HealthCare Chatbot system using artificial intelligence

Client Requirements

The client is a USA-based Healthcare provider. Healthcare communication platform that connects care teams and engages patients in real-time. The client wanted to develop a text interface (Chatbot) capable of answering queries and recording information related to patient activities & provider updates. Such as below:

• Patient Information Retrieval & Update.

• Patient Appointment Booking.

• Patient Diet Tracking.

• Patient Information Gathering via. Surveys or Forms.

Thinkitive Solution

Thinkitive’s business analysts and subject matter experts started with the requirement analysis (Discovery) phase. The Thinkitive team took multiple calls with the client and created detailed SRS requirement documents and low-fidelity wireframes. Thinkitive established a team of professionals with expertise in frontend and backend development. The Thinkitive team leveraged Rasa Framework and developed a Healthcare Chatbot using Rasa NLU (Natural Language Understanding) for intent classification and entity extraction to identify user input and trigger the respective workflow and Rasa Core for deciding the set of actions to perform based on the previous set of user inputs.

• Rasa NLU:

  Rasa NLU (Natural Language Understanding): Rasa NLU is an open-source natural language processing tool for intent classification (decides what the user is asking), extraction of the entity from the bot in the form of structured data, and helping the chatbot understand what the user is saying.

• Tokenizers:

  Tokenizers split the text into tokens. If you want to split intents into multiple labels, e.g., for predicting multiple intents or modeling hierarchical intent structure, the following flags were used with the tokenizer.

• Tokens:

  Tokenization is the task of chopping text up into pieces, called tokens, and at the same time throwing away certain characters, such as punctuation.

• Intent:

  Intent can be understood as labels that are attached based on the overall goal of the user’s message. For example, a user inputs ‘Hello’ or ‘Hi’ may call the intent ‘greet’, and a user inputs ‘bye’ or may call the intent ‘goodbye’. Rasa uses the concept of intent to describe how user messages should be categorized. Rasa NLU will classify the user messages into one or also multiple user intents.

• Entity:

  Intent can be understood as labels attached based on the overall goal of the user’s message. For example, a user inputs ‘Hello’ or ‘Hi’ may call the intent ‘greet,’ and a user inputs ‘bye’ or may call the intent ‘goodbye.’ Rasa uses the concept of intent to describe how user messages should be categorized. Rasa NLU will classify the user messages into multiple user intents.

Solution Highlights

• Appointment Creation:

  Whenever a patient comes to the chatbot, it greets the patient, then the patient asks for an appointment booking, then asks if they are registered with the provider or not. If the patient is not registered, it asks for basic details like Full Name, Date of Birth, Gender, Mobile number, email-based, and Full Address. On this, a new patient's record is created in the EHR, and the chatbot moves to the appointment creation flow. If the patient is registered with the provider-patient confirms their EHR number, and then the chatbot moves to the appointment creation flow.



• Appointment Rescheduling:

  Whenever a patient comes to the chatbot, it greets the patient, then the patient asks for an Appointment Rescheduling, then the chatbot asks for the EHR number, and the chatbot moves to the appointment rescheduling flow.



• Appointment Cancellation:

  Whenever a patient comes to the chatbot, it greets the patient, then asks for an Appointment Cancellation, then the chatbot asks EHR, and the chatbot moves to the appointment cancellation flow.

Value Delivered

• Enable patients to find the best specialist available across the entire U.S.

• Patient access to provider appointment booking facility, appointment rescheduling facility, appointment cancellation facility, and more.

• The provider doesn’t need to manage the EHR anymore; the provider can simply take notes and push to the EHR with a single click.

• A chatbot can not only be used to dispense medical advice but also to send daily reminders to patients.

• The goal of any healthcare provider is to provide exceptional services to their patients. Chatbots are a great way to collect feedback. Healthcare providers can use that feedback to do their practice even better.

• A chatbot can sort the patient according to their urgency for medical attention.