Latest AI and machine learning research in nephrology for healthcare professionals.
Emergent misalignment (EM) -- the broad misbehaviour a language model acquires after fine-tuning on narrow harmful data -- is mediated in Qwen2.5 models by a latent persona direction, and that direction is causal in open weights. Transplanting it into a model that shares only pretraining with its source induces broad EM (2.83 +/- 0.26% misaligned against a random-direction floor of ~1.1%), and abl...
BackgroundGaps in care (GIC) among patients with congenital heart disease (CHD) are associated with adverse outcomes, yet the specific social and healthcare-related factors contributing to GIC and the clinical consequences of delayed re-engagement in care remain poorly characterized. Large electronic medical record datasets often cannot distinguish true GIC from clinically appropriate care pattern...
Abstract Background: Disability prediction in elderly with cardiometabolic multimorbidity (CMM) is limited. We developed a dynamic nomogram and addres...
Background: Standardized evaluation of agentic artificial intelligence (AI) for medication management is lacking. Given the potential lethality of med...
Objective: Predicting health outcomes from electronic health records (EHRs) is challenging because traditional models rely on structured data and ofte...
Background Hypertension remains one of the most challenging healthcare problems in the community. It is a common, measurable, and treatable condition ...
Immune checkpoint inhibitors (ICI) are central to the treatment of metastatic clear cell renal cell carcinoma (ccRCC), yet only a subset of patients d...
BackgroundGraft-versus-host disease (GVHD) remains a major determinant of morbidity and mortality following allogeneic hematopoietic stem cell transpl...
BackgroundInitiation of emergency dialysis, often requiring temporary catheter owing to unprepared definitive vascular access, is associated with infe...
Accurate pulmonary vessel segmentation remains challenging due to the sparse, tortuous, and multi-scale nature of vascular structures, where small bra...
This study introduces a computationally efficient convolutional neural network (CNN) architecture enhanced with transfer learning for multi-cancer det...
Background: Calcium oxalate nephrolithiasis is the most common type of kidney stone disease. Dietary oxalate intake is an important modifiable factor....
Microbial dysbiosis is a hallmark of inflammatory bowel diseases (IBD); however, its drivers and impact on disease pathophysiology are poorly understo...
Kidney-function assessment relies on blood urea as a clinically informative metabolic marker; however, its dependence on venipuncture and centralised ...
Neural network controllers for autonomous decision-making are well-established in cyber-physical systems, yet their deployment in safety-critical heal...
Foundation models (FMs) are central to digital pathology, encoding histology images into dense embeddings for facilitating diagnostic classification, ...
Foreground segmentation is the critical first step of every computational pathology pipeline, yet existing methods rely on hand-tuned heuristics or su...
Multiphasic contrast-enhanced CT (CECT) is widely used for abdominal lesion characterization, yet it carries inherent risks of contrast-induced nephro...
Oppenheim and Lim (1981) showed that natural images stay recognizable when reconstructed from their Fourier phase alone, while the magnitude carries l...
Translating high-dimensional, spatially resolved molecular datasets into testable biological findings remains a major research bottleneck. Here, we pr...