Text size
Line height
Text spacing
Section outline
-
The AI Document Extraction feature has recently been introduced to a limited number of practices to simplify and speed up extracting demographic information directly from uploaded documents. You must be on GoodXweb version 1.20 or higher to use this feature.
Please note that the use of AI Document Extraction may be linked to supplementary features such as AI GoodXpert, a specialised service that requires a separate subscription and carries an additional fee. Please also note that certain functions, such as dispatching SMS notifications, may incur additional costs.
The AI Document Extraction tool is designed to process data from a single document. Any subsequent attempts to extract data will override previously imported information, and loading a new document will repopulate the readable fields with the most recently extracted data. For optimal results, it is recommended to use South African identity documents or accounts where the customer information is clearly presented in the document header. The source document can be saved to the patient file for additional information on custom intake forms not within the debtor file save/fields scope.
-
-
-
- What is the "AI Document Extraction" feature?
- Where to learn more about AI Document Extraction within GoodX:
- How is the “AI Document Extraction” used?
Change Log Date of Change Changes Editor 2026 05 12 Established Page Chendene Barnard
-
-
-
-
-
How is the “AI Document Extraction” used?
The AI Document Extraction service is available when adding a new debtor to the system.
To import a debtor, select the "Import Debtor" button to start the process.
The Import dropdown menu defaults to "Grand Central Members." However, you can expand this menu to select alternative methods, such as importing via "Medical Aid" or utilising the "AI Document Extraction" option.
Once the AI Document Extraction option is selected from the dropdown menu, you will need to complete several additional steps to ensure the data is correctly classified as a Case, Debtor, Patient, or Printout. To ensure accuracy, you must always specify the category of the document being uploaded: Financial, Personal, or Practice.
The settings default to the basic configuration of Debtor – Personal. You can upload the document as a high-resolution image file or a standard PDF.
Once the AI Document Extraction process is complete, the "Import" button will brighten, indicating that it is active and available for selection. After the extracted information appears, you may add the debtor or patient to the system as needed.
In the event that a patient or debtor record is duplicated—often triggered by multiple identifying values such as matching dates of birth or ID numbers—you have the discretion to select and import the record containing the most comprehensive and complete information.
Additionally, you have the option to save the uploaded document and link it directly to the debtor or patient profile currently being created for future reference by selecting the empty "Debtor Documents" box.
Secondary information can be removed by selecting the "X" icon, as highlighted below. It is essential to verify that you are not duplicating existing debtor or patient information during the import process.
If you inadvertently import duplicate information, you can remove the redundant entries in the patient information section. Once marked, the duplicate debtor will display a red strikethrough to indicate its status. Please note that these duplicates cannot be permanently hidden or removed from the system views.
Once you select "Save," the extracted information will automatically populate the corresponding fields. Perform a final review of these areas to ensure the data sourced from statements or alternative documents is accurate and relevant to the specific debtor or patient profile.
While you can have patients pre-fill forms and then use the AI Document Extraction tool to pull that typed data, you must carefully verify the results. Ensure all fields are populated correctly, and manually add, select, or correct any information as required to maintain data integrity.
Review the extracted information to ensure it was captured accurately, then select the correct supplementary details, such as the relevant account email addresses. Left represents information from AI document. Center represents current information in view. Right side would be blank for new debtors, and have information if it's an existing debtor.
Reminder: While the system can read various document types, the AI may currently prioritise the account source over the account holder. This functionality is still under active development and is being further evaluated to improve accuracy.
-
-
-