Wyniki wyszukiwania - BazTech

1

The Development of a Generative Approach for Joint Super-Resolution Image Reconstruction from Highly Sparse Raw Data in the Context of MR-PET Imaging

Malczewski Krzysztof

Machine Graphics and Vision

|

2023

|

Vol. 32, No. 3/4

161--191

EN

The present study introduces a rapid and efficient approach for reconstructing high-resolution images in hybrid MRI-PET scanners. The application of sparsity, compressed sensing (CS), and super-resolution reconstruction (SRR) methodologies can significantly decrease the demands of data acquisition while concurrently attaining high-resolution output. G-guided generative multilevel networks for sparsely sampled MR-PET input are shown here. Compressed Sensing using conjugate symmetry and Partial Fourier methodology speeds up data collection over k-space sampling methods. GANs and k-space adjustments are used in this image domain technique. The employed methodology utilizes discrete preprocessing stages to effectively tackle the challenges associated with the deblurring, reducing motion artifacts, and denoising of layers. Initial trials offer contextual details and accelerate evaluations. Preliminary experiments provide contextual information and expedite assessments.

2

Applications of generative models with a latent observation subspace in vibrodiagnostics

Puchalski Andrzej, Komorska Iwona

Diagnostyka

|

2023

|

Vol. 24, No. 4

art. no. 2023413

EN

The vibration signal is one of the most essential diagnostic signals, the analysis of which allows for determining the dynamic state of the monitored machine set. In the era of cyber-physical industrial systems, making diagnostic decisions involves the study of large databases from previous registers and data downloaded from machines in real-time. However, the recorded signals mainly concern the operational status of the monitored object. Insufficient training data regarding failure states hinders the operation of classification algorithms. Progress in machine learning has created a new avenue for the advancement of diagnostic methods based on models. These methods now have the capability to produce signals through random sampling from a hidden space or generate fresh instances of input data from noise. The article suggests the use of a Generative Adversarial Network (GAN) model as a tool to create synthetic measurement observations for vibration monitoring. The effectiveness of the synthetic data generation algorithm was verified on the example of the vibration signal recorded during tests of the drive system of a motor vehicle.

3

Automated detection of crystalline retinopathy via fundus photography using multistage generative adversarial networks

Young Choi Eun, Han Seung Hoon, Hee Ryu Ik, Kuk Kim Jin, Sik Lee In, Han Eoksoo, Kim Hyungsu, Yul Choi Joon, Keun Yoo Tae

Biocybernetics and Biomedical Engineering

|

2023

|

Vol. 43, no. 4

725--735

EN

Purpose: Crystalline retinopathy is characterized by reflective crystal deposits in the macula and is caused by various systemic conditions including hereditary, toxic, and embolic etiologies. Herein, we introduce a novel application of deep learning with a multistage generative adversarial network (GAN) to detect crystalline retinopathy using fundus photography. Methods: The dataset comprised major classes (healthy retina, diabetic retinopathy, exudative age-related macular degeneration, and drusen) and a crystalline retinopathy class (minor set). To overcome the limited data on crystalline retinopathy, we proposed a novel multistage GAN framework. The GAN was retrained after CutMix combination by inputting the GAN-generated synthetic data as new inputs to the original training data. After the multistage CycleGAN augmented the data for crystalline retinopathy, we built a deep-learning classifier model for detection. Results: Using the multistage CycleGAN facilitated realistic fundus photography synthesis with the characteristic features of retinal crystalline deposits. The proposed method outperformed typical transfer learning, prototypical networks, and knowledge distillation for both multiclass and binary classifications. The final model achieved an area under the curve of the receiver operating characteristics of 0.962 for internal validation and 0.987 for external validation for the detection of crystalline retinopathy. Conclusion: We introduced a deep learning approach for detecting crystalline retinopathy, a potential biomarker of underlying systemic pathological conditions. Our approach enables realistic pathological image synthesis and more accurate prediction of crystalline retinopathy, an essential but minor retinal condition.

4

Research on garden landscape reconstruction based on geographic information system under the background of deep learning

Cui Ying

Acta Geophysica

|

2023

|

Vol. 71, no. 3

1491--1513

EN

In the context of deep learning, this paper combines the arbitration mechanism to propose a GAN (Arbi-DCGAN) model based on the arbitration mechanism. First, the network structure of the proposed improved algorithm is composed of generator, discriminator and arbitrator. Then, the generator and the discriminator will conduct adversarial training according to the training plan and strengthen the ability of generating images and distinguishing the authenticity of the images according to the characteristics learned from the data set. Secondly, the arbitrator is composed of the generator, discriminator and measurement score computation module that have undergone the previous adversarial training. The arbitrator will feed back the results of the metric generator and discriminator adversarial training to the training plan. Finally, a winning limit is added to the network structure to improve the stability of model training, and the Circle loss function is used to replace the BCE loss function, which makes the model optimization process more flexible and the convergence state more clear. On the basis of geographic information system, this paper uses 325 meticulously annotated sample plans to establish a data set for deep learning, and trains the Arbi-DCGAN model to achieve the task of extracting land plots of different land types in the plan, as well as from the plane color block map to the color texture. The rendering and generation of the map complete the reconstruction task of the garden landscape. In addition, we further evaluate the results of the model's reconstruction of the garden landscape from the aspects of image quality, correct standardization and color expression. The training model has the potential to be applied to land type analysis and plane rendering in landscape architecture cases, helping designers improve the efficiency of analysis and drawing.

5

Cell image augmentation for classification task using GANs on Pap smear dataset

Zak Jakub, Grzeszczyk Michal K., Pater Antonina, Roszkowiak Lukasz, Siemion Krzysztof, Korzynska Anna

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 3

995--1011

EN

One of the solutions to the problem of insufficiently large training datasets in image processing is data augmentation. This process artificially extends the size of training datasets to avoid overfitting. Generative Adversarial Networks yield that become increasingly difficult to differentiate from real images, until the differentiation is no longer possible. Thus, artificial images closely resembling original ones can be generated. Inclusion of artificial images contributes to improving the training process. Medical domain is one of the areas where data acquisition is burdened by many procedures, laws, and prohibitions. As a result the potential size of collected datasets is reduced. This article presents the results of training Convolutional Neural Networks on a artificially extended image datasets. The resulting classification accuracy on a cell classification task of models trained with images generated using the proposed method were increased by up to 12.9% in comparison to that of the model trained only with original dataset from the HErlev Pap smear dataset.

6

Underwater image enhancement via efficient generative adversarial network

Qian Xin, Ge Peng

Optica Applicata

|

2021

|

Vol. 51, nr 4

483--497

EN

Underwater image enhancement has been receiving much attention due to its significance in facilitating various marine explorations. Inspired by the generative adversarial network (GAN) and residual network (ResNet) in many vision tasks, we propose a simplified designed ResNet model based on GAN called efficient GAN (EGAN) for underwater image enhancement. In particular, for the generator of EGAN we design a new pair of convolutional kernel size for the residual block in the ResNet. Secondly, we abandon batch normalization (BN) after every convolution layer for faster training and less artifacts. Finally, a smooth loss function is introduced for halo-effect alleviation. Extensive qualitative and quantitative experiments show that our methods accomplish considerable improvements compared to the state-of-the-art methods.

7

Epileptic seizure prediction using scalp electroencephalogram signals

Usman Syed Muhammad, Khalid Shehzad, Bashir Zafar

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 1

211--220

EN

Epilepsy is a brain disorder in which patients undergo frequent seizures. Around 30% of patients affected with epilepsy cannot be treated with medicines/surgical procedures. Abnormal activity, known as the preictal state starts few minutes before the seizure actually occurs. Therefore, it may be possible to deliver medication prior to the occurrence of a seizure if initiation of the preictal state can predicted before the seizure onset. We propose an epileptic seizure prediction method that predicts the preictal state before the seizure onset using electroencephalogram (EEG) monitoring of brain activity. It involves three steps including preprocessing of EEG signals, feature extraction classification of preictal and interictal states. In our proposed method, we have used (i) Empirical model decomposition to remove noise from the EEG signals and Generative Adversarial Networks to generate preictal samples to deal with the class imbalance problem; (ii) Automated features have been extracted with three layer Convolutional Neural Networks and (iii) Classification between preictal and interictal states is done with Long Short Term Memory units. In this study, we have used CHBMIT dataset of scalp EEG signals and have validated our proposed method on 22 subjects of dataset. Our proposed seizure prediction method is able to achieve 93% sensitivity and 92.5% specificity with average time of 32 min to predict the seizure's onset. Results obtained from our method have been compared with recent state-of-the-art epileptic seizure prediction methods. Our proposed method performs better in terms of sensitivity, specificity and average anticipation time.