hospital-care-api-documentation

HOSPITAL CARE API DOCUMENTATION

1. initiate intervew id                 #This endpoint is for generating interview id by taking users age and sex
api = https://api.hospitalcare.co.in/initial
sample payload = {
    "age":"25",
    "age_unit":"year",
    "sex":"male"
    }

2. diagnosis_start                     #This will start the diagnosis by taking a message from the user about their initial symptoms
api = https://api.hospitalcare.co.in/diagnosis_start
sample payload =  {
    "message":"i am feeling vomit",
    "extras":""
    }

3. interview                         #After gathering enough evidence this endpoint will be hit, which will start 1-1 interview.
api = https://api.hospitalcare.co.in/interview
sample payload=
 {
    "id": "s_112",
    "choice_id": "absent",
    "source":""
 }                                  #this will return conditions probablity

4. triage                          #for consultation level
api = https://api.hospitalcare.co.in/triage
sample payload = no body

5. specialist_recommender         #for specialist doctor recommendation
api = https://api.hospitalcare.co.in/specialist_recommender



Important points:

* The "/diagnosis_start" endpoint will return a list of 4 suggested symptoms, from which the user has to choose 1 symptom for now.
 In the future, we will develop it to accept multiple suggested symptoms at a time.

* The "/interview" endpoint's payload has a "source" attribute. Selected "suggested symptom" from the previous endpoint has to be defined its source as "suggest". After hitting the API with the payload it will return the diagnosis question in its response.
After the first hit, it will also return two attributes {"should_stop":bool} and {conditions: }. The interview will keep running until the "should_stop" is not returning "true". when the "should_stop" attribute will return true Interview will end. The condition probabilities will be stored in the variable for the final report generation.

* The "/triage" endpoint can categorize provided patient cases based on the seriousness of reported observations and the severity of likely conditions. This endpoint will be hit without any payload by the "POST" method.

* The "/specialist_recommender" endpoint extends the /triage recommendation by providing the most appropriate specialist by taking into account not only the severity of the most likely conditions but also the presence of alarming symptoms. This endpoint will be hit without any payload by the "POST" method.