CT report generation (CTRG) aims to automatically generate diagnostic reports for 3D volumes, relieving clinicians' workload and improving patient care. Despite clinical value, existing works fail to effectively incorporate diagnostic information from multiple anatomical views and lack related clinical expertise essential for accurate and reliable diagnosis. To resolve these limitations, we propose a novel Multi-view perception Knowledge-enhanced Transformer (MvKeTR) to mimic the diagnostic workflow of clinicians. Just as radiologists first examine CT scans from multiple planes, a Multi-View Perception Aggregator (MVPA) with view-aware attention effectively synthesizes diagnostic information from multiple anatomical views. Then, inspired by how radiologists further refer to relevant clinical records to guide diagnostic decision-making, a Cross-Modal Knowledge Enhancer (CMKE) retrieves the most similar reports based on the query volume to incorporate domain knowledge into the diagnosis procedure. Furthermore, instead of traditional MLPs, we employ Kolmogorov-Arnold Networks (KANs) with learnable nonlinear activation functions as the fundamental building blocks of both modules to better capture intricate diagnostic patterns in CT interpretation. Extensive experiments on the public CTRG-Chest-548K dataset demonstrate that our method outpaces prior state-of-the-art (SOTA) models across almost all metrics. The code will be made publicly available.

Scientists have warned that an underwater volcano off the coast of the northwestern US is likely to blow sometime in 2025. The volcano, called Axial Seamount, is more than 3,600-feet-tall and sits half a mile underwater just 300 miles off the coast of Oregon. Experts made the prediction on December 10 after detecting seafloor swelling around Axial that mimicked a level seen immediately before an eruption in 2015. Seismic activity has also increased, with hundreds of earthquakes generated around the volcano per day and earthquake swarms greater than 500 per day. 'Based on the current trends, and the assumption that Axial will be primed to erupt when it reaches the 2015 inflation threshold, our current eruption forecast window is between now (July 2024) and the end of 2025,' researchers said in the new study.

Integrated sensing and communications (ISAC) is pivotal for 6G communications and is boosted by the rapid development of reconfigurable intelligent surfaces (RISs). Using the channel state information (CSI) across multiple frequency bands, RIS-aided multi-band ISAC systems can potentially track users' positions with high precision. Though tracking with CSI is desirable as no communication overheads are incurred, it faces challenges due to the multi-modalities of CSI samples, irregular and asynchronous data traffic, and sparse labeled data for learning the tracking function. This paper proposes the X2Track framework, where we model the tracking function by a hierarchical architecture, jointly utilizing multi-modal CSI indicators across multiple bands, and optimize it in a cross-domain manner, tackling the sparsity of labeled data for the target deployment environment (namely, target domain) by adapting the knowledge learned from another environment (namely, source domain). Under X2Track, we design an efficient deep learning algorithm to minimize tracking errors, based on transformer neural networks and adversarial learning techniques. Simulation results verify that X2Track achieves decimeter-level axial tracking errors even under scarce UL data traffic and strong interference conditions and can adapt to diverse deployment environments with fewer than 5% training data, or equivalently, 5 minutes of UE tracks, being labeled.

Using 3D CNNs on high resolution medical volumes is very computationally demanding, especially for large datasets like the UK Biobank which aims to scan 100,000 subjects. Here we demonstrate that using 2D CNNs on a few 2D projections (representing mean and standard deviation across axial, sagittal and coronal slices) of the 3D volumes leads to reasonable test accuracy when predicting the age from brain volumes. Using our approach, one training epoch with 20,324 subjects takes 20 - 50 seconds using a single GPU, which two orders of magnitude faster compared to a small 3D CNN. These results are important for researchers who do not have access to expensive GPU hardware for 3D CNNs.

Fully automated classification methods that provide direct physical insights into phase diagrams are of current interest. Interpretable, i.e., fully explainable, methods are desired for which we understand why they yield a given phase classification. Ideally, phase classification methods should also be unsupervised. That is, they should not require prior labeling or knowledge of the phases of matter to be characterized. Here, we demonstrate an unsupervised machine-learning method for phase classification, which is rendered interpretable via an analytical derivation of the functional relationship between its optimal predictions and the input data. Based on these findings, we propose and apply an alternative, physically-motivated, data-driven scheme, which relies on the difference between mean input features. This mean-based method does not rely on any predictive model and is thus computationally cheap and directly explainable. As an example, we consider the physically rich ground-state phase diagram of the spinless Falicov-Kimball model.

Human conscious flows can alter brain states. Such brain activities modulate energy consumptions, which can be manifest in the BOLD effect in fMRI experiment. The goal of this study is to identify whether there is difference in such BOLD effects between experienced Tai Chi master in meditation state and normal control subjects. In this experiment, both the meditator and the controls using their conscious to lead a flow periodically circling in their brain in axial, sagittal, and coronal orientations inside a MRI scanner. The experimental results showed significant differences between the meditator and the controls. The most important one is that the meditator activates frontal medial cortex and precuneous regions without any visual excitation, while the controls only utilize visual cortex and precuneous regions without any frontal medial excitation. These seems suggest that for performing the same tasks, the meditator is in cognitive control state, while the controls are in spatial imagination state.