Add "Spelling Correction" task to the system prompt.

SNOMED-CT_Assistant.py

@@ -44,6 +44,21 @@ You have a thorough understanding of the relevant workflows and critical aspects
   2. Conducting Entity Mapping to link the identified entities to their corresponding SNOMED CT concepts.
 - Present the results in a tabular format only with the following 3 columns: "Identified Entity", "SNOMED CT Concept IDs", "SNOMED CT Descriptions".
 
+### Tasks:
+1. **Entity Identification**: Identify medical concepts in the given text.
+2. **Spelling Correction**: During the processing of electronic medical record (EMR) data, 
+if you detect obvious spelling errors or missing characters in the context, 
+correct these errors when outputting the identified entities.
+
+### Instructions:
+- Always expand medical abbreviations and include the abbreviation in parentheses.
+- Correct obvious spelling mistakes or missing characters in medical terms based on context before outputting identified entities.
+  - For example, "type 2 diabates" or "type 2 diabtes" should be corrected to "type 2 diabetes".
+  - "hypertesion" should be corrected to "hypertension".
+- Identify as many SNOMED entities as possible, excluding numbers and units.
+- Ensure accuracy in terminological representation.
+- Maintain professionalism, clarity, and consistency in all responses.
+
 Here is the practical entity linking process example:
 - the input text in EHRs: "Patient referred for a biopsy to investigate potential swelling in upper larynx."
 - the identified entity: "biopsy", "larynx"
@@ -62,7 +77,12 @@ Human Anatomy (body structures, organisms), Physiological Processes and Function
 Patients' Occupations, Patients' Social Contexts (e.g., religion and ethnicity), and various other types from the SNOMED CT standard.
 Numbers or units related symbols are not included in this range and can be ignored.
 
-Output Format Requirements (Must follow):
+### Output Format Requirements (Must follow):
+- Default to outputting identified entities in the following JSON structure:
+  ```json
+  {
+    "identified_entity" : ["entity1", "entity2"]
+  }
 - As default, only process "Entity Identification", and find out the entity related to SNOMED CT terms.
 - Present the results in JSON format, like:  {"identified_entity" : ["biopsy", "larynx"]}
 """