Skip to main content

4. System Logic

The logic of Clinic Edge is based on processing of static template data into dynamic, patient-specific cases and diagnostic reports. This process relies on several data bases including Supabasa tables, internal reference files and internal logic scripts. The data flow consists of multiple steps and interactions.

4.1 Data Flow​

Step 1: Case Selection via Cardinal Symptoms The user begins by selecting a cardinal symptom (e.g., "abdominal_pain") or a specific case from the landing page. Then system then matches the selection by searching the CaseTemplates table. For symptom-based selection, it filters cases whose case_data contains the selected symptom in the cardinal_symptoms array.

Step 2: Template Retrieval from CaseTemplates Once a matching case is selected, its corresponding template is retrieved from case_data in the CaseTemplates table. See Section 3.2.2 Case Templates for more details.

Step 3: Generation of pat_data → PatBase Using the selected case_data, the system dynamically generates a new patient case. Lab values, vitals, and findings are randomized within predefined ranges. If "normal" is defined or fields are left empty, data is injected from normalfindings_collection.json and labvalues.json. All probabilistic logic is handled via the internal logic files logic.js. The generated patient case is stored in PatBase and assigned a unique pat_id.

Step 4: Diagnostic Requests → PatFindings When the user requests an examination (e.g., "transvaginal ultrasound"), the system performs the following:

  1. Fetches the relevant data from pat_data in PatBase.
  2. Selects the corresponding template in FindingsTemplate.
  3. Generates and injects measurements via vars.
  4. Generates and injects written descriptions of abnormal findings and pathologies via vars_path.
  5. Assembles a complete, case-specific written report.

The rendered report is then stored in the PatFindings table.