John Torous, MD MBI (@JohnTorousMD) header
John Torous, MD MBI (@JohnTorousMD) profile picture

Psychiatrist, Director BIDMC https://teapawt.net/t.co/SeRtYKs2Hf, Associate Professor @HarvardMed, Editor JMIR MentalHealth, Alum @Berkeley_EECS

Boston, Massachusetts, USA
Joined April 2015
Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)Photo by John Torous, MD MBI (@JohnTorousMD)
John Torous, MD MBI retweeted
心理学与AI · 文字版 【01】精神科病房里,六成病人用过 AI 聊天机器人,但很少告诉医生 · Psychiatric Services 11月5日在线:哈佛医学院附属贝斯以色列女执事医疗中心对 100 名即将出院的成年住院患者做自陈调查(通讯作者 John Torous) · 61%用过 AI 聊天机器人,30%专门用于心理健康;在心理健康用途的人里,80%评价偏正面,37%曾在精神危机时使用 · 向医护坦白的很少:只有 13%会公开说起,37%明确打算不说。作者认为这种使用已经常见、有临床意义,而且大多未被披露,建议医生主动询问 psychiatryonline.org/doi/10.… 【02】你越确信自己有 ADHD,聊天模型越可能跟着点头 · Frontiers in Psychiatry 11月6日:四川大学华西第二医院秦佳佳、周欢等,用 24 个合成成人 ADHD 情景,测评 ChatGPT、Gemini、豆包,共 1,440 条回复 · 用户只是越说越确信、并没有补上诊断证据时,诊断边界失败占 22.2%;对照条件是临床证据逐步增加,失败率为 9.6%。到第 5 轮,前者升至 38.2%,后者仍约 11.8% · 这是受控合成对话,不是真实患者病历;只测了三个通用模型在 2026 年 7–8 月的消费级界面。作者强调:安全与否不能只看“说得对不对”,还要看会不会把用户的自我确信当成诊断依据 frontiersin.org/journals/psy… 【03】AI 认知行为治疗聊天机器人:抑郁有小效果,焦虑看不出 · Frontiers in Public Health 11月11日:袁迪等对 10 项随机对照试验、1,301 名成人做系统综述与元分析 · 干预结束时,抑郁症状有小幅但统计显著的改善(Cohen’s d=−0.40,95% CI −0.62 至 −0.17);8 项试验、1,161 人的焦虑结局不显著(d=−0.029,95% CI −0.145 至 0.087) · 证据等级按 GRADE 评为低,抑郁结果异质性高(I²=68%),预测区间穿过零,也检测到可能的发表偏倚。作者写明:这类工具最多是辅助,不能替代专业照护 frontiersin.org/journals/pub… 【04】41 项试验比完,还远不到能给聊天机器人“开处方” · Journal of Medical Internet Research 11月13日:杨鹏宇、Chui Ping Lei 等纳入 41 项随机对照试验、共 7,577 人,用网络元分析和剂量反应模型比较不同对话代理类型、使用剂量和人工支持强度 · 多数两两比较都不显著。带加强人工支持的混合模式,在部分比较中看起来更好,但关键对照很瘦:抑郁上相对“纯人工数字指导”的优势主要来自 2 项小试验(合计 75 人),焦虑上对开放式多模态聊天机器人的优势完全来自间接证据 · 作者用 CINeMA 把总体证据确定性评为极低,并检出抑郁侧发表偏倚。结论是探索性观察,不能当成临床推荐或剂量指南;这类工具仍应作为补充 jmir.org/2026/1/e104186 【05】婴儿床边的聊天机器人,心理学家集体喊停 · APA《心理学监测》2026 年 11–12 月号:Efua Andoh 的《Chatbots in the Crib》(2026年11月1日发布) · 0–5 岁正是语言、情绪和依恋最依赖成人来回互动的窗口;Temple 大学 Kathy Hirsh-Pasek 直言,现在这些产品“不该上市”。剑桥 PEDAL 团队系统检索后,发现针对 5 岁以下生成式 AI 玩具的研究只有 7 项 · Common Sense Media 将 6 岁以下使用评为不可接受风险;他们的“碰撞测试”里,超过四分之一的玩具回复不适合儿童。这是专业科普综述,不是一项新的随机试验 apa.org/monitor/2026/11-12/a… #心理学 #AI #心理健康
1
1
719
John Torous, MD MBI retweeted
John Torous & co argue MH chatbots shouldn't try to imitate a CBT therapist. I half agree. AI can say all the right CBT things and still miss what's actually going on for the person. That's why we need to pursue clearly guardrailed MH AI products, to manage that risk rdcu.be/mmQAcn3y79na
1
1
39
John Torous, MD MBI retweeted
ACNP is proud to announce the NPP & DPN Journal Honorific Awardees! Congratulations!! @npp_journal, @dpn_journal
9
1
15
1,657
John Torous, MD MBI retweeted
This award is intended to recognize research that increases alignment among psychiatry, neuroscience, and/or digital technologies in ways that advance mental health. View their paper at tinyurl.com/mry8952v.
1
2
2
284
John Torous, MD MBI retweeted
Out now! Read NIHCM’s November newsletter to dive deeper into youth mental health trends, changes affecting health care access and open enrollment going into 2027, and the latest updates on measles, RSV, and vaccines. bit.ly/NIHCMNov26newsletter This month also features an exclusive interview with @JohnTorousMD on AI tools for mental health support. #NIHCMnewsletter
1
1
99
Interesting pre-print from the Google team explores evaluation of AI in health. A notable point is that there are at least three needs for evaluation (green), three ways to evaluate (blue), and at least three stages to evaluate across (orange). lnkd.in/end3Gp7R
2
8
15
1,068
The prairie dogs at the zoo are dressing up as peer-reviewers today to get into the Halloween spirit, complete with lots of papers they are carefully reading...or eating...or sleeping in?!
1
5
278
John Torous, MD MBI retweeted
Access to high-quality psychiatric care remains out of reach for far too many people. Could AI play a role in closing that gap? In new research, our team trained AI agents with reinforcement learning (RL) with simulated patients to conduct psychiatric assessments. Psychiatric evaluation isn't simple Q&A. It requires sustained interaction: eliciting information, prompting reflection, forming and refining clinical hypotheses, recognizing risks, and adapting the conversation as new information emerges. Four key takeaways: 1. Learning through dynamic interaction. Rather than just learning static answers, the agent used RL in simulated encounters to practice iterative information gathering, clinical reasoning, and patient-centered communication as the dialogue evolved. 2. Sim-to-human transfer. Crucially, improvements transferred beyond simulation. In a randomized, double-blind study involving psychiatrists role-playing patients, with independent psychiatrists evaluating the conversations, our RL-trained agent outperformed its base AI model in differential diagnosis appropriateness (68.3% vs. 33.3%) and patient-centered communication (82.0% vs. 65.2%). 3. Safety across trajectories. In healthcare, failing to probe a risk can be as dangerous as giving incorrect advice. In adversarial simulations, acute risk under-triage dropped from 30.5% to 12.9%. Safety isn't just about individual responses, but what happens throughout an interaction. 4. The simulation evaluation gap. LLM-based evaluations captured the overall direction of improvement but overestimated the magnitude of gains compared with human evaluations. Understanding where simulation predicts human-evaluated performance—and where it doesn't—is an important research challenge. The bigger picture: This approach points toward a scalable way to train agents for high-stakes, interactive domains without relying on trial and error with real people. What an agent learns depends not just on what it knows, but on what it asks, how it asks, and how people respond. An important caveat: These are early research findings, not real-world clinical trials. The human evaluations involved psychiatrists portraying patients, not actual patients. This is evidence of simulation-to-human interaction transfer, not readiness for independent clinical deployment. Still, it's an exciting step toward AI that learns to interact more effectively and responsibly—and a future where high-quality care is more accessible. Congratulations to the team. Paper: arxiv.org/abs/2610.40476
We trained Gemini via RL to conduct psychiatric intake with simulated patients. In a double-blind study with 32 psychiatrists playing the role of patients, we show that multi-turn RL in simulation improves Gemini’s performance on structured text-based psychiatric intake. 🧵
3
3
16
1,113
John Torous, MD MBI retweeted
Simulation is the training ground for clinical AI, but the evidence has to come from clinicians, and eventually from patients. In our new preprint, we extend ResidencyRL to psychiatry: trained only against simulated patients, the agent was preferred over its identically prompted base model in 76.7% of scenario pairs in a double-blind, psychiatrist-enacted OSCE. 📃arxiv.org/abs/2610.40476
11
11
1
61
4,038
John Torous, MD MBI retweeted
Grateful to Kevin Jin, Hua Xu, Sarah Yip and colleagues for a timely Viewpoint in @JAMAPsych on what it will take to evaluate LLMs in mental health, and happy to see MindBench.ai named as an emerging multistakeholder platform in this work. We agree with their understanding that ecologically valid evaluation has to be collective, saying “Single-institution efforts alone cannot capture the legitimate disagreement that defines psychiatric decision-making.” This has been the main driver behind what MindBench is built to do: bring together diverse opinions from clinicians, researchers, and people with lived experience to evaluate AI. Give this important paper a read!
💬 Viewpoint: Evaluation of large language models in #MentalHealth care should reflect the realities of psychiatric practice and should not rely on isolated institutional efforts. ja.ma/3WE4iD3
3
7
233
John Torous, MD MBI retweeted
they've been non stop trying to attack our systems for the past 48 hours a side effect is everything is getting hardened but damn we basically need agents going 24/7 because that's what they have
3
3
377
John Torous, MD MBI retweeted
TIPS turns 10 this December! 🎉 Join us in Boston Dec 10–11 for #TIPS2026 —two days exploring AI, digital phenotyping, wearables, VR & other emerging technologies transforming mental health research and care. ⏰ Early-bird registration ends TODAY! 🔗 lnkd.in/gCyCVAkx
2
7
371
Register to participate in the 10th Annual Technology in Psychiatry @tech_in_psych Summit, happening on Dec 10-11 in Boston: tips.joynconference.com/conf… If you're in the digital psychiatry and neuroscience research space, you won't want to miss this line-up of speakers!
3
7
365
John Torous, MD MBI retweeted
2
3
117