Emerging Technology Profile:

The Role of Chatbots in Healthcare

Overview

Since the launch of Amazon’s Alexa, chatbots have become increasingly mainstream. The public now knows the power of automatic conversational agents that can interact with humans in sophisticated ways. Even though Blockchain may have a bit more buzz at the moment—read our Blockchain guide here—don’t let that fool you. The increasing power of chatbots suggests that the technologies underlying Alexa have the potential to transform the healthcare industry, alongside banking, retail, and e-commerce. Because of the incredible level of detail and judgment required in many medical conversations between people, chatbots are not going to replace doctors entirely, of course. However, unlike more complex technologies, even simple chatbots built by provider organizations have the potential to play an essential and growing role in the healthcare system, in areas such as patient engagement, provider assistance, and facility operations. With extensive investment in these new tools, chatbots can be quickly deployed in healthcare facilities, thanks to off-the-shelf software like IBM’s Watson API. They will save money and save lives.

If you are committed to understanding the tech changing healthcare, then you’ll want to  spend a few minutes learning the basics of chatbot technologies. How do they work? Why are they important? They will change the nature of healthcare work and how patients interact with care providers, creating autonomous systems that improve over time.

Toehold Learning is here to help you discover how the natural language processing and machine learning at the heart of chatbots will impact the healthcare system and how you can use these technologies to improve and extend healthcare and reduce costs at your facility.

What you'll find here

This section covers the basics of chatbot technology, both generally and in its potential applications for the healthcare sector. We’ll cover why you should care about this technology, discuss the basics of how it works, and outline the aspects of healthcare where the technology has potential to decrease costs, increase efficiency, and improve care quality.

Click on a topic or question to jump directly to that section or explore them to get the full picture.

1. Understand the basics of Chatbots

This section covers the basics of chatbot technology, both generally and in its potential applications for the healthcare sector. We’ll cover why you should care about this technology, discuss the basics of how it works, and outline the aspects of healthcare where the technology has potential to decrease costs, increase efficiency, and improve care quality.

Why should we care about Chatbots in Healthcare?

Because the healthcare industry already relies heavily on interactions and conversations between individuals, and the use of sophisticated software, there is massive potential for this technology to become deeply embedded in the medical system. While many people may think of chatbots as simple digital personal assistants, these new tools offer tremendous potential to transform a wide range of work and interactions in the healthcare system. Conversations that once required two people may now occur between a chatbot and a person. Complex software can be simplified with the use of a chatbot, allowing patients, providers, and payers to have a conversation with the computer or app instead of having to navigate complex screens and systems.

"It's a simple concept, yet it's very powerful in its impact. It is about taking the power of human language and applying it more pervasively to our computing."​

Satya Nadella, CEO at Microsoft Tweet

Chatbots reduce the labor required to accomplish both small and large tasks. Recent advances in artificial intelligence have radically expanded the opportunities for the healthcare sector to create intuitive conversational agents as the interface between humans and machines. While chatbots may be good at answering simple questions—those not requiring a detailed medical history—they also can include potentially lifesaving and cost-saving interactions between patients, providers, payers, and the medical system. Chatbots won’t just interact with patients; they will interact with and improve the efficiency and effectiveness of the people working in healthcare.

What are Chatbots, anyway?

Chatbots are a service that offers human-like interaction through voice or text between a person and a piece of software. Put simply, a chatbot changes the interface between a person and a machine to mimic the types of conversations generally conducted between people, making using the program more comfortable and intuitive for the user.

Let’s look at an example. If you wanted to purchase a product online, usually you would go to a website, browse through the items, make a selection, and then buy the item. With an e-commerce chatbot, you would have a conversation with a friendly chatbot agent who would guide you to the things you might be most interested in, based on your previous shopping preferences and answers to questions. With a chatbot, the experience would closely mirror the conversation you might have with a customer service representative in a retail store. The e-commerce industry is extremely excited about the technology—but the healthcare industry is too.

The possibilities for implementing chatbots in healthcare are seemingly endless. Chatbots can transform any interface between a person and a piece of software into a more conversational interaction that is familiar to a person. Rather than navigating a sophisticated piece of software, the user engages with a chatbot that effortlessly guides them to what they are looking for or trying to accomplish. The user saves considerable time and energy, in two ways: by reducing time spent learning new software, and by reducing the work of filtering results to find the best option.

Thanks to recent advances in artificial intelligence (AI)—through machine learning and natural language processing—chatbots have rapidly become incredibly powerful. They can answer complex questions, learn from past interactions, and process and sort vast quantities of data. These advances mean that AI-powered chatbots can handle conversations that were once only possible between two people: a customer and an expert. Chatbots can do much more than merely schedule an appointment or tell you when the doctor’s office opens—though these are the low-hanging chatbot fruit that will sweep the healthcare industry first.

"Put simply, a chatbot changes the interface between a person and a machine to mimic the types of conversations normally conducted between people."

How does it work?

Using a chatbot is a lot like interacting with a human working at a help desk. Rather than the help-desk employee responding, it’s the chatbot that answers. That’s the fundamental principle behind chatbot technology. (Of course, things get a little more complicated if you pop the hood and take a look at the engine powering the beast that is the modern AI chatbot.)

The simplest chatbots don’t use AI, however. They function on a set of structured questions and answers. Simple chatbots harness the power of a smaller knowledge base and respond with correct output when given a specific set of instructions or questions. The humans must ask the questions in a way the chatbot’s programming understands for the correct answer to result. These simple chatbots can reply with “I’m sorry, I didn’t understand the question” when the user doesn’t ask the ‘right’ question—a response that does not suffice for most medical-based applications of the technology.

More advanced chatbots use the power of artificial intelligence to let users ask complex questions and receive a much broader and more precise array of answers. Both machine language learning and natural language processing, which we will look at separately, have given simple chatbots extraordinary new power.

Machine learning programs analyze previous datasets using sophisticated algorithms to develop the capacity to make predictions and return more nuanced output. Many of the algorithms used in chatbots stem from recent advancements in artificial neural networks, a type of machine learning inspired by the connection between neurons in biological information processing. This biologically inspired network provides the algorithm with the structure needed to process even non-linear relationships between data points—critical functionality considering the complexity of the many large data sets involved in creating chatbots with a high degree of predictive output. As the machine learning algorithms process data, the system learns and improves over time, increasing the accuracy of the results returned to the user.

While machine learning takes care of the analysis and prediction of data responses, natural language processing (NLP) focuses on putting human language into a format that the software can use to interpret a question accurately. NLP uses the structure of language, idioms, and pattern recognition to determine the user’s intent. The NLP identifies what the user wants to do, which entities are in question, and asks any follow-up questions needed for clarification before processing the request. Not all bots use natural language processing, however, because the technology isn’t yet fully capable of handling all adult-level linguistic elements. In the right application and field, the chatbots based on NLP can create a genuinely conversational feel for the user, as opposed to making the user feel like the bot is just filtering through a set of pre-defined menu items.

Between simple and complex chatbots, the variety of different kinds of chatbots available to the healthcare industry is immense. Some situations require machine learning and natural language processing, whereas others don’t. For example, a chatbot designed with NLP and machine learning would do a lousy job at helping patients automatically schedule their next appointment, tending instead to get into conversations with patients on a range of topics. Simple chatbots are best in this setting and can create an inviting interface for users asking specific questions or working to accomplish specific tasks. In contrast, more complex chatbots can harness the power of vast knowledge bases, making that information accessible in a way it has never been before. Given the rapid progression of machine learning and natural language processing, the future of chatbot technology is bright. Chatbots will change healthcare, simplifying and automating processes that once were so complex that they required people to perform them.

What are the core features?

Because chatbots can be elaborate or simple, machine learning and natural language processing are not necessary features in all applications of this technology. Still, all chatbots share five features that shed light on the applicability of this technology to healthcare. Chatbots are:

  1. Asynchronous—Because they are automated, the interaction can take place at any time and across platforms that allow for more flexibility than human-human interactions do.
  2. User-centric interface—This technology creates an interface between humans and software that mirrors human personal conversations, making it more user-friendly.
  3. Structured—With complex natural language processing and machine learning, designers structure chatbots to sort the questions, answers, and interactions with users.
  4. Knowledge-based—Chatbots utilize structured or unstructured knowledge bases to provide answers for and interact with users.
  5. Scalable—Because they are autonomous, chatbots can scale rapidly to interact with increasingly large groups of patients and providers, without the need to add more staff.

These core features of chatbots make them extremely valuable for a wide variety of businesses and industries where the automation and increased efficiency of interaction with customers are essential. The healthcare industry is no exception, given the extensive patient, provider, and business interactions that could be improved or extended with this technology.

Why do Chatbots matter for the world?

Chatbots represent a significant change in the way that humans interact with software technologies, and as such, they offer a variety of benefits for customer engagement. Fundamentally, human interaction with technology has exploded with the invention of the internet, mobile technologies, and messaging apps. Because so many people use smartphones already, they are familiar with the kind of technologies and software that can easily integrate chatbots into their lives, creating a more user-centric, asynchronous interface for humans with various autonomous technologies. Chatbots will improve user engagement with technology and dramatically reduce the labor force required to serve patients, by reducing the complexity of the human-machine interface and automating conversations that once needed humans on both sides.

By serving as conversation-based personal assistants, chatbots also offer the possibility to put powerful AI tools in the hands of millions of people. This technology provides the potential to embed powerful tools for accessing information and solving problems, which would otherwise be considerably more challenging to deploy.

Still unclear about the basics?

Understanding the richness that is chatbots can be a challenge. Check out the two videos below for more information.

Oracle created a solid explainer video that introduces chatbots and suggests their product direction.

This video covers the basics of what chatbots as part of a course on how to build bots. 

Why do Chatbots Matter for Healthcare?

With the basics of chatbots in mind, hopefully the range of potential applications for this technology in the healthcare industry is starting to become clear. The benefits of the technology—autonomous, conversation-based, knowledge-based—give it a wide variety of applications for the healthcare industry. As chatbot entrepreneurs build out this technology into existing messaging platforms and create new applications for specific use cases, the use of chatbots in healthcare is proliferating.

There are already numerous uses of chatbots in healthcare people seeking medical information, like WebMD, but many of those do not engage patients directly with healthcare systems. Perhaps the most extensive use case for chatbots in healthcare focuses on patient engagement, creating new ways that patients can engage with the medical system. The second most popular use case is creating personal assistants and secure solutions for providers to efficiently and effectively find information on patients. Beyond these two use cases, chatbot technology has the potential to streamline and improve the business side of healthcare facilities, including medical billing, human resources, medical research, and environmental health and safety.

Below, we present some of the most promising use cases for chatbots.

Patient Engagement

Booking appointments & reminders

Because of the simple set of interactions required, chatbots excel at providing and automating the appointment booking and reminder process, autonomously extending patient engagement.

Emergency care & understanding the care needed

When integrated with specialist care such as emergency departments, chatbots can react instantaneously to provide help to patients in a potential emergency. Depending on its functionality, a chatbot can give information on how patients should respond in an emergency situation, or even detect that an emergency situation is in process. Furthermore, chatbots can help patients determine whether they need to see a specialist, and, if so, what kind.

Patient information requests

Before consulting with physicians, patients interacting with a chatbot may be able to quickly ask and answer preliminary questions and access necessary medical information. With the ability to view their healthcare records, patients would be able to confirm and access information in their medical records in a seamless manner and take more control over their health information.

Follow-up care

With the 30-day readmit rule, hospitals and care facilities are becoming increasingly responsible for following up with patients, to ensure that they are taking their medications on time and continuing to engage in the ongoing behavioral changes required to become healthy again. The personalization of chatbots can make it easier for patients to interact with a friendly and fun automated assistant that helps them continue their care plan, reducing penalties for care facilities and improving treatment.

Mental health

Chatbot designers are building a wide range of new applications for mental health patients. These allow patients to engage in the care process between appointments continuously. With excellent natural language processing systems at their core, these new psychological health chatbots can provide emotionally sensitive, empathetic answers in a way that mimics human interactions. Because the bots are always online, patients have access to the care they need whenever they need it, not just when they can get an appointment.

Voice-activated assistants

While not all chatbots are voice-based, the technology offers a seemingly endless number of ways to create voice-activated assistants to provide information and assistance to healthcare patients suffering from lack of mobility or vision problems. The potential of chatbots for senior care is enormous, as the technology would create an easy way to continually remind patients of their daily activities and to take their medications.

Provider Engagement Applications

Patient information for providers

Chatbots could make the process of accessing patient information considerably easier for providers than is presently possible with traditional EHRs. By automating and creating a simple system for providers to use, they could more easily access the components of the medical record that are immediately relevant to the particular case. Imagine a doctor asking a chatbot “What were Mr. Jones’ highest blood pressure measurements in the past 24 hours?” without having to log out of the EHR system to find the answers.

Clinician medication & information upload

The chatbot interface for direct care providers could create an easy way for them to upload and file prescriptions and other medical notes securely, quickly, and on the go. Chatbots using intelligent voice dictation would save considerable time and reduce errors in submitting notes. When designers embed the chatbot into an app, doctors could also upload photos of wounds or other medical images directly after answering questions and confirming the patient’s information—a process that used to require a medical photographer and considerable time.

Medical learning assistants

Doctors face a tremendous challenge when sifting through the impossibly large and ever-growing body of facts about their medical specialty. The sheer scale of data is immense, making it impossible for doctors to stay current with all of the relevant information at any moment. Chatbots offer a potential solution to this challenge, by enabling care providers to quickly and easily locate and retrieve specific medical information about a disease or treatment plan.

Facility Operations

Customer engagement & new patient leads

Facilities can develop chatbots to extend their interaction with their service community, bringing in new patients for their doctors. By employing chatbots and engagements through common application platforms like Slack, Facebook Messenger, and Kik, facilities can provide information to potential clients on medical care and questions, creating a new way to engage with potential customers and guide people to their facility.

Other facility applications

Chatbots hold tremendous potential to automate and streamline communications between various departments and their customers. Taking the human resources knowledge base and creating a chatbot application for employees to use is just one of many possible examples where the functionality of chatbots can improve user experience and create savings for the healthcare facility or system. With chatbots, the sky is the limit, and facility operations managers should evaluate the application of this technology to their domain within the organization.

To summarize

The list of possible use cases for chatbots in healthcare is immense, and entrepreneurs have just scratched the surface. The list of use cases will expand tremendously in the coming years, with facilities, providers, and patients all targeted for the improvements brought by automation and conversational agents.

2. The major players

The Chatbot landscape is changing rapidly, with dozens of new companies emerging over the few three years. Dive into the new companies and the people that are shaping the future of Chatbots use in healthcare. 

Exciting healthcare companies using Chatbots.

Patient Engagement

Sensely

An app designed to connect patients and healthcare plan members to the information and health resources they need.

SimplifiMed

A simple autonomous chatbot designed to provide reminders to patients and improve staff productivity

Conversa

Using Conversa’s innovative and scalable "Conversational AI" technology, healthcare organizations deliver automated, personalized doctor-patient conversation experiences that lead to more informed and meaningful patient relationships, effective population management and, ultimately, better clinical and financial outcomes.

Logo for Gomo Health's Chatbot tools

GoMo Health

GoMo Health uses a wide range of mobile health tools (including chatbots) to improve patient acquisition, retention, and outcomes. They have apps for care facilities and community education solutions that support the full continuum of care that present chatbots as a bridge between patients and caregivers.

Full Healthcare Solutions

Babylon Health

Using Conversa’s innovative and scalable "Conversational AI" technology, healthcare organizations deliver automated, personalized doctor-patient conversation experiences that lead to more informed and meaningful patient relationships, effective population management and, ultimately, better clinical and financial outcomes.

iCliniq

A collection of 2500 healthcare providers, iCliniq deploys a mixed chatbot and human services solution to provide answers to medical questions to thousands of people daily.

Provider Engagement

MedXnote

This app is designed to connect medical teams and integrate messaging and team connections in a fully secure platform.

Avaamo

A larger platform for improving healthcare operations, provider engagement, and operations.

3. Further resources

If you want to dive deeper into the healthcare Chatbot world, we’ve got even more resources for you to use to expand your inderstaanding. Explore these articles, and courses to find out even more about the future of Chatbot’s use in healthcare.

Excellent articles on Chatbots in healthcare

Check out these outstanding articles on Chatbot deployment in healthcare. Click the title to access the full article.

Author: The Medical Futurist

There are a wide range of healthcare chatbots on the market. Sifting through what’s available can be a challenge. The medical futurist has selected 12 of the best bots that are changing healthcare. These are well worth looking through to get a sense of where the market is heading.

Amazon has an early lead, but technologists can apply lessons from cutting-edge providers to Apple Siri, Google Home and Assistant, and Microsoft Cortana.

Chatbots, virtual assistants, and ambient computing devices may offer healthcare organizations a more accessible entry point into the world of artificial intelligence.

A quick article discussing the features of chatbots that make them beautifully suited to deal with low patient engagement

Chatbots and virtual personal assistants are helping to evaluate symptoms, manage medications and monitor chronic conditions.

Courses on Chatbots

While there are limited online course offerings on the role of Chatbot technology in healthcare—something Toehold Learning is working on—

Author: IBM

Audience: People wanting to understand how to create empathetic chatbots.

Length: 3 hours

Price: Free

Imagine if you could build a chatbot that understands how someone is feeling. Your chatbot could play music or make suggestions to help them feel happier. During this mission, you will:

  • Learn how to design a dialog for your specific users by using Design Thinking
  • Play with a chatbot and learn about the three basic components of Watson Conversation: intents, entities, and dialog
  • Build a simple chatbot using the Watson Conversation service
  • Analyze tone and emotions in the text by using Watson Tone Analyzer and brainstorm ways you can use the service to improve your chatbot

Authors:  Sachin Bhatnagar

Audience: Developers and professionals who understand coding in JavaScript and NodeJS

Length: 8.5 hours

Price: $59.99 (sometimes discounted)

Learning outcomes:

  • Understand and learn about conversational interfaces and agents
  • Understand the role of Natural Language Processing and AI
  • Build chatbots using Motion ai
  • Use Wit.ai for building chatbots
  • Deploying a bot on the Facebook messenger platform
  • Deploying a bot on the Facebook Messenger Platform
  • Deploying a bot on Slack
  • Build conversational agents or chatbots using JavaScript & NodeJS
  • Build a Chatbot with zero coding with Chatfuel
  • Build and use your interface to Facebook Messenger
  • Build chatbots using API ai (DialogFlow
  • Understand and use Regular Expressions (RegEx) for building basic Chatbots
  • Deploying a bot on Skype
  • Deploying chatbot apps on cloud platforms such as Heroku

“Welcome to the most comprehensive and complete bot developer course. Learn concepts, tools, and techniques that you will need to build fully functional chatbots for business and enterprise.

Chatbots are computer programs that can interact with humans through a simple conversational interface. They are designed to simulate interaction with another human. Couple this with the fact that more than 90% of smartphone users spend most of their time in messaging apps such as Facebook Messenger and you have an excellent opportunity to build highly interactive chatbot based services that can empower business and enterprise like never before.

In this course, we will go from zero to pro as we build multiple chatbots using a variety of techniques and platforms. We will explore chatbot platforms that do not require you to write code, and all the way to code intensive chatbots that can be built for specialized scenarios.”

Subscribe to our newsletter

Don't miss any updates when we add new content.
Scroll to Top