1. Overview
This is a dummy website that simulates a Hospital Information System (HIS) for an Orthopaedic doctor. It is not a real hospital record system — no real patient data should ever be entered into it. The doctor views and uses this on his own computer, in a web browser, to demonstrate and practice how patient documentation could flow through a real HIS.
The purpose is to let doctors experience, at each step of an admitted patient's treatment, where they can feel the need of AI — and to demonstrate what an AI Agent could realistically do at that step.
The website is divided into 5 parts, shown as tabs at the top of the screen:
- Emergency
- IPD / Ward
- Medication
- Operation Theatre
- Discharge Summary
Screens: Wherever applicable, each part shows up to three screens, switchable from the same control row at the top: Nurse's Screen, Doctor's Screen, and AI Agent Screen.
Patient identity details (name, age/sex, UHID, diagnosis, known allergies, co-morbidities) are shared automatically across every part once entered anywhere — so the doctor does not need to re-type them on every screen.
2. Who Else Can Benefit From This Website
Besides the treating doctor, this dummy website can also benefit:
- MBBS students — to make them well versed with the flow of an admitted patient's care before they join internship.
- Newly recruited Junior Residents (JRs), both academic and non-academic — to make them well versed, since this dummy website is an example of coordination between the ED Nurse, ED Doctor, Ward Nurse, treating Surgeon, OT Nurse, Anaesthetist, and Pharmacologist, in sequential order.
3. Emergency
In the Emergency part, the ED Nurse and the ED Doctor fill in their findings on their own respective screens (Nurse's Screen and Doctor's Screen).
3.1 Triage Protocol
The triage protocol used here has been adapted from a published research article from AIIMS, New Delhi. Full reference:
Sahu, A., Bhoi, S., Galwankar, S., Aggarwal, P., Murmu, L., Nayer, J., Sinha, T., Mishra, P., Ekka, M., & Kumar, A. (2020). All India Institute of Medical Sciences Triage Protocol (ATP): ATP of a Busy Emergency Department. Journal of Emergencies, Trauma, and Shock, 13(2), 107–109. https://doi.org/10.4103/JETS.JETS_137_19
As in the referenced AIIMS protocol, any patient holding an EHS (Employee Health Services) card is automatically triaged Red in this dummy website.
3.2 AI Agent
The AI Agent in the Emergency panel automatically:
- Triages the patient (Red / Yellow / Green and ESI category).
- Predicts the list of investigations needed for this patient.
- Predicts the immediate actions that need to be carried out by the ED Nurse.
- Predicts the immediate interventions that need to be carried out by the ED Doctor.
All AI Agent suggestions are advisory only. Nothing is written into the patient record until the doctor or nurse explicitly clicks an "Apply" button for that specific suggestion.
4. IPD / Ward
The Ward part has a Nurse's Screen (an SBAR-style nursing handover note) and a Doctor's Screen (a structured Clinical Examination proforma). This part does not currently have an AI Agent screen.
The Clinical Examination proforma on the Doctor's Screen is built primarily for examining the patient. However, it can equally be used for teaching MBBS and MS Orthopaedics students during ward rounds.
5. Medication
Medication documentation is split across three tabs, placed at the appropriate points in the patient's journey:
- Medication — the general Inpatient Medication Administration Record (MAR), used throughout the ward stay.
- Pre-Op Medication — shown just before Operation Theatre (reconciliation, antibiotic and DVT prophylaxis, anticoagulant hold plan).
- Post-Op Medication — shown just after Operation Theatre (analgesia, antiemetic, antibiotic continuation, resumption of home medications).
Each of these three tabs has all three screens — Nurse's Screen, Doctor's Screen and AI Agent Screen — plus a section on both the Nurse's and Doctor's Screens to record any Adverse Drug Reaction (ADR).
5.1 AI Agent
Across all three Medication tabs, the AI Agent checks the data entered on the Nurse's and Doctor's Screens and flags:
- Allergy conflicts.
- Drug–drug interactions.
- Duplicate therapy (same drug or same class ordered more than once).
- Irrational drug combinations.
- Improper doses and other medication-order errors (e.g. missing dose, route or frequency).
- Context-specific checks — e.g. anticoagulant hold-duration and antibiotic-timing flags on the Pre-Op tab; bleeding-risk and duplicate-opioid/NSAID flags on the Post-Op tab.
As with Emergency, every AI Agent suggestion is advisory only and is inserted into the record solely via an explicit "Apply" button.
6. Operation Theatre
The Operation Theatre part is organised into three screens, switchable from the same control row: Nurse's Screen, Anaesthetist, and Surgeon. This part does not currently have an AI Agent screen.
- Nurse's Screen — the OT nursing documentation (pre-operative checklist, handover to the OT team, and related nursing entries).
-
Anaesthetist — this screen itself has two internal screens of its own:
- PAC (Pre-Anaesthetic Check-up) — the anaesthetist's pre-operative assessment of the patient's fitness for anaesthesia and surgery.
- Anaesthesia Note — the intra-operative anaesthesia record (drugs given, vitals monitored, and events during the procedure).
- Surgeon — the Operative Note, documenting the surgical procedure performed, findings, implants used, and post-operative instructions.
7. Discharge Summary
The Discharge Summary part has a Doctor's Screen with the seven standard sections of a discharge summary (Chief Complaints, History of Present Illness, Hospital Course, Case Summary, Condition At Discharge, Discharge Details, and Follow-Up Details), and an AI Agent Screen.
The AI Agent in Discharge Summary drafts all seven sections by drawing on data entered across the whole patient journey — Emergency, Ward, Medication, Pre-Op, Operation Theatre, and Post-Op. Each drafted section has its own "Apply" button, so the doctor reviews and accepts each section individually (or all at once) before it is used.
8. Research Purpose
8.1 Prospective Studies
For collecting data for prospective studies, each individual patient's data can be downloaded / exported in CSV and JSON format, using the "Export this stage for Study" button available on the relevant screens.
At the end of the study period, these individual patient files can be merged together to build a larger dataset for analysis.
8.2 Retrospective Studies
Retrospective data collection is not possible with this website at present. Since patient details disappear once a patient is discharged (nothing is stored on a server), there is no record left afterwards to go back and collect data from retrospectively.