Large language models (LLMs) have performed well across various clinical natural language processing tasks, despite not being directly trained on electronic health record (EHR) data. In this work, we examine how popular open-source LLMs learn clinical information from large mined corpora through two crucial but understudied lenses: (1) their interpretation of clinical jargon, a foundational…
OBJECTIVES: To develop the first core Critical Care Data Dictionary (C2D2) with common data elements (CDEs) to characterize critical illness and injuries. DESIGN: Group consensus process using modified Delphi approach. SETTING: Electronic surveys and in-person meetings. SUBJECTS: A multidisciplinary workgroup of clinicians and researchers with expertise in the care of the critically ill and injured….
Diffuse correlation spectroscopy (DCS) allows label-free, non-invasive investigation of microvascular dynamics deep within tissue, such as cerebral blood flow (CBF). However, the signal-to-noise ratio (SNR) in DCS limits its effective cerebral sensitivity in adults, in which the depth to the brain, through the scalp and skull, is substantially larger than in infants.Therefore, we aim to…
Technology research is the driving force of the innovations that shape the world. Sony Group Corporation (Sony) and Nature partnered together to launch the Sony Women in Technology Award to recognize three outstanding early to mid-career researchers from the field of technology. They interviewed the winners of the inaugural 2024 award on the inspirations behind their outstanding…
Monitoring disease development in arteries, which supply oxygen and nutrients to the body, is crucial and can be assessed using hemodynamic metrics. Hemodynamic metrics can be calculated via computational fluid dynamic simulation of patient-specific geometries. These simulations are known to be heavily influenced by boundary conditions, such as time-dependent inlet flow. However, the effects of…
Virtual planning tools that provide intuitive user interaction and immediate hemodynamic feedback are crucial for cardiologists to effectively treat coronary artery disease. Current FDA-approved tools for coronary intervention planning require days of preliminary processing and rely on conventional 2D displays for hemodynamic evaluation. Immersion offered by extended reality (XR) has been found to benefit intervention…
Parallel agent-based models of the adaptive immune response can efficiently recapitulate emerging spatiotemporal properties of T-cell motility during clonal selection across multiple length and time scales. Here, we present a distributed, three-dimensional (3D) computational model of T-cell priming, and associated parallel data structures and algorithms that enable fully deterministic cell simulations at scale. We demonstrate…
Over the past four decades, high-performance computing (HPC) has enabled considerable advances in scientific discovery and engineering, spurring technological development across the globe. However, with the demand for precision and fidelity of computational models continuing to grow, HPC faces bottlenecks in data handling, algorithm efficiency, and the scalability of new architectures, especially in fields such…
Fusion oncoproteins, a class of chimeric proteins arising from chromosomal translocations, are major drivers of various pediatric cancers. These proteins are intrinsically disordered and lack druggable pockets, making them highly challenging therapeutic targets for both small molecule-based and structure-based approaches. Protein language models (pLMs) have recently emerged as powerful tools for capturing physicochemical and functional…
Designing binders to target undruggable proteins presents a formidable challenge in drug discovery. In this work, we provide an algorithmic framework to design short, target-binding linear peptides, requiring only the amino acid sequence of the target protein. To do this, we propose a process to generate naturalistic peptide candidates through Gaussian perturbation of the peptidic…