D. Mandic

Zehua Chen, Yuyang Miao, Liyuan Wang et al.

Cardiovascular signals such as photoplethysmography (PPG), electrocardiography (ECG), and blood pressure (BP) are inherently correlated and complementary, together reflecting the health of cardiovascular system. However, their joint utilization in real-time monitoring is severely limited by diverse acquisition challenges from noisy wearable recordings to burdened invasive procedures. Here we propose UniCardio, a multi-modal diffusion transformer that reconstructs low-quality signals and synthesizes unrecorded signals in a unified generative framework. Its key innovations include a specialized model architecture to manage the signal modalities involved in generation tasks and a continual learning paradigm to incorporate varying modality combinations. By exploiting the complementary nature of cardiovascular signals, UniCardio clearly outperforms recent task-specific baselines in signal denoising, imputation, and translation. The generated signals match the performance of ground-truth signals in detecting abnormal health conditions and estimating vital signs, even in unseen domains, while ensuring interpretability for human experts. These advantages position UniCardio as a promising avenue for advancing AI-assisted healthcare.

Nina Moutonnet, Gregory Scott, Danilo P. Mandic

The performance of deep learning methods critically depends on the quality and quantity of the available training data. This is especially the case for physiological time series, which are both noisy and scarce, which calls for data augmentation to artificially increase the size of datasets. Another issue is that the time-evolving statistical properties of nonstationary signals prevent the use of standard data augmentation techniques. To this end, we introduce a novel method for augmenting nonstationary time series. This is achieved by combining offline changepoint detection with the iterative amplitude-adjusted Fourier transform (iAAFT), which ensures that the time-frequency properties of the original signal are preserved during augmentation. The proposed method is validated through comparisons of the performance of i) a deep learning seizure detection algorithm on both the original and augmented versions of the CHB-MIT and Siena scalp electroencephalography (EEG) databases, and ii) a deep learning atrial fibrillation (AF) detection algorithm on the original and augmented versions of the Computing in Cardiology Challenge 2017 dataset. By virtue of the proposed method, for the CHB-MIT and Siena datasets respectively, accuracy rose by 4.4% and 1.9%, precision by 10% and 5.5%, recall by 3.6% and 0.9%, and F1 by 4.2% and 1.4%. For the AF classification task, accuracy rose by 0.3%, precision by 2.1%, recall by 0.8%, and F1 by 2.1%.

Yiming Jiang, Jinlan Liu, Dongpo Xu et al.

Adam-type algorithms have become a preferred choice for optimisation in the deep learning setting, however, despite success, their convergence is still not well understood. To this end, we introduce a unified framework for Adam-type algorithms (called UAdam). This is equipped with a general form of the second-order moment, which makes it possible to include Adam and its variants as special cases, such as NAdam, AMSGrad, AdaBound, AdaFom, and Adan. This is supported by a rigorous convergence analysis of UAdam in the non-convex stochastic setting, showing that UAdam converges to the neighborhood of stationary points with the rate of $\mathcal{O}(1/T)$. Furthermore, the size of neighborhood decreases as $β$ increases. Importantly, our analysis only requires the first-order momentum factor to be close enough to 1, without any restrictions on the second-order momentum factor. Theoretical results also show that vanilla Adam can converge by selecting appropriate hyperparameters, which provides a theoretical guarantee for the analysis, applications, and further developments of the whole class of Adam-type algorithms.

M. Chang, N. Nishikawa, Z. R. Struzik et al.

The aim of this study is to provide a comprehensive test of three spatial speller settings, for the auditory, visual, and audiovisual paradigms. For rigour, the study is conducted with 16 BCI-naïve subjects in an experimental set-up based on five Japanese hiragana characters. Auditory P300 responses give encouragingly longer target vs. non-target latencies during the training phase, however, real-world online BCI experiments in the multimodal setting do not validate this potential advantage. Our case studies indicate that the auditory spatial unimodal paradigm needs further development in order to be a viable alternative to the established visual domain speller applications, as far as BCI-naïve subjects are concerned.

H. Mori, Y. Matsumoto, Z. R. Struzik et al.

We study the extent to which vibrotactile stimuli delivered to the head of a subject can serve as a platform for a brain computer interface (BCI) paradigm. Six head positions are used to evoke combined somatosensory and auditory (via the bone conduction effect) brain responses, in order to define a multimodal tactile and auditory brain computer interface (taBCI). Experimental results of subjects performing online taBCI, using stimuli with a moderately fast inter-stimulus interval (ISI), validate the taBCI paradigm, while the feasibility of the concept is illuminated through information transfer rate case studies.