Complex basilar apex aneurysms (CBAAs) present a significant challenge due to their unfavorable natural history and difficulty with conventional treatments. This study aimed to provide insights into flow alteration strategies by combining a systematic review using PRISMA methodology with a multicenter experience from South Korea. We analyzed 57 cases, finding that flow preservation with aneurysm obliteration was performed in 12.7%, while mild, moderate, and maximum flow reduction were applied in 77.2%, 7.0%, and 3.5% respectively. Outcomes showed that 75.8% of patients with available imaging achieved satisfactory aneurysm obliteration. A good clinical outcome (mRS 0–2) was observed in 49.1% of cases. However, poor outcomes (mRS 4–6) were reported in 31.6%, with a mortality rate of 17.5%. Beyond simply reducing intra-aneurysmal flow, computational fluid dynamics (CFD) simulations revealed that alterations in flow balance and direction significantly influenced hemodynamic stress. Given the severe prognosis of CBAAs, flow alteration strategies can serve as viable alternatives when conventional treatments are not feasible. Furthermore, CFD simulations might hold promise in identifying optimal strategies that can maximize aneurysm control while minimizing procedural risks.

The activation of integrins by Mn²⁺ is a crucial area of research, yet the underlying mechanisms remain poorly understood. Previous studies have shown that substituting Mg²⁺ with Mn²⁺ at the metal ion-dependent adhesion site (MIDAS) enhances the affinities of high-affinity open and low-affinity closed integrins. However, the molecular effect of Mn²⁺ and how it compares to physiological activation mediated by Mg²⁺/Ca²⁺ remain unclear. This is partly due to the lack of experimental techniques capable of detecting these processes dynamically. In this study, we used equilibrium molecular dynamics simulations to examine the effects of Mn²⁺ on the binding site of platelet integrin α_IIbβ₃. Our findings show that Mn²⁺ accelerates conformational changes related to activation. Specifically, Mn²⁺ promotes an earlier displacement of M335 in the β6-α7 loop away from the ADMIDAS site (adjacent to the MIDAS site) and a rapid downward movement of the α7 helix in the βI domain. Additionally, Mn²⁺ leads to faster stabilization of the α1 helix, strengthening the interactions between the α_IIbβ₃ ligand-binding site and the RGD motif. These results suggest that Mn²⁺ accelerates high-affinity rearrangements at the ligand-binding site, resembling those seen in physiological activation, but occurring more rapidly than with Mg²⁺/Ca²⁺. Overall, our data suggest that Mn²⁺-induced affinity modulation proceeds through similar early activation steps, even without full integrin extension.

The think-aloud (TA) protocol is a useful method for evaluating user interfaces, including data visualizations. However, TA studies are time-consuming to conduct and hence often have a small number of participants. Crowdsourcing TA studies would help alleviate these problems, but the technical overhead and the unknown quality of results have restricted TA to synchronous studies.

To address this gap we introduce CrowdAloud, a system for creating and analyzing asynchronous, crowdsourced TA studies. CrowdAloud captures audio and provenance (log) data as participants interact with a stimulus. Participant audio is automatically transcribed and visualized together with events data and a full recreation of the state of the stimulus as seen by participants.

To gauge the value of crowdsourced TA studies, we conducted two experiments: one to compare lab-based and crowdsourced TA studies, and one to compare crowdsourced TA studies with crowdsourced text prompts. Our results suggest that crowdsourcing is a viable approach for conducting TA studies at scale.

Large vision-language contrastive models (VLCMs), such as CLIP, have become foundational, demonstrating remarkable success across a variety of downstream tasks. Despite their advantages, these models, akin to other foundational systems, inherit biases from the disproportionate distribution of real-world data, leading to misconceptions about the actual environment. Prevalent datasets like ImageNet are often riddled with non-causal, spurious correlations that can diminish VLCM performance in scenarios where these contextual elements are absent. This study presents an investigation into how a simple linear probe can effectively distill task-specific core features from CLIP’s embedding for downstream applications. Our analysis reveals that the CLIP text representations are often tainted by spurious correlations, inherited in the biased pre-training dataset. Empirical evidence suggests that relying on visual representations from CLIP, as opposed to text embedding, is more effective to refine the skewed perceptions in VLCMs, emphasizing the superior utility of visual representations in overcoming embedded biases. Our code can be found here.

Multi-fidelity methods that use an ensemble of models to compute a Monte Carlo estimator of the expectation of a high-fidelity model can significantly reduce computational costs compared to single-model approaches. These methods use oracle statistics, specifically the covariance between models, to optimally allocate samples to each model in the ensemble. However, in practice, the oracle statistics are estimated using additional model evaluations, whose computational cost and induced error are typically ignored. To address this issue, this paper proposes an adaptive algorithm to optimally balance the resources between oracle statistics estimation and final multi-fidelity estimator construction, leveraging ideas from multilevel best linear unbiased estimators in Schaden and Ullmann (2020) and a bandit-learning procedure in Xu et al. (2022). Under mild assumptions, we demonstrate that the multi-fidelity estimator produced by the proposed algorithm exhibits mean-squared error commensurate with that of the best linear unbiased estimator under the optimal allocation computed with oracle statistics. Our theoretical findings are supported by detailed numerical experiments, including a parametric elliptic PDE and an ice-sheet mass-change modeling problem. 

The challenge of being able to scale application codes based on the Asynchronous Many-Task (AMT) Uintah framework on the Department of Energy (DOE) Aurora exascale system is addressed in this work by considering a challenging Reverse Monte Carlo Ray Tracing radiation benchmark calculation. This benchmark involves potentially global all-to-all communication and uses adaptive mesh refinement and ray tracing to achieve scalability. This benchmark has been used as part of previous scalability studies on a number of pre-exascale systems and on the DOE Frontier exascale system. This paper describes steps taken to enable this benchmark to run successfully on up to 10,240 nodes and 122,880 Intel® Ponte Vecchio Xe stacks on the DOE Aurora exascale system. This scalability was achieved through a limited number of experiments on Aurora, given machine loads and its uniqueness. These experiments constitute valuable lessons learned to achieve scalability at this level. The resulting scalability runs, while few in number, demonstrate relatively good strong-scaling characteristics. A detailed analysis of these results provides important indications about the path to scalability on Aurora for future work. Overall, these results continue the remarkable ability of this AMT approach to produce scalable solutions for challenging problems at extreme scale on heterogeneous architectures.

Fatigue-induced crack growth is a leading cause of structural failure across critical industries such as aerospace, civil engineering, automotive, and energy. Accurate prediction of stress intensity factors (SIFs) -- the key parameters governing crack propagation in linear elastic fracture mechanics -- is essential for assessing fatigue life and ensuring structural integrity. While machine learning (ML) has shown great promise in SIF prediction, its advancement has been severely limited by the lack of rich, transparent, well-organized, and high-quality datasets.
To address this gap, we introduce SIFBench, an open-source, large-scale benchmark database designed to support ML-based SIF prediction. SIFBench contains over 5 million different crack and component geometries derived from high-fidelity finite element simulations across 37 distinct scenarios, and provides a unified Python interface for seamless data access and customization. We report baseline results using a range of popular ML models -- including random forests, support vector machines, feedforward neural networks, and Fourier neural operators -- alongside comprehensive evaluation metrics and template code for model training, validation, and assessment. By offering a standardized and scalable resource, SIFBench substantially lowers the entry barrier and fosters the development and application of ML methods in damage tolerance design and predictive maintenance. 

Chest diseases rank among the most prevalent and dangerous global health issues. Object detection and phrase grounding deep learning models interpret complex radiology data to assist healthcare professionals in diagnosis. Object detection locates abnormalities for classes, while phrase grounding locates abnormalities for textual descriptions. This paper investigates how text enhances abnormality localization in chest X-rays by comparing the performance and explainability of these two tasks. To establish an explainability benchmark, we proposed an automatic pipeline to generate image regions for report sentences using radiologists’ eye-tracking data. The better performance - mIoU = 36% vs. 20% - and explainability - Containment ratio 48% vs. 26% - of the phrase grounding model infers the effectiveness of text in enhancing chest X-ray abnormality localization.

Topological descriptors such as contour trees are widely utilized in scientific data analysis and visualization, with applications from materials science to climate simulations. It is desirable to preserve topological descriptors when data compression is part of the scientific workflow for these applications. However, classic error-bounded lossy compressors for volumetric data do not guarantee the preservation of topological descriptors, despite imposing strict pointwise error bounds. In this work, we introduce a general framework for augmenting any lossy compressor to preserve the topology of the data during compression. Specifically, our framework quantifies the adjustments (to the decompressed data) needed to preserve the contour tree and then employs a custom variable-precision encoding scheme to store these adjustments. We demonstrate the utility of our framework in augmenting classic compressors (such as SZ3, TTHRESH, and ZFP) and deep learning-based compressors (such as Neurcomp) with topological guarantees.

Hemozoin crystal formation is a major antimalarial drug target as it is essential for survival of Plasmodium parasites that cause malaria, one of the world’s most devastating infectious diseases. Hemozoin crystals rapidly tumble inside the parasite food vacuole, and the mechanism and significance of this motion have been mysterious. Using quantitative live-cell imaging and computational modeling, we found that hemozoin motion is driven by catalytic decomposition of hydrogen peroxide on the crystal surface, a mechanism analogous to synthetic nanomotors. Hemozoin crystals have been viewed as inert detoxification products. Our work reframes the physiological role of these crystals as catalytically active nanoparticles whose mechanism of propulsion helps to neutralize toxic hydrogen peroxide generated by parasite digestion of hemoglobin during blood-stage infection. Malaria parasites infect red blood cells where they digest host hemoglobin and release free heme inside a lysosome-like organelle called the food vacuole. To detoxify excess heme, parasites form hemozoin crystals that rapidly tumble inside this compartment. Hemozoin formation is critical for parasite survival and central to antimalarial drug activity. Although the static structural properties of hemozoin have been extensively investigated, crystal motion and its underlying mechanism have remained puzzling. We used quantitative image analysis to determine the timescale of motion, which requires the intact vacuole but does not require the parasite itself. Using single particle tracking and Brownian dynamics simulations with experimentally derived interaction potentials, we found that hemozoin motion exhibits unexpectedly tight confinement but is much faster than thermal diffusion. Hydrogen peroxide, which is generated at high levels in the food vacuole, has been shown to stimulate the motion of synthetic metallic nanoparticles via surface-catalyzed peroxide decomposition that generates propulsive kinetic energy. We observed that peroxide stimulated the motion of isolated crystals in solution and that conditions that suppress peroxide formation slowed hemozoin motion inside parasites. These data suggest that surface-exposed metals on hemozoin catalyze peroxide decomposition to drive crystal motion. This work reveals hemozoin motion in malaria parasites as a biological example of an endogenous self-propelled nanoparticle. This mechanism of propulsion likely serves a physiological role to reduce oxidative stress to parasites from hydrogen peroxide produced by large-scale hemoglobin digestion during blood-stage infection.

We investigate whether tactile charts support comprehension and learning of complex visualizations for blind and low-vision (BLV) individuals and contribute four tactile chart designs and an interview study. Visualizations are powerful tools for conveying data, yet BLV individuals typically can rely only on assistive technologies -- primarily alternative texts -- to access this information. Prior research shows the importance of mental models of chart types for interpreting these descriptions, yet BLV individuals have no means to build such a mental model based on images of visualizations. Tactile charts show promise to fill this gap in supporting the process of building mental models. Yet studies on tactile data representations mostly focus on simple chart types, and it is unclear whether they are also appropriate for more complex charts as would be found in scientific publications. Working with two BLV researchers, we designed 3D-printed tactile template charts with exploration instructions for four advanced chart types: UpSet plots, violin plots, clustered heatmaps, and faceted line charts. We then conducted an interview study with 12 BLV participants comparing whether using our tactile templates improves mental models and understanding of charts and whether this understanding translates to novel datasets experienced through alt texts. Thematic analysis shows that tactile models support chart type understanding and are the preferred learning method by BLV individuals. We also report participants' opinions on tactile chart design and their role in BLV education. 

A key challenge faced when preparing codes for Department of Energy (DOE) exascale systems was designing scalable applications for systems featuring hardware and software not yet available at leadership class scale. With such systems now available, it is important to evaluate scalability of the resulting software solutions on these target systems. One such code designed with the exascale DOE Aurora and DOE Frontier systems in mind is the Uintah Computational Framework, an open-source asynchronous many-task (AMT) runtime system. To prepare for exascale, Uintah adopted a portable MPI+X hybrid parallelism approach using the Kokkos performance portability library (i.e., MPI+Kokkos). This paper complements recent work with additional details and an evaluation of the resulting approach on Aurora and Frontier. Results are shown for a challenging benchmark demonstrating interoperability of 3 portable codes essential to Uintah-related combustion research. These results demonstrate single-source portability across Aurora and Frontier with scaling characteristics shown to 3,072 Aurora nodes and 9,216 Frontier nodes. In addition to showing results run to new scales on new systems, this paper also discusses lessons learned through efforts preparing Uintah for exascale systems.

The increasing demand for larger and higher fidelity simulations has made Adaptive Mesh Refinement (AMR) and unstructured mesh techniques essential to focus compute effort and memory cost on just the areas of interest in the simulation domain. The distribution of these meshes over the compute nodes is often determined by balancing compute, memory, and network costs, leading to distributions with jagged nonconvex boundaries that fit together much like puzzle pieces. It is expensive, and sometimes impossible, to re-partition the data posing a challenge for in situ and post hoc visualization as the data cannot be rendered using standard sort-last compositing techniques that require a convex and disjoint data partitioning. We present a new distributed volume rendering and compositing algorithm, Approximate Puzzlepiece Compositing, that enables fast and high-accuracy in-place rendering of AMR and unstructured meshes. Our approach builds on Moment-Based Ordered-Independent Transparency to achieve a scalable, order-independent compositing algorithm that requires little communication and does not impose requirements on the data partitioning. We evaluate the image quality and scalability of our approach on synthetic data and two large-scale unstructured meshes on HPC systems by comparing to state-of-the-art sort-last compositing techniques, highlighting our approach’s minimal overhead at higher core counts. We demonstrate that Approximate Puzzlepiece Compositing provides a scalable, high-performance, and high-quality distributed rendering approach applicable to the complex data distributions encountered in large-scale CFD simulations.

In applications where efficiency is critical, developers may examine their compiled binaries, seeking to understand how the compiler transformed their source code and what performance implications that transformation may have. This analysis is challenging due to the vast number of disassembled binary instructions and the many-to-many mappings between them and the source code. These problems are exacerbated as source code size increases, giving the compiler more freedom to map and disperse binary instructions across the disassembly space. Interfaces for disassembly typically display instructions as an unstructured listing or sacrifice the order of execution. We design a new visual interface for disassembly code that combines execution order with control flow structure, enabling analysts to both trace through code and identify familiar aspects of the computation. Central to our approach is a novel layout of instructions grouped into basic blocks that displays a looping structure in an intuitive way. We add to this disassembly representation a unique block-based mini-map that leverages our layout and shows context across thousands of disassembly instructions. Finally, we embed our disassembly visualization in a web-based tool, DisViz, which adds dynamic linking with source code across the entire application. DizViz was developed in collaboration with program analysis experts following design study methodology and was validated through evaluation sessions with ten participants from four institutions. Participants successfully completed the evaluation tasks, hypothesized about compiler optimizations, and noted the utility of our new disassembly view. Our evaluation suggests that our new integrated view helps application developers in understanding and navigating disassembly code. 

Adverse drug events represent a key challenge in public health, especially concerning drug safety profiling and drug surveillance. Drug-drug interactions represent one of the most popular types of adverse drug events. Most computational approaches to this problem have used different types of drug-related information utilizing different machine-learning algorithms to predict potential drug interactions. In this work, we focus on genetic information about the drugs, particularly the protein sequence and protein structure of protein targets in drug interaction networks, to predict potential drug interactions. We collected various drug information like drug-drug interaction (DDI), Drug attributes like drug active ingredients, protein targets, protein sequence, protein structure etc. We proposed a similarity-based Neural Network framework called protein sequence-structure similarity network (PS3N) and used this to predict novel DDI’s. The drug-drug similarities are computed using different categories of drug information based on multiple similarity metrics. Our method outperforms the state-of-the-art and achieves competitive results. Our performance evaluations on different datasets showed the predictive performance as follows: Precision 91%–98%, Recall 90%–96%, F1 Score 86%–95%, Area Under Curve (AUC) 88%–99%, and Accuracy 86%–95%. Our evaluation demonstrates the effectiveness of PS3N in predicting DDI’s, including the clinical significance of some new DDI’s discovered by the model.

Anatomy evaluation is crucial for understanding the physiological state, diagnosing abnormalities, and guiding medical interventions. Statistical shape modeling (SSM) is vital in this process, particularly in medical image analysis and computational anatomy. By enabling the extraction of quantitative morphological shape descriptors from medical imaging data such as MRI and CT scans, SSM provides comprehensive descriptions of anatomical variations within a population. However, the effectiveness of SSM in anatomy evaluation hinges on the quality and robustness of the shape models, which face challenges due to substantial nonlinear variability in human anatomy. While deep learning techniques show promise in addressing these challenges by learning complex nonlinear representations of shapes, existing models still have limitations and often require pre-established shape models for training. To overcome these issues, we propose Mesh2SSM++, a novel approach that learns to estimate correspondences from meshes in an unsupervised manner. This method leverages unsupervised, permutation-invariant representation learning to estimate how to deform a template point cloud into subject-specific meshes, forming a correspondence-based shape model. Additionally, our probabilistic formulation allows learning a population-specific template, reducing potential biases associated with template selection. A key feature of Mesh2SSM++ is its ability to quantify aleatoric uncertainty, which captures inherent data variability and is essential for ensuring reliable model predictions and robust decision-making in clinical tasks, especially under challenging imaging conditions. Through extensive validation across diverse anatomies, evaluation metrics, and downstream tasks, we demonstrate that Mesh2SSM++ outperforms existing methods. Its ability to operate directly on meshes, combined with computational efficiency and interpretability through its probabilistic framework, makes it an attractive alternative to traditional and deep learning-based SSM approaches. Github: https://github.com/iyerkrithika21/Mesh2SSMJournal

High-performance computing (HPC) systems are becoming increasingly water-intensive due to their reliance on water-based cooling and the energy used in power generation. However, the water footprint of HPC remains relatively underexplored-especially in contrast to the growing focus on carbon emissions. In this paper, we present ThirstyFLOPS - a comprehensive water footprint analysis framework for HPC systems. Our approach incorporates region-specific metrics, including Water Usage Effectiveness, Power Usage Effectiveness, and Energy Water Factor, to quantify water consumption using real-world data. Using four representative HPC systems - Marconi, Fugaku, Polaris, and Frontier - as examples, we provide implications for HPC system planning and management. We explore the impact of regional water scarcity and nuclear-based energy strategies on HPC sustainability. Our findings aim to advance the development of water-aware, environmentally responsible computing infrastructures.