Better Aggregation in Test-Time Augmentation

Shanmugam, Divya; Blalock, Davis; Balakrishnan, Guha; Guttag, John

doi:10.1109/iccv48922.2021.00125

Cited by 83 publications

(71 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, there is a new tendency developing among the research community on deep learning in which samples are enhanced through the test data augmentation method [ 69 , 70 , 71 , 72 ]. Test data augmentation can help increase the robustness of a trained model [ 73 , 74 , 75 ]. Test data augmentation may be utilized to enhance deep network prediction performance and open up new intriguing possibilities for medical image interpretation [ 76 , 77 , 78 ].…”

Section: Datasetsmentioning

confidence: 99%

LungNet22: A Fine-Tuned Model for Multiclass Classification and Prediction of Lung Disease Using X-ray Images

Shamrat

Azam

Karim

et al. 2022

JPM

View full text Add to dashboard Cite

In recent years, lung disease has increased manyfold, causing millions of casualties annually. To combat the crisis, an efficient, reliable, and affordable lung disease diagnosis technique has become indispensable. In this study, a multiclass classification of lung disease from frontal chest X-ray imaging using a fine-tuned CNN model is proposed. The classification is conducted on 10 disease classes of the lungs, namely COVID-19, Effusion, Tuberculosis, Pneumonia, Lung Opacity, Mass, Nodule, Pneumothorax, and Pulmonary Fibrosis, along with the Normal class. The dataset is a collective dataset gathered from multiple sources. After pre-processing and balancing the dataset with eight augmentation techniques, a total of 80,000 X-ray images were fed to the model for classification purposes. Initially, eight pre-trained CNN models, AlexNet, GoogLeNet, InceptionV3, MobileNetV2, VGG16, ResNet 50, DenseNet121, and EfficientNetB7, were employed on the dataset. Among these, the VGG16 achieved the highest accuracy at 92.95%. To further improve the classification accuracy, LungNet22 was constructed upon the primary structure of the VGG16 model. An ablation study was used in the work to determine the different hyper-parameters. Using the Adam Optimizer, the proposed model achieved a commendable accuracy of 98.89%. To verify the performance of the model, several performance matrices, including the ROC curve and the AUC values, were computed as well.

show abstract

Section: Datasetsmentioning

confidence: 99%

LungNet22: A Fine-Tuned Model for Multiclass Classification and Prediction of Lung Disease Using X-ray Images

Shamrat

Azam

Karim

et al. 2022

JPM

View full text Add to dashboard Cite

show abstract

“…The aim of augmenting is to reduce the risk of incorrect predictions based on overfitting or too strict pattern learning [57], [58], [60]. In our analysis, we reused the Baseline models and applied random rotations as well as mirroring on all axes for inference.…”

Section: ) Augmentingmentioning

confidence: 99%

An Analysis on Ensemble Learning Optimized Medical Image Classification With Deep Convolutional Neural Networks

Müller

Soto-Rey

2022

IEEE Access

View full text Add to dashboard Cite

Novel and high-performance medical image classification pipelines are heavily utilizing ensemble learning strategies. The idea of ensemble learning is to assemble diverse models or multiple predictions and, thus, boost prediction performance. However, it is still an open question to what extent as well as which ensemble learning strategies are beneficial in deep learning based medical image classification pipelines. In this work, we proposed a reproducible medical image classification pipeline for analyzing the performance impact of the following ensemble learning techniques: Augmenting, Stacking, and Bagging. The pipeline consists of state-of-the-art preprocessing and image augmentation methods as well as 9 deep convolution neural network architectures. It was applied on four popular medical imaging datasets with varying complexity. Furthermore, 12 pooling functions for combining multiple predictions were analyzed, ranging from simple statistical functions like unweighted averaging up to more complex learning-based functions like support vector machines. Our results revealed that Stacking achieved the largest performance gain of up to 13% F1-score increase. Augmenting showed consistent improvement capabilities by up to 4% and is also applicable to single model based pipelines. Cross-validation based Bagging demonstrated significant performance gain close to Stacking, which resulted in an F1-score increase up to +11%. Furthermore, we demonstrated that simple statistical pooling functions are equal or often even better than more complex pooling functions. We concluded that the integration of ensemble learning techniques is a powerful method for any medical image classification pipeline to improve robustness and boost performance.

show abstract

“…To reduce this effect, we rely on test-time augmentation (TTA) [55], which can be seen as a form of ensembling: we average several runs on different subsamples. However, aggregating final results, as often done in TTA [97], would be very time consuming in our case as we would have to do it to answer the occupancy of each query, basically multiplying the inference running time by the number of subsamples.…”

Section: Refinementsmentioning

confidence: 99%

POCO: Point Convolution for Surface Reconstruction

Boulch¹,

Marlet²

2022

Preprint

View full text Add to dashboard Cite

Implicit neural networks have been successfully used for surface reconstruction from point clouds. However, many of them face scalability issues as they encode the isosurface function of a whole object or scene into a single latent vector. To overcome this limitation, a few approaches infer latent vectors on a coarse regular 3D grid or on 3D patches, and interpolate them to answer occupancy queries. In doing so, they loose the direct connection with the input points sampled on the surface of objects, and they attach information uniformly in space rather than where it matters the most, i.e., near the surface. Besides, relying on fixed patch sizes may require discretization tuning. To address these issues, we propose to use point cloud convolutions and compute latent vectors at each input point. We then perform a learning-based interpolation on nearest neighbors using inferred weights. Experiments on both object and scene datasets show that our approach significantly outperforms other methods on most classical metrics, producing finer details and better reconstructing thinner volumes. The code is available at https://github.com/ valeoai/POCO.

show abstract

Better Aggregation in Test-Time Augmentation

Cited by 83 publications

References 10 publications

LungNet22: A Fine-Tuned Model for Multiclass Classification and Prediction of Lung Disease Using X-ray Images

LungNet22: A Fine-Tuned Model for Multiclass Classification and Prediction of Lung Disease Using X-ray Images

An Analysis on Ensemble Learning Optimized Medical Image Classification With Deep Convolutional Neural Networks

POCO: Point Convolution for Surface Reconstruction

Contact Info

Product

Resources

About