Artificial image objects for classification of breast cancer biomarkers with transcriptome sequencing data and convolutional neural network algorithms

Background Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively for clinical applications due to the high dimension of the data and the highly correlated expression between individual genes. Methods We proposed a method to transform RNA sequencing data into artificial image objects (AIOs) and applied convolutional neural network (CNN) algorithms to classify these AIOs. With the AIO technique, we considered each gene as a pixel in an image and its expression level as pixel intensity. Using the GSE96058 (n = 2976), GSE81538 (n = 405), and GSE163882 (n = 222) datasets, we created AIOs for the subjects and designed CNN models to classify biomarker Ki67 and Nottingham histologic grade (NHG). Results With fivefold cross-validation, we accomplished a classification accuracy and AUC of 0.821 ± 0.023 and 0.891 ± 0.021 for Ki67 status. For NHG, the weighted average of categorical accuracy was 0.820 ± 0.012, and the weighted average of AUC was 0.931 ± 0.006. With GSE96058 as training data and GSE81538 as testing data, the accuracy and AUC for Ki67 were 0.826 ± 0.037 and 0.883 ± 0.016, and that for NHG were 0.764 ± 0.052 and 0.882 ± 0.012, respectively. These results were 10% better than the results reported in the original studies. For Ki67, the calls generated from our models had a better power for prediction of survival as compared to the calls from trained pathologists in survival analyses. Conclusions We demonstrated that RNA sequencing data could be transformed into AIOs and be used to classify Ki67 status and NHG with CNN algorithms. The AIO method could handle high-dimensional data with highly correlated variables, and there was no need for variable selection. With the AIO technique, a data-driven, consistent, and automation-ready model could be developed to classify biomarkers with RNA sequencing data and provide more efficient care for cancer patients.

Motion Magnification in Solar Imaging Data Sequences in the Sub-pixel Regime

The capability of the motion-magnification technique for the detection of transverse oscillations, such as kink oscillations of solar coronal loops observed with an imaging telescope, in the sub-pixel regime is investigated. The technique is applied to artificial-image sequences imitating harmonic transverse displacements of the loop, observed in the optically thin regime. Motion magnification is found to work well on the analysis of sub-pixel, $\geq 0.01$ ≥ 0.01 pixel oscillations, and it is characterised by linear scaling between the magnified amplitude and input amplitude. Oscillations of loops with transverse density profiles of different steepness are considered. After magnification, the original transverse profiles are preserved sufficiently well. The motion-magnification performance is found to be robust in noisy data, for coloured noise with spectral indices ranging from 0 to 3, and additional Poisson noise with a signal-to-background-noise ratio down to unity. Our findings confirm the reliability of the motion-magnification technique for applications in magnetohydrodynamic seismology of the solar corona.

Deep Learning-Based Surrogate Modeling via Physics-Informed Artificial Image (PIAI) for Strongly Coupled Multidisciplinary Engineering Systems

The design of strongly coupled multidisciplinary engineering systems is challenging since it is characterized by the complex interaction of different disciplines. Such complexity cannot be easily captured by explicit analytical solutions, which motivates the development of surrogate modeling. It enables the prediction of the systems’ behavior without analytical formulations. Among existing surrogate modeling techniques, deep learning has gained significant interest because of the flexibility of non-linear formulation and applicability to data-driven analysis. Notably, the convolution neural networks-based deep surrogate model augments the precision of prediction and estimation of system behavior once image-based inputs representing physical experiments and simulation are employed. Nevertheless, the feasibility of the deep surrogate model is often flawed due to the miserable correlation representation between design parameters and the corresponding responses. Massive training costs also degrade the performance of the predictive model. To address those issues, this research proposes a physics-informed artificial image (PiAI) that incubates geometry-informed CAD, location-clarified filter, and essential simulation conditions, which augments the prediction credibility. Moreover, in lieu of employing multimodalities or multiple image channels, the proposed method employs a unimodal-based single image input to increase computational efficiency. The proposed framework’s efficacy and applicability are addressed in practical engineering design applications: cantilever beam and stretchable strain sensor.

Thoughts on the Ageing Body

Recent studies have taken into account the fact that the lives of older people have changed drastically in the past fifty years. Older people today engage more with society and are also expected to maintain an active role in their communities. In order to maintain a positive social status, todays older adults need both to challenge negative stereotypes and also to achieve the “unachievable” positive representations in the media. Society plays a complex game of bodily images: the artificial image of the human body in the media, the image that individuals try to project, and the image that society reflects back to the individual. When the three don’t coincide, the collision creates a distancing effect. To truly understand the lived experiences of older adults in contemporary society we must explore the changing perceptions of the body. This review will illustrate the arguments both classical and contemporary through an exploration of the ageing female body, which remains the focus of most of the literature.

Artificial Image Objects for Classification of Breast Cancer Biomarkers with Transcriptome Sequencing Data and Convolutional Neural Network Algorithms

Background: Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively to due to the high dimension of the data and the high correlation of gene expression. Methods: We propose a novel method that transforms RNA sequencing data into artificial image objects (AIOs) and apply convolutional neural network (CNN) algorithm to classify these AIOs. The AIO technique considers each gene as a pixel in digital image, standardizes and rescales gene expression levels into a range suitable for image display. Using the GSE81538 (n = 405) and GSE96058 (n = 3,373) datasets, we create AIOs for the subjects and design CNN models to classify biomarker Ki67 and Nottingham histologic grade (NHG). Results: With 5-fold cross validation, we accomplish a classification accuracy and AUC of 0.797 ± 0.034 and 0.820 ± 0.064 for Ki67 status. For NHG, the weighted average of categorical accuracy is 0.726 ± 0.018, and the weighted average of AUC is 0.848 ± 0.019. With GSE81538 as training data and GSE96058 as testing data, the accuracy and AUC for Ki67 are 0.772 ± 0.014 and 0.820 ± 0.006, and that for NHG are 0.682 ± 0.013 and 0.808 ± 0.003 respectively. These results are comparable to or better than the results reported in the original study. For both Ki67 and NHG, the calls from our models have similar predictive power for survival as the calls from trained pathologists in survival analyses. Comparing the calls from our models and the pathologists, we find that the discordant subjects for Ki67 are a group of patients for whom estrogen receptor, progesterone receptor, PAM50 and NHG could not predict their survival rate, and their responses to chemotherapy and endocrine therapy are also different from the concordant subjects. Conclusions: RNA sequencing data can be transformed into AIOs and be used to classify the status of Ki67 and NHG by CNN algorithm. The AIO method can handle high dimension data with highly correlated variables with no requirement for variable selection, leading to a data-driven, consistent and automation-ready approach to model RNA sequencing data.

Criminal Face Recognition Using GAN

Forensics and security at present often use low technological resources. Security measures often fail to update with the upcoming technology. This project is based on implementing an automatic face recognition of criminals or specific targets using machine-learning approach. Given a set of features to a Generative Adversarial Network(GAN), the algorithm generates an image of the target with the specified feature set. The input to the machine can either be a given set of features or a set of portraits varying from frontals to side profiles from which these features can be extracted. The accuracy of the system is directly proportional to the number of epochs trained in the network. The generated output image can vary from primitive, low resolution images to high quality images where features are more recognizable. This is then compared with a predefined database of existing people. Thus, the target can immediately be recognized with the generation of an artificial image with the given biometric feature set, which will be again compared by a discriminator network to check the true identity of the target.