A study has found that undiagnosed heart failure can be detected in people with high-risk diabetes using a simple, GP-led care pathway.
An in-home engagement and usability study of GeRI: an open-source platform for remote symptom assessment and wearable activity monitoring in men with prostate cancer
Geriatric assessment (GA) is underused in oncology because clinic-based implementation is time- and resource-intensive, limiting routine evaluation of frailty and treatment tolerance. Existing digital tools often rely on proprietary devices and closed analytic pipelines. We developed the Geriatric Remote Initiative (GeRI), an open-source platform integrating a wrist-worn accelerometer, smart scale, and tablet interface with reproducible […]
Amgen’s BiTE for lung cancer backed for EU approval
Amgen’s fast-growing bispecific T-cell engager Imdylltra should be cleared as a second-line therapy for small cell lung cancer, said the CHMP.
Inside the stealthy startup that pitched brainless human clones
After operating in secrecy for years, a startup company called R3 Bio, in Richmond, California, suddenly shared details about its work last week—saying it had raised money to create nonsentient monkey “organ sacks” as an alternative to animal testing. In an interview with Wired, R3 listed three investors: billionaire Tim Draper, the Singapore-based fund Immortal […]
“Reimaging a triage system with midwives, for midwives”: exploring preferences for a midwife-Led triage system in South Africa through a user-centered approach
IntroductionTriage in the maternity unit is critical to ensuring the delivery of timely and appropriate care. It is regarded as an initiative to reduce maternal mortality by accelerating the provision of appropriate care at the appropriate time. However, maternity units in South Africa lack standardized triage systems. Most pregnant women often wait for hours and […]
Technology-driven diabetes care: innovation without equity?
Post Content
Extraction and processing of intensive care chart data from a patient data management system
BackgroundRoutine clinical data captured in Patient Data Management Systems (PDMS) in intensive care and perioperative settings are an invaluable resource for clinical research. However, the proprietary, fragmented, and transaction-oriented architecture of many systems severely limits secondary data use and requires extensive Extract, Transform, and Load (ETL) processing.MethodsWe developed a modular, Python-based ETL framework that enables […]
When Perplexity Lies: Generation-Focused Distillation of Hybrid Sequence Models
arXiv:2603.26556v1 Announce Type: cross Abstract: Converting a pretrained Transformer into a more efficient hybrid model through distillation offers a promising approach to reducing inference costs. However, achieving high-quality generation in distilled models requires careful joint design of both the student architecture and the distillation process. Many prior distillation works evaluate downstream multiple-choice benchmarks by ranking […]
Humanline: Online Alignment as Perceptual Loss
arXiv:2509.24207v2 Announce Type: replace Abstract: Online alignment (e.g., GRPO) is generally more performant than offline alignment (e.g., DPO) — but why? Drawing on prospect theory from behavioral economics, we propose a human-centric explanation. We prove that online on-policy sampling better approximates the human-perceived distribution of what the model can produce, and PPO/GRPO-style clipping — originally […]
QHap: Quantum-Inspired Haplotype Phasing
arXiv:2603.25762v1 Announce Type: new Abstract: Haplotype phasing, the process of resolving parental allele inheritance patterns in diploid genomes, is critical for precision medicine and population genetics, yet the underlying optimization is NP-hard, posing a scalability challenge. To address this, we introduce QHap, a haplotype phasing tool that leverages quantum-inspired optimization. By reformulating haplotype phasing as […]
A Human-Inspired Decoupled Architecture for Efficient Audio Representation Learning
arXiv:2603.26098v1 Announce Type: cross Abstract: While self-supervised learning (SSL) has revolutionized audio representation, the excessive parameterization and quadratic computational cost of standard Transformers limit their deployment on resource-constrained devices. To address this bottleneck, we propose HEAR (Human-inspired Efficient Audio Representation), a novel decoupled architecture. Inspired by the human cognitive ability to isolate local acoustic features […]
DUET-VLM: Dual stage Unified Efficient Token reduction for VLM Training and Inference
arXiv:2602.18846v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) have achieved remarkable multimodal understanding and reasoning capabilities, yet remain computationally expensive due to dense visual tokenization. Existing efficiency approaches either merge redundant visual tokens or drop them progressively in language backbone, often trading accuracy for speed. In this work, we propose DUET-VLM, a versatile plug-and-play dual […]