healthcare

NOAH introduces a generative transformer model for comprehensive representation and forecasting of longitudinal multimodal patient data, enabling tasks like zero-shot classification and counterfactual simulation in clinical settings.

Allia is releasing a new Partner API that allows third-party products to integrate with Allia’s provider network to offer care services directly to users.

Personal Therapist is an AI agent that provides continuous mental health support by listening without agenda, remembering context across conversations, and offering honest feedback, available anytime without appointments.

A physician argues that unified multimodal AI agents will soon autonomously outperform doctors in clinical encounters, rendering physicians obsolete as middlemen between data collection and analysis, yet welcomes this as a net good for patient care.

This paper introduces ReTA, a reinforcement learning-based framework for dynamically augmenting electronic health record graphs with external knowledge graphs to improve prediction tasks like diagnosis and mortality.

This paper introduces contrastive explanations for Quantitative Bipolar Argumentation Frameworks, explaining differences between two topic arguments to enhance AI explainability, with applications in healthcare and bias identification.

This paper proposes AI Morbidity and Mortality (AI M&M), a blameless framework for case-based review of clinical AI failures, aiming to convert individual errors into actionable institutional learning.

This paper proposes MiNER, a fine-tuned BioBERT model for extracting biomedical entities from malaria-related clinical texts, and releases a human-labeled dataset for future research.

This paper compares Arabic and English large language models for assessing suicide risk from crisis helpline transcripts, showing that both can effectively identify high-risk cases without compromising privacy by keeping audio local.

This paper presents a preliminary study evaluating the accuracy of large language models in verifying causal medical hypotheses, finding that while they exhibit strong recall, they often fail to provide valid scientific evidence or reject unsupported claims.

a16z has raised an $8.5 billion Growth Fund to invest in six super trends including Enterprise AI, Consumer AI, and Healthcare, focusing on transformative technology cycles.

Arintra Health has raised $25M in Series B funding to develop AI that automates medical billing coding, improving revenue capture and efficiency for hospitals.

CARE introduces a causally-aligned reasoning exploration framework for medical large language models to enhance training stability and reduce spurious correlations.

The article questions AI agent builders on methods for tracking competitor features, deciding product roadmaps, and understanding buyer preferences in competitive verticals.

China's NMPA has begun reviewing Novo Nordisk's marketing application for Rybelsus tablet, marking a key regulatory milestone for the drug in the Chinese market.

HealthBench-Psych introduces a mental health subset of OpenAI's HealthBench benchmark to evaluate LLMs in mental health conversations, validated through clinician review and assessing 20 models as a reusable resource.

An AI skill named /fuck-cancer is being open-sourced to help cancer patients and caregivers navigate diagnosis and treatment by creating a practical medical brief.

The paper proposes a multimodal prompt-learning framework to handle missing modalities in electronic health records for robust clinical prediction in intensive care units, introducing four prompt types to capture dependencies and interactions.

The UN utilizes AI tools across agencies to promote peace, security, and human rights, with examples including UNICEF's accessible digital textbooks, UNEP's methane leak detection system, and WHO's AI-assisted tuberculosis screening.

The article introduces llava-medical-8B-clip-vit-stage2, a specialized vision-language model fine-tuned for healthcare, enabling AI to interpret medical images like X-rays and MRIs and answer related questions.