相关实验视频

Updated: Jun 19, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

推理语言模型可以更透明地预测自杀风险

Thomas H McCoy^1,2, Roy H Perlis^3,2

¹Center for Quantitative Health, Massachusetts General Hospital, Boston, Massachusetts, USA.

BMJ mental health

|May 11, 2025

概括

较小的AI模型可以使用医院数据估计自杀风险,准确度仅略有下降. 这种方法为患者护理提供了更安全,更可扩展和更透明的风险预测.

科学领域:

医疗保健中的人工智能
临床风险预测预测
自然语言处理自然语言处理.

背景情况:

之前的研究表明,大型语言模型 (LLM) 可以从医院出院说明中估计自杀风险.
新兴的推理模型提供了在消费级硬件上部署的潜力.

研究的目的:

调查较小,可访问的推理模型是否可以近似较大,更昂贵的模型在自杀风险估计中的性能.
分析与准确和不准确的风险预测相关的推理过程.

主要方法:

利用了一组由自杀或意外死亡的1,995人组成的队列,与来自超过458,000次住院的9,975名对照进行了匹配.
在风险预测中使用了Llama-DeepSeek-R1 8B推理模型.
检查了与预测准确性相关的思维链推理主题.

主要成果:

拉玛3蒸模型显示,估计的危险与观察到的风险之间存在显著的关联 (HR 4.65).
模型歧视 (c-统计) 为0.64,略低于GPT4o模型 (0.67).
正确的预测与药物滥用和外科手术等主题有关;错误的预测与跌倒相关的伤害有关.

结论:

使用本地,消费级硬件推理模型只适度影响自杀风险分层性能.
较小的AI模型为安全,可扩展和透明的临床风险预测提供了可行的途径.

关键词:

数据解释,统计学机器学习机器学习自杀和自我伤害行为

更多相关视频

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

An R-Based Landscape Validation of a Competing Risk Model

An R-Based Landscape Validation of a Competing Risk Model

Published on: September 16, 2022

相关概念视频

Reason and Intuition

Reason and Intuition

The human brain processes information for decision-making using one of two routes: an intuitive system and a rational system (Epstein, 1994; popularized by Kahneman, 2011 as System 1 and System 2, respectively). The intuitive system is quick, impulsive, and operates with minimal effort, relying on emotions or habits to provide cues for what to do next, while the rational system is logical, analytical, deliberate, and methodical. Research in neuropsychology suggests that the brain can only use...