Search Results (3)

Despite the proliferation of deep learning techniques for accelerated MRI acquisition and enhanced image reconstruction, the construction of large and diverse MRI datasets continues to pose a barrier to effective clinical translation of these technologies. One major challenge is in collecting the MRI raw data (required for image reconstruction) from clinical scanning, as only magnitude images are typically saved and used for clinical assessment and diagnosis. The image phase and multi-channel RF coil information are not retained when magnitude-only images are saved in clinical imaging archives. Additionally, preprocessing used for data in clinical imaging can lead to biased results. While several groups have begun concerted efforts to collect large amounts of MRI raw data, current databases are limited in the diversity of anatomy, pathology, annotations, and acquisition types they contain. To address this, we present a method for synthesizing realistic MR data from magnitude-only data, allowing for the use of diverse data from clinical imaging archives in advanced MRI reconstruction development. Our method uses a conditional GAN-based framework to generate synthetic phase images from input magnitude images. We then applied ESPIRiT to derive RF coil sensitivity maps from fully sampled real data to generate multi-coil data. The synthetic data generation method was evaluated by comparing image reconstruction results from training Variational Networks either with real data or synthetic data. We demonstrate that the Variational Network trained on synthetic MRI data from our method, consisting of GAN-derived synthetic phase and multi-coil information, outperformed Variational Networks trained on data with synthetic phase generated using current state-of-the-art methods. Additionally, we demonstrate that the Variational Networks trained with synthetic k-space data from our method perform comparably to image reconstruction networks trained on undersampled real k-space data. Full article

The recent explosive growth in the number of smart technologies relying on data collected from sensors and processed with machine learning classifiers made the training data imbalance problem more visible than ever before. Class-imbalanced sets used to train models of various events of interest are among the main reasons for a smart technology to work incorrectly or even to completely fail. This paper presents an attempt to resolve the imbalance problem in sensor sequential (time-series) data through training data augmentation. An Unrolled Generative Adversarial Networks (Unrolled GAN)-powered framework is developed and successfully used to balance the training data of smartphone accelerometer and gyroscope sensors in different contexts of road surface monitoring. Experiments with other sensor data from an open data collection are also conducted. It is demonstrated that the proposed approach allows for improving the classification performance in the case of heavily imbalanced data (the F1 score increased from 0.69 to 0.72, $p<0.01$, in the presented case study). However, the effect is negligible in the case of slightly imbalanced or inadequate training sets. The latter determines the limitations of this study that would be resolved in future work aimed at incorporating mechanisms for assessing the training data quality into the proposed framework and improving its computational efficiency. Full article

Co-speech gestures are a crucial, non-verbal modality for humans to communicate. Social agents also need this capability to be more human-like and comprehensive. This study aims to model the distribution of gestures conditioned on human speech features. Unlike previous studies that try to find injective functions that map speech to gestures, we propose a novel, conditional GAN-based generative model to not only convert speech into gestures but also to approximate the distribution of gestures conditioned on speech through parameterization. An objective evaluation and user study show that the proposed model outperformed the existing deterministic model, indicating that generative models can approximate real patterns of co-speech gestures better than the existing deterministic model. Our results suggest that it is critical to consider the nature of randomness when modeling co-speech gestures. Full article