| Both sides previous revision
Previous revision
Next revision
|
Previous revision
|
chatehr [2026/08/03 18:14]
nigam [Automations] |
chatehr [2026/08/05 09:26] (current)
nigam [Automations] |
| ChatEHR is a vendor-agnostic, internally governed platform that places the full longitudinal patient record—up to twenty million tokens—directly into a model's context window in real time. It is accessible both through an interface embedded in the EHR (40% usage) and through APIs (60% usage) that automate high-volume rote tasks. This [[https://arxiv.org/abs/2602.00074|ArXiv submission]] describes the usage in the first 3 months of launch. | ChatEHR is a vendor-agnostic, internally governed platform that places the full longitudinal patient record—up to twenty million tokens—directly into a model's context window in real time. It is accessible both through an interface embedded in the EHR (40% usage) and through APIs (60% usage) that automate high-volume rote tasks. This [[https://arxiv.org/abs/2602.00074|ArXiv submission]] describes the usage in the first 3 months of launch. |
| | |
| | ===== Automations ===== |
| | |
| | Automations are fixed prompts applied to defined sets of patients. Across seven operational workflows, these automations convert high-volume, repetitive review into focused, actionable work embedded in the EHR. They help place patients in the right care setting, identify surgery cases who may need medical co-management, strengthen infection surveillance, accelerate referral triage, surface patients who may benefit from inpatient hospice, standardize complex orthopedic referrals, and validate billing accounts. |
| | |
| | {{ ::chatgpt_image_aug_4_2026_03_12_58_pm.png?nolink |}} |
| | |
| | |
| | * [[https://pulse.stanfordmedicine.org/articles/ai-tool-will-streamline-patient-screening-at-sequoia | Streamlined patient screening]] turns a 5 to 20 minutes chart review for 150 patient charts to roughly 30 highly probable cases. |
| | * [[https://pulse.stanfordmedicine.org/articles/chatehr-automation-categorizes-surgical-patients-for-better-care-coordination | Identifying patients for surgical co-management]] lets hospitalists focus on the right neurosurgery, orthopedics, and ENT patients. |
| | * [[https://pulse.stanfordmedicine.org/articles/chatehr-automation-supports-enhanced-surveillance-of-surgical-site-infections | Surveillance of surgical site infections]] reduces chart review by 40%, and improves the timeliness of infection reporting. |
| | * [[https://pulse.stanfordmedicine.org/articles/enterprise-contact-center-launches-chatehr-automation-referral-triage-indexing | Referral triage at the Contact Center]], cuts time by 17.7%, raises throughput by 17.8%, and reduced backlog from five to two days. |
| | * [[https://pulse.stanfordmedicine.org/articles/how-ai-can-help-identify-eligibility-for-inpatient-hospice | Identifying eligibility for Inpatient Hospice]], screens 600 patient charts to find 10–40 eligible cases to avoid 18–20 hours each day. |
| | * [[https://pulse.stanfordmedicine.org/articles/orthopedics-launches-chatehr-automation-referral-summary-tool | Accelerated referral triage in Orthopedics]], processes over 4,700 documents and saves at least 1.7 staff hours each day. |
| | * [[https://pulse.stanfordmedicine.org/articles/patient-financial-services-launches-chatehr-automation-for-billing-review | Review of accounts with an LoA stop bill]], saves 10–15 min/account per resolution and lowers underpayment recovery cost. |
| | |
| | \\ |
| |
| ===== User Interface ===== | ===== User Interface ===== |
| {{ :screenshot_2026-07-30_at_4.34.09 pm.png?nolink& }} | {{ :screenshot_2026-07-30_at_4.34.09 pm.png?nolink& }} |
| |
| | \\ |
| ==== Testimonials ==== | ==== Testimonials ==== |
| |
| {{ :chatehr-testimonials.png?nolink& }} | {{ :chatehr-testimonials.png?nolink& }} |
| |
| | \\ |
| ==== Current usage ==== | ==== Current usage ==== |
| |
| The panels below show current usage as of July 2026. The top left panel shows usage by role and to top right shows usage by service line. The bottom left panel shows usage by time and day of week, and the bottom right panel shows the top 20 tasks performed by the users. | The panels below show current usage as of July 2026, with 3400+ active users, 150,000+ sessions, and 140B tokens processed. The top left panel shows usage by role and to top right shows usage by service line. The bottom left panel shows usage by time and day of week, and the bottom right panel shows the top 20 tasks performed by the users. |
| |
| {{ ::screenshot_2026-08-03_at_1.42.18 pm.png?nolink |}} | {{ ::screenshot_2026-08-03_at_1.42.18 pm.png?nolink |}} |
| |
| ===== Automations ===== | |
| |
| Automations are fixed prompts applied to defined sets of patients, and performance is quantified using a clinician-curated gold-standard truth set and impact is measured during monitoring. There are seven automations for which information is publicly available. | ---- |
| |
| - [[https://pulse.stanfordmedicine.org/articles/ai-tool-will-streamline-patient-screening-at-sequoia | Streamline patient screening]] | \\ |
| - [[https://pulse.stanfordmedicine.org/articles/chatehr-automation-categorizes-surgical-patients-for-better-care-coordination | Identify patients for medical co-management during surgery]] | ===== Our Journey ===== |
| - [[https://pulse.stanfordmedicine.org/articles/chatehr-automation-supports-enhanced-surveillance-of-surgical-site-infections | Enhanced surveillance of surgical site infections]] | |
| - [[https://pulse.stanfordmedicine.org/articles/enterprise-contact-center-launches-chatehr-automation-referral-triage-indexing | Referral triage at the Enterprise Contact Center]] | |
| - [[https://pulse.stanfordmedicine.org/articles/how-ai-can-help-identify-eligibility-for-inpatient-hospice | Identify eligibility for Inpatient Hospice]] | |
| - [[https://pulse.stanfordmedicine.org/articles/orthopedics-launches-chatehr-automation-referral-summary-tool | Accelerate referral triage in Orthopedics]] | |
| - [[https://pulse.stanfordmedicine.org/articles/patient-financial-services-launches-chatehr-automation-for-billing-review | Review of accounts with an LoA stop bill]] | |
| |
| | ChatEHR was made possible by [[https://med.stanford.edu/news/all-news/2022/03/nigam-shah-inaugural-chief-data-scientist-stanford-health-care.html | embedding the data science team]], which includes software engineers, product managers, and data scientists with direct access to academic researchers, in the IT organization. Because of this set up, what began as a sandbox in April 2023 was conceptualized as ChatEHR in November 2023, underwent multiple prototypes by mid-2024, leading to a pilot in early 2025 that scaled to 150 clinical users by August 2025, and a health system-wide roll out by September 2025. |
| | |
| | {{ ::chatgpt_image_aug_3_2026_06_24_36_pm.png?nolink |}} |
| | |
| | \\ |
| ===== News coverage ===== | ===== News coverage ===== |
| |
| - [[https://www.statnews.com/2026/01/28/stanford-research-ai-validation-tools-ai-prognosis/|Stanford’s health AI validation tools you should know about]] | - [[https://www.statnews.com/2026/01/28/stanford-research-ai-validation-tools-ai-prognosis/|StatNews - Stanford’s health AI validation tools you should know about]] |
| - [[https://www.statnews.com/2026/01/28/stanford-penn-hospitals-build-electronic-health-record-chatbots/|Why some hospitals are making their own ChatGPTs for patient records]] | - [[https://www.statnews.com/2026/01/28/stanford-penn-hospitals-build-electronic-health-record-chatbots/| StatNews - Why some hospitals are making their own ChatGPTs for patient records]] |
| - [[https://med.stanford.edu/news/all-news/2025/06/chatehr.html|Stanford News - Clinicians can ‘chat’ with medical records through new AI software, ChatEHR]] | - [[https://med.stanford.edu/news/all-news/2025/06/chatehr.html|Stanford News - Clinicians can ‘chat’ with medical records through new AI software, ChatEHR]] |
| - [[https://news.stanford.edu/stories/2025/06/ai-medical-records-chat-ehr|AI lets clinicians ‘chat’ with medical records]] | - [[https://news.stanford.edu/stories/2025/06/ai-medical-records-chat-ehr| Stanford News - AI lets clinicians ‘chat’ with medical records]] |
| - [[https://www.mobihealthnews.com/news/stanford-medicines-chatehr-expedites-chart-review-process|MobiHealthNews - Stanford Medicine's ChatEHR expedites the chart review process]] | - [[https://www.mobihealthnews.com/news/stanford-medicines-chatehr-expedites-chart-review-process|MobiHealthNews - Stanford Medicine's ChatEHR expedites the chart review process]] |
| - [[https://www.healthcareitnews.com/news/new-chatehr-tool-enables-clinical-conversation-stanford|HealthcareITNews - New 'ChatEHR' tool enables clinical conversation at Stanford]] | - [[https://www.healthcareitnews.com/news/new-chatehr-tool-enables-clinical-conversation-stanford|HealthcareITNews - New 'ChatEHR' tool enables clinical conversation at Stanford]] |
| - [[https://yesilscience.com/stanford-medicine-introduces-chatehr-for-efficient-chart-reviews/|Stanford Medicine Introduces ChatEHR for Efficient Chart Reviews]] | - [[https://yesilscience.com/stanford-medicine-introduces-chatehr-for-efficient-chart-reviews/|Stanford Medicine Introduces ChatEHR for Efficient Chart Reviews]] |
| - [[https://www.medicaleconomics.com/view/stanford-s-chatehr-lets-doctors-talk-to-the-chart-and-it-talks-back|Stanford’s ChatEHR lets doctors talk to the chart — and it talks back]] | - [[https://www.medicaleconomics.com/view/stanford-s-chatehr-lets-doctors-talk-to-the-chart-and-it-talks-back|Stanford’s ChatEHR lets doctors talk to the chart — and it talks back]] |
| |
| ===== Timeline ===== | |
| |
| * Apr 06, 2023 - Data Science team sets up SHCOpenAISandbox at [[http://goto.stanford.edu/shcopenaisandbox|http://goto.stanford.edu/shcopenaisandbox]] | |
| * Aug - 2023 - Work begins on setting up securegpt.stanfordhealthcare.org | |
| * Nov 01, 2023 - chatEHR is conceptualized at TDS while building [[https://securegpt.stanfordhealthcare.org/|https://securegpt.stanfordhealthcare.org/]] | |
| * Nov 08, 2023 - A working demo is created by a student as an [[https://shahlab.stanford.edu/ai-care|AI-CARE]] project. | |
| * Jan 29, 2024 - SecureGPT, launched for Stanford School of Medicine and Stanford Health Care | |
| * Jan - March, 2024 - Proof of concept work in ENT referral and Nursing summarization using SecureGPT. | |
| * May - 2024 - the concept of using an LLM in context with a single patient record is fleshed out | |
| * June 10, 2024 - second demo and prototype | |
| * June 19, 2024 - presentation to Data Science Exec Committee, and permission to build | |
| * July 2024 - First automation use-case implemented to automate chart review for inter-facility transfers | |
| * Aug 15, 2024 - Backend architecture overhaul | |
| * Aug 23, 2024 - Alpha testers get access to the UI | |
| * Jan - 2025 - First user gets access for UI | |
| * June 5, 2025 - Announcement of the pilot, with 30+ active users | |
| * July 2, 2025 - Ramp up to 89 users (clinicians, nurses, residents, fellows, PAs, care managers, MD students) | |
| * Aug 1, 2025 - Pilot ramped up to ~150 users, and closed new enrollment into the pilot | |
| * Sept 9, 2025 - Broad roll out to all providers, and APPs. | |
| |
| |
