Steps to Build a Multilingual Healthcare AI Chatbot

August 22, 2024

Key Takeaways

Healthcare startups are focusing on making healthcare advice more accessible across India’s diverse demographics.
By combining an LLM fine-tuned for the healthcare domain with Retrieval-Augmented Generation (RAG) architecture, along with a robust sector-specific dataset, it is now possible to create advanced healthcare AI solutions.
Given the importance of data sovereignty, it is advisable to deploy such a chatbot on AI-first cloud platforms that are MeitY-empanelled, such as E2E Cloud.
The future of healthcare is AI-powered, and startups should consider exploring emerging healthcare LLMs and advanced RAG architectures to stay ahead.

Introduction

Multilingual healthcare AI chatbots are increasingly being adopted by startups and businesses in the healthcare domain to enhance accessibility to health advice across India's linguistically diverse user base. With the rapid advancements in artificial intelligence, particularly in open-source multilingual large language models (LLMs), developing a chatbot that can understand and respond in multiple languages has become a viable reality.

In this guide, we will walk you through the steps needed to build a multilingual healthcare AI chatbot, right from selecting the right technologies to explaining the correct deployment approach. By the end of this article, you'll have a clear roadmap to building a powerful tool that bridges language barriers and improves accessibility to healthcare services for everyone.

Understanding the Technical Architecture

Creating any intelligent chatbot involves several steps, from loading and processing documents to generating meaningful responses based on user queries. In order to build this chatbot, we will use the following architecture:

‍

LangChain framework: We will leverage LangChain framework, which is designed to simplify the development of applications that integrate LLMs. It offers tools for chaining together LLMs with various data sources, enabling more complex and dynamic AI-driven workflows. LangChain's modular approach allows you to easily build and customize applications such as chatbots, data analyzers, and automated content generators by connecting different components like prompts, memory, and knowledge bases.

Llama 3.1: We will use the cutting-edge open-weight LLM from Meta - Llama 3.1. Llama 3.1 is the latest iteration of the Llama language model series, known for its enhanced efficiency and accuracy in natural language understanding and generation. Building on its predecessors, Llama 3.1 offers improved contextual comprehension and multilingual capabilities, making it a powerful tool for diverse applications in AI, including content creation, translation, and conversational agents. This version also features optimizations that reduce computational requirements, making it more accessible for broader use cases.

Qdrant: In order to build our RAG architecture, we will use the Qdrant vector store. Qdrant is optimized for handling high-dimensional data, making it ideal for applications in machine learning, recommendation systems, and AI-driven search. With its scalable, open-source architecture, Qdrant enables efficient and accurate similarity searches, allowing you to build powerful applications that leverage large datasets and complex queries.

For our dataset, we will use the “A Z Family Medical Encyclopedia” dataset. Finally, to showcase the responses generated by our Chatbot, we will use Gradio.

While we are demonstrating the chatbot's responses using Gradio, we recommend building APIs and leveraging WebSocket when developing for production deployments.

Building on India’s Top MeitY-Empanelled Cloud

We will leverage E2E Cloud to build and deploy this chatbot. Beyond being the most price-performant cloud in the Indian market, E2E Cloud is also MeitY-empanelled. This designation means that E2E Cloud meets the stringent security and compliance standards set by the Indian government, ensuring that your data, especially customer data, is protected and managed in accordance with Indian IT laws.

Additionally, being MeitY-empanelled signifies that E2E Cloud is trusted for handling sensitive information, making it a reliable choice for healthcare applications where data security and privacy are paramount.

Note: Data security and sovereignty are ultimately a shared responsibility. It is crucial to have robust internal security policies and practices in place to complement the cloud provider's security measures. This includes implementing strong encryption protocols, regularly updating security settings, conducting thorough access management, and continuously monitoring your systems for potential vulnerabilities. By combining E2E Cloud's secure infrastructure with diligent in-house security practices, you can better safeguard your data and maintain compliance with relevant regulations.

Steps to Build a Healthcare AI Chatbot

First, sign up to E2E Cloud using MyAccount. Next, launch a cloud GPU node. Since you are going to use Llama 3.1, you need a cloud GPU which has a minimum RAM of 16 GB (ideally 32 GB).

Before launching the node, you need to add your SSH key so you can login to E2E Cloud easily.

Prerequisites

Once you have SSH-ed into the node, go ahead and create a Python virtual environment.


$ python3 -m venv .venv
$ source .venv/bin/activate

Now you can either use VS Code with Remote Explorer extension, or start a Jupyter Lab.


$ pip install jupyterlab
$ jupyter lab

You can also use TIR, which will allow you to skip the two steps mentioned above entirely. Explore TIR by clicking on "TIR AI Platform" in the top navbar:

Once you have your Jupyter environment up, go ahead and install the following libraries:


! pip install sentence-transformers
! pip install qdrant_client
! pip install gradio

Step 1: Loading and splitting the PDF

We will load the PDF document from our dataset using LangChain's PyPDFLoader and split it into manageable chunks with the Recursive Character Splitter:


from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter

loaders = [PyPDFLoader("/path/to/A-Z Family Medical Encyclopedia.pdf")]
docs = []
for loader in loaders:
    docs.extend(loader.load())

chunk_size = 500
chunk_overlap = 250
r_splitter = RecursiveCharacterTextSplitter(chunk_size=chunk_size, chunk_overlap=chunk_overlap)
splits = r_splitter.split_documents(docs)
chunks = [doc.page_content for doc in splits]

Step 2: Deploying the LLM

We will use Ollama to deploy the LLM. Alternatively, you can easily create a TIR endpoint using vLLM serving:

‍

Select vLLM in the “Launch Inference” step, and the above UI will launch. You can then select the model from the “Model” dropdown menu.

Alternatively, to use Ollama with your Cloud GPU, you can follow these steps:


$ curl -fsSL https://ollama.com/install.sh | sh
$ ollama pull llama3.1

Step 3: Encoding the chunks using a pre-trained embedding model

You can use a pretrained model like neuml/pubmedbert-base- embeddings for turning chunks into embeddings by using the sentence-transformers library:


from sentence_transformers import SentenceTransformer

model = SentenceTransformer("neuml/pubmedbert-base-embeddings")
vectors = model.encode(chunks)

Step 4: Storing the embeddings in Qdrant

Now, you can store these embeddings in a database like Qdrant, which can also be used for semantic searches. The choice of the vector database is yours.


from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams

client = QdrantClient(":memory:")

client.recreate_collection(
    collection_name="medical-encyclopaedia",
    vectors_config=VectorParams(size=len(vectors[0]), distance=Distance.COSINE),
)

client.upload_collection(
    collection_name="medical-encyclopaedia",
    ids=[i for i in range(len(chunks))],
    vectors=vectors,
)

Step 5: Implementing the Context Generation Function

We will now create a function that will fetch the context based on the query vector. It will use similarity search to find document chunks closest to the query:


def create_context(question):
    ques_vector = model.encode(question)
    result = client.query_points(collection_name="medical-encyclopedia", query=ques_vector)
    
    sim_ids = [i.id for i in result.points]
    context = "".join([chunks[i] for i in sim_ids[:2]])
    return context

Step 6: Generating responses

When encoding our query vector, we have used the same function that we used to embed our documents in the model.encode function.

When we call the create_context function, it uses similarity search to fetch the documents and generate the context.

In the context, we additionally specify the language we want the responses in. Since Llama 3.1 is a multilingual LLM, we can use its language ability to create a multilingual chatbot.


def respond(question, language):
    context = create_context(question)
    chat_completion = groq_api.chat.completions.create(
        messages=[
            {
                "role": "user",
                "content": f"This is the question asked by the user: '{question}' and the context given is: '{context}'. Answer this question based on the context provided and in '{language}'.",
            }
        ],
        model="llama-3.1-70b-versatile",
    )
    return chat_completion.choices[0].message.content

Step 7: Integrating a web interface

You can use Gradio to build a web interface for the chatbot. Users can ask questions and receive meaningful responses based on the context provided:


import gradio as gr

def respond(user_query):
    return f"This is a placeholder response for your query: {user_query}"

with gr.Blocks() as demo:
    gr.Markdown("Healthcare Chatbot")
    gr.Markdown("Hello! Ask me anything about health.")
    
    user_query = gr.Textbox(placeholder="E.g., Tell me about indigestion?", label="Enter your question:")
    output = gr.HTML()

    def generate_response(user_query):
        bot_response = respond(user_query)
        return f"Answer:
{bot_response}"

    submit_button = gr.Button("Answer")
    submit_button.click(fn=generate_response, inputs=user_query, outputs=output)

demo.launch()

Output:

And that’s it! We have our chatbot ready.

Next Steps

Building a multilingual healthcare AI chatbot is a crucial step toward making healthcare more accessible and personalized for users from diverse linguistic backgrounds. By leveraging advanced AI technologies and cloud infrastructure, you can create a powerful tool that not only breaks down language barriers but also delivers timely and accurate medical assistance to those who need it most. As we’ve outlined, the process involves careful planning, the right technology stack, and a commitment to data security and compliance.

To bring your chatbot to life and ensure it runs efficiently, consider deploying it on E2E Cloud. With its MeitY empanelment and industry-leading price-performance ratio, E2E Cloud provides the secure and scalable infrastructure you need to support your AI applications.

Take the next step in your AI journey—sign up for E2E Cloud today and start building your multilingual healthcare chatbot on a platform designed for success.

Sign up for Free Trial

Latest Blogs

November 4, 2024

A Complete Guide To Customer Acquisition For Startups

Any business is enlivened by its customers. Therefore, a strategy to constantly bring in new clients is an ongoing requirement. In this regard, having a proper customer acquisition strategy can be of great importance.

So, if you are just starting your business, or planning to expand it, read on to learn more about this concept.

The problem with customer acquisition

As an organization, when working in a diverse and competitive market like India, you need to have a well-defined customer acquisition strategy to attain success. However, this is where most startups struggle. Now, you may have a great product or service, but if you are not in the right place targeting the right demographic, you are not likely to get the results you want.

To resolve this, typically, companies invest, but if that is not channelized properly, it will be futile.

So, the best way out of this dilemma is to have a clear customer acquisition strategy in place.

How can you create the ideal customer acquisition strategy for your business?

Define what your goals are

You need to define your goals so that you can meet the revenue expectations you have for the current fiscal year. You need to find a value for the metrics –

MRR – Monthly recurring revenue, which tells you all the income that can be generated from all your income channels.
CLV – Customer lifetime value tells you how much a customer is willing to spend on your business during your mutual relationship duration.
CAC – Customer acquisition costs, which tells how much your organization needs to spend to acquire customers constantly.
Churn rate – It tells you the rate at which customers stop doing business.

All these metrics tell you how well you will be able to grow your business and revenue.

Identify your ideal customers

You need to understand who your current customers are and who your target customers are. Once you are aware of your customer base, you can focus your energies in that direction and get the maximum sale of your products or services. You can also understand what your customers require through various analytics and markers and address them to leverage your products/services towards them.

Choose your channels for customer acquisition

How will you acquire customers who will eventually tell at what scale and at what rate you need to expand your business? You could market and sell your products on social media channels like Instagram, Facebook and YouTube, or invest in paid marketing like Google Ads. You need to develop a unique strategy for each of these channels.

Communicate with your customers

If you know exactly what your customers have in mind, then you will be able to develop your customer strategy with a clear perspective in mind. You can do it through surveys or customer opinion forms, email contact forms, blog posts and social media posts. After that, you just need to measure the analytics, clearly understand the insights, and improve your strategy accordingly.

Combining these strategies with your long-term business plan will bring results. However, there will be challenges on the way, where you need to adapt as per the requirements to make the most of it. At the same time, introducing new technologies like AI and ML can also solve such issues easily. To learn more about the use of AI and ML and how they are transforming businesses, keep referring to the blog section of E2E Networks.

‍

Reference Links

https://www.helpscout.com/customer-acquisition/

https://www.cloudways.com/blog/customer-acquisition-strategy-for-startups/

https://blog.hubspot.com/service/customer-acquisition

This is a decorative image for: Constructing 3D objects through Deep Learning

October 18, 2022

Image-based 3D Object Reconstruction State-of-the-Art and trends in the Deep Learning Era

3D reconstruction is one of the most complex issues of deep learning systems. There have been multiple types of research in this field, and almost everything has been tried on it — computer vision, computer graphics and machine learning, but to no avail. However, that has resulted in CNN or convolutional neural networks foraying into this field, which has yielded some success.

The Main Objective of the 3D Object Reconstruction

Developing this deep learning technology aims to infer the shape of 3D objects from 2D images. So, to conduct the experiment, you need the following:

Highly calibrated cameras that take a photograph of the image from various angles.
Large training datasets can predict the geometry of the object whose 3D image reconstruction needs to be done. These datasets can be collected from a database of images, or they can be collected and sampled from a video.

By using the apparatus and datasets, you will be able to proceed with the 3D reconstruction from 2D datasets.

State-of-the-art Technology Used by the Datasets for the Reconstruction of 3D Objects

The technology used for this purpose needs to stick to the following parameters:

Input

Training with the help of one or multiple RGB images, where the segmentation of the 3D ground truth needs to be done. It could be one image, multiple images or even a video stream.

The testing will also be done on the same parameters, which will also help to create a uniform, cluttered background, or both.

Output

The volumetric output will be done in both high and low resolution, and the surface output will be generated through parameterisation, template deformation and point cloud. Moreover, the direct and intermediate outputs will be calculated this way.

Network architecture used

The architecture used in training is 3D-VAE-GAN, which has an encoder and a decoder, with TL-Net and conditional GAN. At the same time, the testing architecture is 3D-VAE, which has an encoder and a decoder.

Training used

The degree of supervision used in 2D vs 3D supervision, weak supervision along with loss functions have to be included in this system. The training procedure is adversarial training with joint 2D and 3D embeddings. Also, the network architecture is extremely important for the speed and processing quality of the output images.

Practical applications and use cases

Volumetric representations and surface representations can do the reconstruction. Powerful computer systems need to be used for reconstruction.

Given below are some of the places where 3D Object Reconstruction Deep Learning Systems are used:

3D reconstruction technology can be used in the Police Department for drawing the faces of criminals whose images have been procured from a crime site where their faces are not completely revealed.
It can be used for re-modelling ruins at ancient architectural sites. The rubble or the debris stubs of structures can be used to recreate the entire building structure and get an idea of how it looked in the past.
They can be used in plastic surgery where the organs, face, limbs or any other portion of the body has been damaged and needs to be rebuilt.
It can be used in airport security, where concealed shapes can be used for guessing whether a person is armed or is carrying explosives or not.
It can also help in completing DNA sequences.

So, if you are planning to implement this technology, then you can rent the required infrastructure from E2E Networks and avoid investing in it. And if you plan to learn more about such topics, then keep a tab on the blog section of the website.

‍

Reference Links

https://tongtianta.site/paper/68922

https://github.com/natowi/3D-Reconstruction-with-Deep-Learning-Methods

This is a decorative image for: Comprehensive Guide to Deep Q-Learning for Data Science Enthusiasts

October 18, 2022

A Comprehensive Guide To Deep Q-Learning For Data Science Enthusiasts

For all data science enthusiasts who would love to dig deep, we have composed a write-up about Q-Learning specifically for you all. Deep Q-Learning and Reinforcement learning (RL) are extremely popular these days. These two data science methodologies use Python libraries like TensorFlow 2 and openAI’s Gym environment.

So, read on to know more.

What is Deep Q-Learning?

Deep Q-Learning utilizes the principles of Q-learning, but instead of using the Q-table, it uses the neural network. The algorithm of deep Q-Learning uses the states as input and the optimal Q-value of every action possible as the output. The agent gathers and stores all the previous experiences in the memory of the trained tuple in the following order:

State> Next state> Action> Reward

The neural network training stability increases using a random batch of previous data by using the experience replay. Experience replay also means the previous experiences stocking, and the target network uses it for training and calculation of the Q-network and the predicted Q-Value. This neural network uses openAI Gym, which is provided by taxi-v3 environments.

Now, any understanding of Deep Q-Learning is incomplete without talking about Reinforcement Learning.

What is Reinforcement Learning?

Reinforcement is a subsection of ML. This part of ML is related to the action in which an environmental agent participates in a reward-based system and uses Reinforcement Learning to maximize the rewards. Reinforcement Learning is a different technique from unsupervised learning or supervised learning because it does not require a supervised input/output pair. The number of corrections is also less, so it is a highly efficient technique.

Now, the understanding of reinforcement learning is incomplete without knowing about Markov Decision Process (MDP). MDP is involved with each state that has been presented in the results of the environment, derived from the state previously there. The information which composes both states is gathered and transferred to the decision process. The task of the chosen agent is to maximize the awards. The MDP optimizes the actions and helps construct the optimal policy.

For developing the MDP, you need to follow the Q-Learning Algorithm, which is an extremely important part of data science and machine learning.

What is Q-Learning Algorithm?

The process of Q-Learning is important for understanding the data from scratch. It involves defining the parameters, choosing the actions from the current state and also choosing the actions from the previous state and then developing a Q-table for maximizing the results or output rewards.

The 4 steps that are involved in Q-Learning:

Initializing parameters – The RL (reinforcement learning) model learns the set of actions that the agent requires in the state, environment and time.
Identifying current state – The model stores the prior records for optimal action definition for maximizing the results. For acting in the present state, the state needs to be identified and perform an action combination for it.
Choosing the optimal action set and gaining the relevant experience – A Q-table is generated from the data with a set of specific states and actions, and the weight of this data is calculated for updating the Q-Table to the following step.
Updating Q-table rewards and next state determination – After the relevant experience is gained and agents start getting environmental records. The reward amplitude helps to present the subsequent step.

In case the Q-table size is huge, then the generation of the model is a time-consuming process. This situation requires Deep Q-learning.

Hopefully, this write-up has provided an outline of Deep Q-Learning and its related concepts. If you wish to learn more about such topics, then keep a tab on the blog section of the E2E Networks website.

Reference Links

https://analyticsindiamag.com/comprehensive-guide-to-deep-q-learning-for-data-science-enthusiasts/

https://medium.com/@jereminuerofficial/a-comprehensive-guide-to-deep-q-learning-8aeed632f52f

October 13, 2022

GAUDI: A Neural Architect for Immersive 3D Scene Generation

The evolution of artificial intelligence in the past decade has been staggering, and now the focus is shifting towards AI and ML systems to understand and generate 3D spaces. As a result, there has been extensive research on manipulating 3D generative models. In this regard, Apple’s AI and ML scientists have developed GAUDI, a method specifically for this job.

An introduction to GAUDI

The GAUDI 3D immersive technique founders named it after the famous architect Antoni Gaudi. This AI model takes the help of a camera pose decoder, which enables it to guess the possible camera angles of a scene. Hence, the decoder then makes it possible to predict the 3D canvas from almost every angle.

What does GAUDI do?

GAUDI can perform multiple functions –

The extensions of these generative models have a tremendous effect on ML and computer vision. Pragmatically, such models are highly useful. They are applied in model-based reinforcement learning and planning world models, SLAM is s, or 3D content creation.
Generative modelling for 3D objects has been used for generating scenes using graf, pigan, and gsn, which incorporate a GAN (Generative Adversarial Network). The generator codes radiance fields exclusively. Using the 3D space in the scene along with the camera pose generates the 3D image from that point. This point has a density scalar and RGB value for that specific point in 3D space. This can be done from a 2D camera view. It does this by imposing 3D datasets on those 2D shots. It isolates various objects and scenes and combines them to render a new scene altogether.
GAUDI also removes GANs pathologies like mode collapse and improved GAN.
GAUDI also uses this to train data on a canonical coordinate system. You can compare it by looking at the trajectory of the scenes.

How is GAUDI applied to the content?

The steps of application for GAUDI have been given below:

Each trajectory is created, which consists of a sequence of posed images (These images are from a 3D scene) encoded into a latent representation. This representation which has a radiance field or what we refer to as the 3D scene and the camera path is created in a disentangled way. The results are interpreted as free parameters. The problem is optimized by and formulation of a reconstruction objective.
This simple training process is then scaled to trajectories, thousands of them creating a large number of views. The model samples the radiance fields totally from the previous distribution that the model has learned.
The scenes are thus synthesized by interpolation within the hidden space.
The scaling of 3D scenes generates many scenes that contain thousands of images. During training, there is no issue related to canonical orientation or mode collapse.
A novel de-noising optimization technique is used to find hidden representations that collaborate in modelling the camera poses and the radiance field to create multiple datasets with state-of-the-art performance in generating 3D scenes by building a setup that uses images and text.

To conclude, GAUDI has more capabilities and can also be used for sampling various images and video datasets. Furthermore, this will make a foray into AR (augmented reality) and VR (virtual reality). With GAUDI in hand, the sky is only the limit in the field of media creation. So, if you enjoy reading about the latest development in the field of AI and ML, then keep a tab on the blog section of the E2E Networks website.

‍

Reference Links

https://www.researchgate.net/publication/362323995_GAUDI_A_Neural_Architect_for_Immersive_3D_Scene_Generation

https://www.technology.org/2022/07/31/gaudi-a-neural-architect-for-immersive-3d-scene-generation/

https://www.patentlyapple.com/2022/08/apple-has-unveiled-gaudi-a-neural-architect-for-immersive-3d-scene-generation.html

Build on the most powerful infrastructure cloud

Contact Sales Get Started for Free

A vector illustration of a tech city using latest cloud technologies & infrastructure

Steps to Build a Multilingual Healthcare AI Chatbot

Table of Contents

Key Takeaways

Introduction

Understanding the Technical Architecture

Building on India’s Top MeitY-Empanelled Cloud

Steps to Build a Healthcare AI Chatbot

Prerequisites

Step 1: Loading and splitting the PDF

Step 2: Deploying the LLM

Step 3: Encoding the chunks using a pre-trained embedding model

Step 4: Storing the embeddings in Qdrant

Step 5: Implementing the Context Generation Function

Step 6: Generating responses

Step 7: Integrating a web interface

Healthcare Chatbot

Answer:

Next Steps

Inside the H200 Tensor Core GPU: An In-Depth Architectural Analysis

Launching and Using Pixtral-12B on TIR AI Platform: Bill Parsing with vLLM

Step-by-Step Guide to Bulk Invoice Processing Using Llama 3.2-11B

Towards a Tech Ren(AI)ssance: Insights From Jensen Huang

A Detailed Comparison of the NVIDIA H200 and H100 Architectures for Developers

Building a Llama 3.2-11B-Based RAG System with Vision Model Integration

A Complete Guide To Customer Acquisition For Startups

Image-based 3D Object Reconstruction State-of-the-Art and trends in the Deep Learning Era

The Main Objective of the 3D Object Reconstruction

A Comprehensive Guide To Deep Q-Learning For Data Science Enthusiasts

GAUDI: A Neural Architect for Immersive 3D Scene Generation

Build on the most powerful infrastructure cloud