11-19 更新 实时
首页/ 科技公司/ 正文

我们分享更安全、更有用的医疗 AI 构建框架

科技公司 来源:Google Blog 2026-11-04 22:00 0 阅读 机器翻译
我们分享更安全、更有用的医疗 AI 构建框架

莎拉·蒙哥马利-泰勒博士

Google for Health 临床评估负责人

我们中的许多人曾带着健康问题求助于 AI,或者接受过使用 AI 的医生的诊疗。健康领域的技术应用正在加速,每天都影响着数十亿人。但我们如何评估这些工具是否安全、是否有用?

我们与健康专家 Elizabeth McGlynn 博士和 Dana Gelb Safran 博士合作,发布了一份白皮书,详细阐述了一套端到端的评估方法,用于评估 AI 在满足人们健康需求方面的应用场景。该框架从三个方面评估模型:准确性、安全性和有用性。它结合了封闭式问题基准测试、自动评分器和专家人工审核,来评估常见及高风险的医疗健康场景。

我们构建了这一框架,并持续用它来评估我们的模型是否能在健康场景中提供有用、准确的回答,从而促进安全的交互。通过分享我们的方法论,我们希望在处理这一重要领域时提升透明度,并支持整个健康生态——从开发者到研究人员再到临床医生——共同构建人们可以信任的 AI 工具。

译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。

查看英文原文(We're sharing our framework for building safer and more helpful AI for health.)

Dr Sarah Montgomery-Taylor

Clinical Evaluation Lead, Google for Health

Many of us have turned to AI with a health question or received care from a doctor who’s using it. The adoption of technology for health is accelerating and impacts billions of people every day. But how can we assess if these tools are safe and helpful?

In partnership with health experts Dr. Elizabeth McGlynn and Dr. Dana Gelb Safran, we published a whitepaper detailing an end-to-end evaluation approach to assess AI use cases for addressing people’s health needs. The framework evaluates models across three areas: accuracy, safety, and usefulness. It combines closed-question benchmarks, autoraters, and expert human review to assess common and high-risk health scenarios.

We built and continue to utilize this framework to assess whether our models provide useful, accurate responses that promote safe interactions in health contexts. By sharing our methodology, we hope to promote transparency in how we approach this important area, and seek to support the entire health ecosystem — from developers to researchers to clinicians — in building AI tools people can trust.

A healthcare worker in green scrubs leans over to look at a smartphone held by a patient sitting in a clinical office.

正文保留英文原文,标题与摘要由机器翻译。

本文由采集器自动同步

原文发布于 Google Blog(We're sharing our framework for building safer and more helpful AI for health.),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。

阅读原文 ↗