Tag
OpenHealthAtlas is an open-source tool that helps AI agents manage personal health data locally to provide more accurate analysis and reduce hallucinations by optimizing context windows.
PhenoBench is an executable benchmark for evaluating AI models on clinical tasks using a deeply phenotyped human cohort, demonstrating limited aggregate improvements of foundation models over standard methods.
Un piratage dû à un accès phpMyAdmin mal protégé a permis l'extraction de millions de données personnelles, dont des données de santé.
Finch is a health data platform that provides a single place for managing personal health data and test results.
Introduces Apple Health MCP Server, an open-source Model Context Protocol server that enables SQL querying and DuckDB-based analysis of exported Apple Health data, featuring natural-language queries, health reports, and lazy loading.
This paper presents an AI-driven multimodal mediation framework using variational autoencoders to discover latent pathways linking socioeconomic disadvantage, psychosocial factors, and cardiometabolic multimorbidity in the All of Us Research Program cohort.
This paper evaluates unsupervised subgrouping methods combined with causal discovery and policy evaluation for budget-constrained health interventions using observational data, finding no single method consistently outperforms others in held-out evaluation.
A WIRED security roundup reveals that period tracker apps, notably Stardust, share intimate health data with third-party analytics firms and Facebook, raising serious privacy concerns.
A BBC article highlights a new Mozilla Foundation report revealing that some popular period tracking apps share sensitive health data with third parties like Google, Meta, and TikTok, while others have strong privacy protections. It provides guidance on how users can assess privacy risks in such apps.
A developer shares their open-source HealthClaw Guardrails project that enforces safety guardrails for LLM agents accessing real health records, including PHI redaction, audit logging, and human-in-the-loop confirmation, with a conformance endpoint for testing.
The tweet shares a personal conversion to wearables with AI, emphasizing the potential of AI agents acting on body/activity context. Anjney Midha envisions a future where a team of AIs monitors health data in real time.
Lawmakers have proposed a new version of the Health and Location Data Protection Act that would ban the sale of Americans' health and location data, including data shared with AI chatbots like ChatGPT and Claude, with stronger enforcement measures.
A user shares their positive experience using openclaw to aggregate messy health data from three sources (fitness watch, text messages, handwritten notes) for a cardiology appointment, including resolving scheduling conflicts, merging data, and ensuring privacy.
freddy is a CLI tool that acts as a personal MCP server for health data, allowing AI agents to query wearables like Garmin. It supports OAuth device flow and can pull sleep, HRV, and other metrics from multiple sources.
A user details the challenges of using an AI assistant to convert 6 months of Apple Watch sleep data into a sleep clinic's diary format, including timezone conversions, date offsets, and fabricated values. The post shares lessons on correctly interpreting health data sources for medical forms.
Freddy CLI is an open-source tool that lets users connect their wearable health data to AI agents via MCP, supporting devices like Oura, Polar, and Garmin.
A personal health MCP server that gives AI agents like OpenClaw access to real health and performance data from wearables (Oura, Garmin, WHOOP, etc.) via a unified interface and OAuth device-code flow.
Amnesty reports that Palantir and other contractors were given unlimited access to identifiable NHS England patient information, raising privacy concerns.
Oura, the health wearable maker, acknowledges receiving government requests for user data but refuses to disclose how many requests it receives or how often it complies. The article highlights Oura's lack of end-to-end encryption and the potential for government access to sensitive health data.
Fitbit Air launched with a new Google Health API that allows developers to build AI agents and services on top of 31 health data points including sleep, heart rate, and SpO2, with webhooks and granular permissions.