Search

Recent advancements in the field of image synthesis have led to the development of Neural Style Transfer (NST) and Generative Adversarial Networks (GANs) which have proven to be powerful tools for data augmentation and realistic data generation. While GANs have been widely used for both data augmentation and generation, NST has not been employed for data generation tasks. Nonetheless, the simpler structure of NST compared to GANs makes it a promising alternative. In this research, we introduce an NST-based method for data generation, which to the best of our knowledge, is the first of its kind. By taking advantage of simplified architecture of NST models attributed to the utilization of a real image as the style input, our method enhances performance in data generation tasks under limited resource conditions. Additionally by utilizing patch-based training and high-resolution inference process high quality images are synthesized with limited resources. Furthermore multi-model and noised input is utilized for increased diversity with the novel NST-based data generation approach. Our proposed method utilizes binary segmentation maps as the condition input, representing the cell and wound regions. We evaluate the performance of our proposed NST-based method and compare it with a modified and fine-tuned conditional GAN (C-GAN) methods for the purpose of conditional generation of phase-contrast wound healing assay images. Through a series of quantitative and qualitative analyses, we demonstrate that our NST-based method outperforms the modified C-GAN while utilizing fewer resources. Additionally, we show that our NST-based method enhances segmentation performance when used as a data augmentation method. Our findings provide compelling evidence regarding the potential of NST for data generation tasks and its superiority over traditional GAN-based methods. The NST for data generation method was implemented in Python language and will be accessible at https://github.com/IDU-CVLab/NST%5ffor%5fGen under the MIT licence. • First instance of data generation using the Neural Style Transfer method proposed for low-cost data synthesis. • The synthesis of phase-contrast microscopy images of wound-healing assays is reported for the first time. • Comparison of the performance of Neural Style Transfer and Conditional Generative Adversarial Networks for data augmentation and data synthesis tasks using qualitative and quantitative methods. [ABSTRACT FROM AUTHOR]

• Karyotyping is a popular method of chromosome analysis in cytogenetic laboratories to diagnose the genetic disorders. • Finding analyzable metaphase images is essential for karyotyping that needs to be done automatically and accurately. • Automation of finding analyzable metaphase images reduces the interobservor variation and speeds-up the karyotyping. • Enhancing the sensitivity and specificity of this task, has big effect on overal accuracy of karyotyping resulting to the more accurate diagnosis. • An automated high accurate metaphase finder decreases the clinical and sampling work and costs. Finding analyzable metaphase chromosome images is an essential step in karyotyping which is a common task for clinicians to diagnose cancers and genetic disorders precisely. This step is tedious and time-consuming. Hence developing automated fast and reliable methods to assist clinical technicians becomes indispensable. Previous approaches include methods with feature extraction followed by rule or quality based classifiers, component analysis, and neural networks. A two-stage automated metaphase-finding scheme, consisting of an image processing based metaphase detection stage, and a deep convolutional neural network based selection stage is proposed. The first stage detects metaphase images from 10× scan of specimen slides. The selection stage, on the other hand, selects the analyzable ones among them. The proposed scheme has a 99.33% true positive rate and 0.34% of the false positive rate of metaphase finding. This study demonstrates an effective scheme for the automated finding of analyzable metaphase images with high True positive and low False positive rates. [ABSTRACT FROM AUTHOR]

The joint estimation of the parameters and the states of the hemodynamic model from the blood oxygen level dependent (BOLD) signal is a challenging problem. In the functional magnetic resonance imaging (fMRI) literature, quite interestingly, many proposed algorithms work only as a filtering method. This makes the estimation of hidden states and parameters less reliable compared with the algorithms that use smoothing. In standard implementations, smoothing is performed only once. However, joint state and parameter estimation can be improved substantially by iterating smoothing schemes such as the extended Kalman smoother (IEKS). In the fMRI literature, extended Kalman filtering is thought to be less accurate than standard particle filtering (PF). We compared EKF with PF and observed that the contrary is true. We improved the EKF performance by adding smoother. By iterative scheme joint hemodynamic and parameter estimation is improved substantially. We compared IEKS performance with the square-root cubature Kalman smoother (SCKS) algorithm. We show that its accuracy for the state and the parameter estimation is better and much faster than iterative SCKS. SCKS was found to be a better estimator than the dynamic expectation maximization (DEM), EKF, local linearization filter (LLF) and PF methods. We show in this paper that IEKS is a better estimator than iterative SCKS under different process and measurement noise conditions. As a result, IEKS seems to be the best method we evaluated in all aspects. [ABSTRACT FROM AUTHOR]

[Display omitted] • Visual expression of breast cancer with IHC allows evaluation of CerbB2 receptors. • CerbB2 mutated breast carcinomas are suitable for targeted therapy. • Breast tumors are evaluated in four different scores as 0, 1, 2, and 3. • Scoring determines the applicability of CerbB2 protein specific treatment. • Pathologists decide the scores by eye. • This is a laborious task with high inter-observer variability. • We propose cell-based image analysis techniques to determine CerbB2/HER2 scores. • This is in accordance with ASCO/CAP recommendations. • The method provides an explainable AI solution for HER2 scoring. Background and Objective: Visual expression of invasive breast cancer with immunohistochemistry (IHC) allows evaluation of CerbB2 receptors, such that CerbB2 mutated breast carcinomas are suitable for targeted therapy. Breast tumors are evaluated in four different scores as 0, 1, 2, 3 to decide if it is suitable for the CerbB2 protein specific treatment or not. Pathologists try to decide the scores by eye, which is laborious, and error-prone work with high inter-observer variability. Methods: We propose cell based image analysis termine the CerbB2/HER2 scores in breast tissue images in accordance with ASCO/CAP recommendations, automatically. The proposed ASCO/CAP recommendations compliant image analysis approach provides an explainable artificial intelligence solution for HER2 tissue scoring. Firstly, tissue images are separated into hematoxylin and diaminobenzidine color channels with color deconvolution. Cell nuclei and boundaries are segmented with a hybrid multi-level thresholding and radial line based method on hematoxylin channel. Following ASCO/CAP recommendations, cell based features representing the intensity and completeness of circumferential membrane staining are extracted with the proposed Membrane Intensity Histogram (MIH) method. Extracted features are, then, fed into a classifier, such as, k-nearest neighbours, decision trees and long-short term memory, to determine cell based HER2 scores. Individual cell scores are combined according to ASCO/CAP recommendations to obtain the final CerbB2/HER2 tissue score. Another contribution of the paper is the introduction of two publicly available image data sets on CerbB2/HER2 tissue scoring. Clinical data sets, ITU-MED-1 and ITU-MED-2, are created by digitizing IHC slides from real patients, that have ground truth CerbB2/HER2 scores. Result: The proposed automatic scoring method is tested on these clinical data sets, as well as, on a HER2 Contest data set. Performance of the proposed explainable artificial intelligence approach for HER2 tissue scoring is evaluated and compared with state-of-the-art techniques in the literature. Conclusion: Results suggest that, the proposed method is highly effective in HER2 tissue scoring on both balanced and unbalanced data sets. Significance: A hand-crafted feature extraction approach for CerbB2/HER2 scoring is proposed which provides an explainable artificial intelligence framework. The proposed HER2 scoring method can be adapted to updates in ASCO/CAP recommendations without the need for re-training and/or re-designing the model. Moreover, two publicly available data sets, namely, ITU-MED-1 and ITU-MED-2 are introduced with corresponding score labels. [ABSTRACT FROM AUTHOR]