From early disease detection by medical chatbots, to prescribed video games as therapy, AI in Healthcare is transforming the patient experience. In this blog we’ll guide you through successful healthcare AI use cases and show how they have improved the patient journey.
In our recent blog post, ‘Five principles of applying AI in Pharma and Life Science,’ we laid out the best practices and cultural shifts that all companies should be encouraging today if they want to be successful, early adopters of AI.
We then took a deep dive into an emerging machine learning technique in our post, ‘Federated learning: Healthcare AI that doesn’t risk patient data privacy,’ that’s showing promise in helping Pharma and Life Science companies get deeper insights out of highly sensitive patient data.
In today’s post, we’d like to focus on the most important stakeholders involved in the digital transformation of Pharma and Life Science: ourselves, meaning patients. More specifically, we’ll be focusing on two critical phases of the patient journey that are being revolutionised by AI: pre-care and diagnosis, including real-life examples of recent projects in these spaces.
1. Pre-care: Healthcare AI Use Cases for Early Disease Detection and Wellness Management
It’s hard for a healthcare provider to have an overview of every single patient round the clock. It’s arguably even harder for every patient – or potential patient – to determine whether their symptoms require a clinic visit.
For these reasons, patients showing early symptoms of diseases – or very gradually worsening conditions – have tended to fly under the radar of healthcare providers until things get a lot more serious. Fortunately, AI in healthcare is about to change all of this.
By playing the role of a patient’s 24/7 personal health advisor, while being able to interpret sets of symptoms with accuracy and make smart, validated recommendations on the patient’s next best moves, AI applications at the pre-care phase of the patient journey are going to transform patient care for good.
Medical AI Wearables
Within the last decade we’ve seen wearable devices like FitBit and Apple Watch take the world by storm. Today, wearables on the market range from wellbeing devices that gather a wide range of patient-generated health data such as user’s sleep-wake cycles and activity levels, all the way to medical grade, clinically proven devices which can track ECG data and be applied in clinical studies.
In fact, Google now uses FitBit data to power its new cloud healthcare API – combining healthcare data standards like FHIR, DICOM, and HL7 with machine learning. The aim is to make it easier to combine individual exercise data with electronic medical records, so that everyday movement and workout data gathered by FitBits can be integrated into owners’ patient profiles, potentially unlocking new insights into conditions like diabetes and hypertension.
At Zühlke, we recently supported the development of one of the latest breakthroughs in AI-powered wearable technologies, the fertility tracker AVA. The wearable collects data 25 times per second while users sleep, then analyses the data to pinpoint fertility in real time using four different algorithms. Although the main use case of AVA is currently fertility tracking, the technology involved has the potential to support AI-driven extension to completely new indication areas and business fields.
Medical AI Chatbots
One very exciting example of an AI application in this space is actually one of our own. We recently partnered with a leading medical device manufacturer – the world leader in hearing aids – to build a revolutionary chatbot system that optimises the hearing experience through AI.
The goal was to improve the hearing experience for users in all kinds of acoustic situations. It works like this: based on the patient’s answers and their unique hearing environment, the system we’ve created suggests optimal settings and can even apply them directly to the user’s hearing aid. Over time, the system then learns from the patient’s – and the community’s – feedback via an AI algorithm.
The project has been a success, and thanks to the agile methodology we established with the company, we managed to go from ideation to execution in under 12 months.
Another example here is a very different kind of chatbot – one that supervises your daily life: Amazon’s Alexa. So far, Amazon’s digital assistant Alexa has mostly been used to play music, inform people about the weather and remind them of appointments. In the future however, there’s potential for the technology to play a part in virtual disease detection, monitoring and consultation at home. We wrote a post all about the potential of Amazon AI healthcare.
2. Diagnosis: Healthcare AI Use Cases for Making Smarter, Faster, Safer Decisions About Patient Care
Medical data is a perfect candidate for machine learning applications – mainly because a lot of the data involved is image or text data, where machine learning has made tremendous advancement in recent years –. It’s why AI in healthcare has the potential to revolutionise medical diagnostics, such as through clinical decision support systems (CDSS) – a use case that’s trending in Pharma to help companies engage better with direct clients: care providers.
For example, Duodecim EBMEDS CDSS – a system developed to optimize quality of care by promoting evidence-based decision-making – was created less than ten years ago, and is now being used by more than 60% of Finnish public health care doctors.
So why hasn’t every Life Science and Pharmaceutical company involved in diagnostics jumped at the opportunity to build AI-powered solutions? Well, because in order for medical AI to work in a diagnostic context it has to be medically compliant and transparent – which is of course a good thing – however it’s also a very complex challenge from the perspective of traditional software development.
At Zühlke, we recently helped one of the largest Pharmaceutical companies to overcome this challenge. Our partner wanted to provide physicians with a diagnostics application that could support the differential diagnosis of specific subclasses of diseases. However, since the recommendations of this software would directly affect the treatment of real-life patients, it had to be implemented in an entirely safe and compliant way.
To deal with this challenge we put together a team of experts covering machine learning, data engineering and software development as well as regulatory experts, who performed the operationalization of the existing AI research prototype. In just six months, we created an AI healthcare application for diagnostics that was ready for clinical validation.
The result? An application that supports physicians – sometimes referred to as an “expert-in-the-loop” – not only with an accurate recommendation for diagnosis of disease, but with something called “XAI”, or “explainable AI”. This is the ability to fully explain how a final result came about for each physician and patient touched by an AI model – a huge triumph for building trust and transparency.
Want to Know More About Building Powerful AI Chatbots and Diagnostic Systems?
We’ve only just scratched the surface by two trending application areas of AI in Pharma and Life Sciences that’ll contribute to the digital revolution of the patient journey.
If you are interested in more details and additional context, this is exactly what we have talked about at this year’s House of Pharma event on September 2nd, 2020. Our colleague, Dr. Stefan Weiss, has presented successful, marketable healthcare AI use cases that span the patient journey, as well as their impact on existing business models and the wider healthcare ecosystem.
If you are interested in the slides of the workshop or if you want to secure a live brainstorm with our experts on building transformative digital patient experiences with Zühlke, click the link below:Let us talk about AI in Healthcare
Dr. Stefan Weiss ist Lead Business Innovation Consultant bei der Zühlke Gruppe und verfügt über breites Hintergrundwissen in den Neurowissenschaften kombiniert mit einer breiten Expertise in Geschäftsmodellen und Innovationsmanagement. Vor seiner Zeit bei Zühlke, gestaltete Stefan die Zukunft von Healthcare und Life Sciences im Innovationszentrum der Merck KGaA aktiv mit. Mit Leidenschaft treibt er die Digitalisierung der Pharma- und MedTech-Industrie mit Fokus auf innovative Geschäftsmodelle voran, bei denen er seine wissenschaftliche und wirtschaftliche Expertise optimal anwenden kann. Bei Zühlke erweitert er die technische Exzellenz um domänenspezifisches Know-How und stärkt damit die Partnerschaften mit Pharma- und MedTech-Kunden.