Unravelling the effect of data augmentation transformations in polyp segmentation

Sánchez-Peralta, Luisa F.; Picón, Artzai; Sánchez‐Margallo, Francisco M.; Pagador, J. Blas

doi:10.1007/s11548-020-02262-4

Cited by 27 publications

(12 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hussain et al, 2017 ). Another report proposed to change image pixel brightness and contrast to improve the deep learning training results ( Sánchez-Peralta, Picón, Sánchez-Margallo, and Pagador, 2020 ). There are many methods of data augmentation.…”

Section: Methodsmentioning

confidence: 99%

Lumbar Disc Herniation Automatic Detection in Magnetic Resonance Imaging Based on Deep Learning

Tsai

Hung

Guo

et al. 2021

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Background: Lumbar disc herniation (LDH) is among the most common causes of lower back pain and sciatica. The causes of LDH have not been fully elucidated but most likely involve a complex combination of mechanical and biological processes. Magnetic resonance imaging (MRI) is a tool most frequently used for LDH because it can show abnormal soft tissue areas around the spine. Deep learning models may be trained to recognize images with high speed and accuracy to diagnose LDH. Although the deep learning model requires huge numbers of image datasets to train and establish the best model, this study processed enhanced medical image features for training the small-scale deep learning dataset.Methods: We propose automatic detection to assist the initial LDH exam for lower back pain. The subjects were between 20 and 65 years old with at least 6 months of work experience. The deep learning method employed the YOLOv3 model to train and detect small object changes such as LDH on MRI. The dataset images were processed and combined with labeling and annotation from the radiologist’s diagnosis record.Results: Our method proves the possibility of using deep learning with a small-scale dataset with limited medical images. The highest mean average precision (mAP) was 92.4% at 550 images with data augmentation (550-aug), and the YOLOv3 LDH training was 100% with the best average precision at 550-aug among all datasets. This study used data augmentation to prevent under- or overfitting in an object detection model that was trained with the small-scale dataset.Conclusions: The data augmentation technique plays a crucial role in YOLOv3 training and detection results. This method displays a high possibility for rapid initial tests and auto-detection for a limited clinical dataset.

show abstract

Section: Methodsmentioning

confidence: 99%

Lumbar Disc Herniation Automatic Detection in Magnetic Resonance Imaging Based on Deep Learning

Tsai

Hung

Guo

et al. 2021

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

show abstract

“…In this particular task, data transformation has been considered due to the relative absence of large annotated image databases of polyps. These approaches have included changing the image dimensions, changing pixel values, and adding in external conditions with the goal to maintain or achieve better generalizability for ML tools to detect polyps (Sánchez-Peralta et al, 2020).…”

Section: Case Study: Colonoscopy With Computer-aided Detectionmentioning

confidence: 99%

The Clinician's Guide to the Machine Learning Galaxy

et al. 2021

View full text Add to dashboard Cite

Machine learning (ML): a subfield of artificial intelligence, broadly refers to the ability of a computational platform to learn from data and make predictions or recommendations based on this data without being explicitly programmed. In general, there are two major categories of machine learning, supervised and unsupervised. Supervised learning is conducted with the concept of "truth" where the model tries to approximate the relationship between inputs and labeled outputs. For example, given images of cats and dogs, where each image has a correct answer, can you train a model that accurately identifies of cats versus dogs? Unsupervised learning is performed without data labels and the goal is for the computer to infer inherent structure or patterns in the data. For example, given a set of heart rate, accelerometer, and location data from a wearable fitness monitor, can the computer identify periods of rest versus exercise based on differences in the raw data?Neural networks (NN): a form of machine learning with a basic architecture consisting of nodes and connections existing in multiple layers, loosely analogous to neurons and synapses in the biological brain. This broad category is inclusive of many kinds of modern machine learning models which are used in tasks such as computer vision, voice recognition, bioinformatics, and among others. Deep learning:A broad family of neural network architectures that have multiple layers (aka deep).

show abstract

“…To compare the proposed data augmentation method with existing ones, we have augmented the dataset with height shift [20], rotation [21], and width shift [22]. The results of training the CNN on the data augmented with these methods are presented in Table 3.…”

Section: Comparison Of Different Data Augmentation Methodsmentioning

confidence: 99%

Hypertrophic Cardiomyopathy Diagnosis Based on Cardiovascular Magnetic Resonance Using Deep Learning Techniques

et al. 2021

View full text Add to dashboard Cite

Hypertrophic cardiomyopathy (HCM) can lead to serious cardiac problems. HCM is often diagnosed by an expert using cardiovascular magnetic resonance (CMR) images obtained from patients. In this research, we aimed to develop a deep learning technique to automate HCM diagnosis. CMR images of 37421 healthy and 21846 HCM patients were obtained during two years. Images obtained from female patients form 53% of the collected dataset. The mean and standard deviation of the dataset patients' age are 48.2 and 19.5 years, respectively. Three experts inspected images and determined whether a case has HCM or not. New data augmentation was used to generate new images by employing color filtering on the existing ones. To classify the augmented images, we used a deep convolutional neural network (CNN). To the best of our knowledge, this is the first time CNN is used for HCM diagnosis. We designed our CNN from scratch to reach acceptable diagnosis accuracy. Comparing the designed algorithm output with the experts' opinions, the method could achieve accuracy of 95.23%, recall of 97.90%, and specificity of 93.06% on the original dataset. The same performance metrics for the designed algorithm on the augmented dataset were 98.53%, 98.70%, and 95.21%, respectively. We have also experimented with different optimizers (e.g. Adadelta and Adagrad) and other data augmentation methods (e.g. height shift and rotation) to further evaluate the proposed method. Using our data augmentation method, accuracy of 98.53% were achieved which is higher than the best accuracy (95.83%) obtained by the other data augmentation methods which have been evaluated. The upper bound on difference between true error rate and empirical error rate of the proposed method has also been provided in order to present better performance analysis. The advantages of employing the proposed method are elimination of contrast agent and its complications, decreased CMR examination time, lower costs for patients and cardiac imaging centers.

show abstract

Unravelling the effect of data augmentation transformations in polyp segmentation

Cited by 27 publications

References 41 publications

Lumbar Disc Herniation Automatic Detection in Magnetic Resonance Imaging Based on Deep Learning

Lumbar Disc Herniation Automatic Detection in Magnetic Resonance Imaging Based on Deep Learning

The Clinician's Guide to the Machine Learning Galaxy

Hypertrophic Cardiomyopathy Diagnosis Based on Cardiovascular Magnetic Resonance Using Deep Learning Techniques

Contact Info

Product

Resources

About