Kezhi Li

Automated chest X-ray disease screening using large language models and deep convolutional neural networks on the MIMIC-CXR dataset

Qingyu Zhang, Pardeep Vasudev, Kezhi Li, Holger Kunz

Introduction Radiologists in resource-limited settings often face high workloads, especially in chest X-ray interpretation. Manual annotation of large-scale imaging datasets remains costly and time-consuming. This study aims to explore the feasibility of using large language models (LLMs), specifically GPT-4o, to generate binary …

gb (code pays fourni par la source)

Cognitive Overcapacity: How Economies Absorban Abundance of Machine Intelligence

Kezhi Li

This concept note introduces cognitive overcapacity, or more informally a cognitive glut, as a potential economic consequence of rapid advances in artificial intelligence. Cognitive overcapacity arises when the effective capacity to produce cognitive services grows faster than final paid demand for those …

gb (code pays fourni par la source)

Cognitive Overcapacity: How Economies Absorban Abundance of Machine Intelligence

Kezhi Li

This concept note introduces cognitive overcapacity, or more informally a cognitive glut, as a potential economic consequence of rapid advances in artificial intelligence. Cognitive overcapacity arises when the effective capacity to produce cognitive services grows faster than final paid demand for those …

gb (code pays fourni par la source)

A Scoping Review of Bias in Large AI Models for Medicine and Healthcare

Showing; id_orcid 0000-0002-8624-5740 Xiao, Zhenpeng Chen, Jen-tse Huang, Wenting Chen et autres

cn, us, gb, ch (code pays fourni par la source)

Retrieval-Augmented Large Language Model Counseling for Continuous Glucose Monitoring in Diabetes: Source-Masked Multirater Comparative Evaluation

Zhijun Guo, Alvina Grace Lai, Emmanouil Korakas, Aristeidis Vagenas et autres

Background: Continuous glucose monitoring (CGM) is central to diabetes care, but explaining CGM patterns consistently and empathetically remains time-intensive in clinical practice. Large language model (LLM)-based systems may support patient-facing interpretation of CGM data, but evidence remains limited for retrieval-grounded tools evaluated …

gb (code pays fourni par la source)

Feasibility and user evaluation of HopeBot: An LLM-powered conversational chatbot for depression screening

Zhijun Guo, Alvina Lai, Julia Ive, Alexandru Petcu et autres

Clinician-administered Patient Health Questionnaire-9 (PHQ-9) interviews allow clarification of ambiguous responses but are resource-intensive and difficult to scale for routine use. Self-administered versions are widely adopted for depression screening yet offer little opportunity for interaction or clarification, which may limit engagement and …

gb (code pays fourni par la source)

An Assessment of Trajectory Analysis for Disease Risk Prediction: A Scoping Review

Freya Pollington, Spiros Denaxas, Kezhi Li, Johan H. Thygesen et autres

Structured electronic health records (EHRs) are increasingly being used to predict risk of disease by analysing 'patient trajectories'. We conducted a PubMed and Web of Science search for studies using temporal EHR sequences to identify disease signatures or predict disease. This scoping …

gb (code pays fourni par la source)

Non-invasive glucose monitoring vs iCGM: a systematic review and meta-analysis of accuracy and methodological challenges

Hengrui Zhang, Ellen Dromard, Kevin C.H. Tsang, Amparo Güemes et autres

Abstract Non-invasive glucose monitoring (NIGM) has been pursued for decades, yet no device has achieved regulatory approval despite numerous studies reporting high accuracy. This systematic review and meta-analysis of 32 studies (38 cohorts: 20 NIGM, 18 iCGM; N = 1,693) investigated methodological …

gb (code pays fourni par la source)

Blinded Multi-Rater Comparative Evaluation of a Large Language Model and Clinician-Authored Responses in CGM-Informed Diabetes Counseling

Zhijun Guo, Alvina Grace Lai, Emmanouil Korakas, Aristeidis Vagenas et autres

Continuous glucose monitoring (CGM) is central to diabetes care, but explaining CGM patterns clearly and empathetically remains time-intensive. Evidence for retrieval-grounded large language model (LLM) systems in CGM-informed counseling remains limited. To evaluate whether a retrieval-grounded LLM-based conversational agent (CA) could support …

Blinded Multi-Rater Comparative Evaluation of a Large Language Model and Clinician-Authored Responses in CGM-Informed Diabetes Counseling (Preprint)

Zhijun Guo, Alvina Grace Lai, Emmanouil Korakas, Aristeidis Vagenas et autres

BACKGROUND Continuous glucose monitoring (CGM) is central to modern diabetes care, but explaining CGM patterns clearly, consistently, and empathetically remains time-intensive in practice. Large language model (LLM)–based systems may support patient-facing interpretation of CGM data, but evidence remains limited for retrieval-grounded tools …

AC-Refiner: Efficient Arithmetic Circuit Optimization Using Conditional Diffusion Models

Chenhao Xue, Kezhi Li, Yi Ren, Zhengyuan Shi et autres

cn, hk (code pays fourni par la source)

Development of a machine learning model to predict intensive care unit bed demand for adult elective surgical patients at a large United Kingdom National Health Service Trust

Jennifer M. Hunter, Hrisheekesh Jayant Vaidya, Sonya Crowe, Martin Utley et autres

Background Elective surgical admissions form a growing share of demand for ICU beds, a constrained resource. Capacity planning for these admissions is feasible, but hospitals often lack reliable systems estimating daily elective surgical ICU bed demand before the day of surgery. Comprehensive …

gb (code pays fourni par la source)