical possible images. Though flipping a medical image such as an MRI scan would cause a scan one would almost never see in the clinical setting, it is often claimed to be an effective strategy ( 39 , 46 ).…”

Section: Proposed Methodologymentioning

confidence: 99%

MNet-10: A robust shallow convolutional neural network model performing ablation study on medical images assessing the effectiveness of applying optimal data augmentation technique

et al. 2022

View full text Add to dashboard Cite

Interpretation of medical images with a computer-aided diagnosis (CAD) system is arduous because of the complex structure of cancerous lesions in different imaging modalities, high degree of resemblance between inter-classes, presence of dissimilar characteristics in intra-classes, scarcity of medical data, and presence of artifacts and noises. In this study, these challenges are addressed by developing a shallow convolutional neural network (CNN) model with optimal configuration performing ablation study by altering layer structure and hyper-parameters and utilizing a suitable augmentation technique. Eight medical datasets with different modalities are investigated where the proposed model, named MNet-10, with low computational complexity is able to yield optimal performance across all datasets. The impact of photometric and geometric augmentation techniques on different datasets is also evaluated. We selected the mammogram dataset to proceed with the ablation study for being one of the most challenging imaging modalities. Before generating the model, the dataset is augmented using the two approaches. A base CNN model is constructed first and applied to both the augmented and non-augmented mammogram datasets where the highest accuracy is obtained with the photometric dataset. Therefore, the architecture and hyper-parameters of the model are determined by performing an ablation study on the base model using the mammogram photometric dataset. Afterward, the robustness of the network and the impact of different augmentation techniques are assessed by training the model with the rest of the seven datasets. We obtain a test accuracy of 97.34% on the mammogram, 98.43% on the skin cancer, 99.54% on the brain tumor magnetic resonance imaging (MRI), 97.29% on the COVID chest X-ray, 96.31% on the tympanic membrane, 99.82% on the chest computed tomography (CT) scan, and 98.75% on the breast cancer ultrasound datasets by photometric augmentation and 96.76% on the breast cancer microscopic biopsy dataset by geometric augmentation. Moreover, some elastic deformation augmentation methods are explored with the proposed model using all the datasets to evaluate their effectiveness. Finally, VGG16, InceptionV3, and ResNet50 were trained on the best-performing augmented datasets, and their performance consistency was compared with that of the MNet-10 model. The findings may aid future researchers in medical data analysis involving ablation studies and augmentation techniques.

show abstract

“…To analyze horizontal OCT images of right and left eyes together, OCT images of the left eyes were flipped horizontally. Data augmentation is a promising way to increase the performance of classification tasks 25 by generating more samples from the available images 26 . Brightness control and image shift cropping were adopted for data augmentation 27 .…”

Section: Preprocessing and Data Augmentationmentioning

confidence: 99%

Deep learning models for screening of high myopia using optical coherence tomography

Choi

Roh

et al. 2021

Sci Rep

View full text Add to dashboard Cite

This study aimed to validate and evaluate deep learning (DL) models for screening of high myopia using spectral-domain optical coherence tomography (OCT). This retrospective cross-sectional study included 690 eyes in 492 patients with OCT images and axial length measurement. Eyes were divided into three groups based on axial length: a “normal group,” a “high myopia group,” and an “other retinal disease” group. The researchers trained and validated three DL models to classify the three groups based on horizontal and vertical OCT images of the 600 eyes. For evaluation, OCT images of 90 eyes were used. Diagnostic agreements of human doctors and DL models were analyzed. The area under the receiver operating characteristic curve of the three DL models was evaluated. Absolute agreement of retina specialists was 99.11% (range: 97.78–100%). Absolute agreement of the DL models with multiple-column model was 100.0% (ResNet 50), 90.0% (Inception V3), and 72.22% (VGG 16). Areas under the receiver operating characteristic curves of the DL models with multiple-column model were 0.99 (ResNet 50), 0.97 (Inception V3), and 0.86 (VGG 16). The DL model based on ResNet 50 showed comparable diagnostic performance with retinal specialists. The DL model using OCT images demonstrated reliable diagnostic performance to identify high myopia.

show abstract

Image pre-processing: enhance the performance of medical image classification using various data augmentation technique

Cited by 10 publications

References 7 publications

Comparison of simple augmentation transformations for a convolutional neural network classifying medical images

Comparison of simple augmentation transformations for a convolutional neural network classifying medical images

MNet-10: A robust shallow convolutional neural network model performing ablation study on medical images assessing the effectiveness of applying optimal data augmentation technique

Deep learning models for screening of high myopia using optical coherence tomography

Contact Info

Product

Resources

About