MindFlow keeps every patient's information in one connected record, demographics, visit history, diagnoses, prescriptions, vitals, questionnaire scores, and attached lab reports. All personal health data is encrypted at rest and in transit, and every access is logged for audit purposes.
Each patient has a single canonical record that all clinical activity, notes, prescriptions, vitals, assessments, links back to. Searching by name, phone, or unique ID returns results instantly regardless of how many records the clinic has.
Patient photos are stored with access controls, they are never served via a guessable public URL. Access requires an authenticated session, and every photo retrieval is logged.
Clinics can define their own intake fields without any system changes. Additional fields are stored as validated structured data attached to the patient record.
All patient-identifying information is treated as sensitive personal data under India's DPDP Act 2023. Encryption is applied at rest for sensitive fields and in transit for all communications.
Every time a note is saved, the previous version is preserved immutably. The clinician always sees the latest version, but the full history is accessible, nothing is ever overwritten or lost.
Once a clinician signs a note, it is locked. Any amendment after signing creates a new version linked to the original, with the signing clinician's identity, timestamp, and session details recorded.
After a session, the AI layer can generate a draft note from structured session data and dictation. The draft is presented for review, it is never auto-saved. The clinician edits, approves, and signs before the note is committed to the record.
Configurable alert thresholds, such as a high-severity assessment score or an extreme vital reading, trigger immediate notifications to the assigned clinician via the platform's notification system.
ICD-11 (International Classification of Diseases, 11th Revision), the WHO standard, and the version required for ABDM/FHIR record exchange in India.
Diagnosis codes are fetched from the official WHO ICD-11 API using secure authenticated calls. Results are cached locally to ensure the search is fast even during a consultation.
Each diagnosis entry stores the ICD-11 code, the WHO display term, the clinical modifier added by the doctor, and the date. Multiple diagnoses per visit are supported.
When health records are shared through the ABDM network, diagnoses are transmitted as standard FHIR Condition resources coded against the official WHO ICD-11 URI, readable by any compliant system.
PHQ-9 (depression), GAD-7 (anxiety), HAM-D (Hamilton Depression Rating), YMRS (Young Mania Rating), PANSS (psychosis), MDQ (mood disorder screening), and AUDIT (alcohol use) are pre-built. Scores are calculated automatically and risk bands, such as PHQ-9 ≥ 20 triggering a severe-range alert, are enforced server-side.
Clinics can define their own questionnaires, question text, response options, scoring weights, and total-score formula, which appear in the workflow alongside built-in scales.
A clinician can send a questionnaire to the patient to complete from their portal. When the patient submits, the scored result is automatically attached to their clinical record and any risk threshold breaches trigger notifications.
Each prescription supports multiple drugs, each with dose, unit, frequency, route, duration, and instructions. Lab orders and vitals readings can be linked directly to a prescription for context.
Prescriptions render using the clinic's configured letterhead, logo, address, registration number, in a print-ready format. PDF generation is handled server-side for consistent output across devices.
The clinician speaks the prescription aloud. The audio is sent to an Indian-language speech recognition service (Sarvam AI), transcribed, and then parsed by the AI layer into structured prescription fields. The result pre-fills the form, the clinician reviews and confirms before saving.
An autocomplete-powered drug library, covering generic names, common brands, standard doses and frequencies, speeds up prescription entry and reduces transcription errors.
Clinics choose which vital signs to track (blood pressure, pulse, weight, BMI, SpO₂, temperature, blood sugar, etc.) and define the normal ranges and units for their practice.
For patients where the standard normal range doesn't apply clinically, a known hypertensive, for example, the alert thresholds can be set individually, so alerts reflect the patient's own baseline.
Patients can optionally authorise sharing of activity, sleep, and heart-rate data from Google Fit or Apple Health. Once authorised, the data syncs automatically and appears in the patient's vitals timeline alongside clinician-recorded readings.
The demo comes pre-loaded with real sample data, so you can click through actual appointments, patients and prescriptions.