Imagine: AI technology to pioneer patient-led healthcare

Share:
LinkedInFacebookTwitterEmail

 

 

 

 

Adrian, Aleksander and Mélanie work on Imagine in LEO Innovation Lab to help patients suffering from skin conditions take back control of their disease.

As the number of patients in need of specialist care rises, the skin tracking app Imagine seeks to empower patients and augment doctors in their work

It’s no secret that the healthcare landscape of tomorrow will be drastically different to that of today. As the burden of chronic diseases rapidly increases worldwide alongside a growing shortage of healthcare professionals, patients who need ongoing support to manage their health condition will be forced to look to other means for their care. In fact, by 2035, there will be an estimated shortage of 13 million healthcare workers to meet the needs of patients living with chronic diseases. The future of healthcare, therefore, calls for patients to take on a more active role in the care of their own disease. It calls for more patient-led healthcare. In LEO Innovation Lab, we took on the challenge of meeting this demand by developing the mobile application Imagine: a ground-breaking tracking tool to help patients monitor and track changes in their skin condition.

 

Bridging the gaps in patient care

Because of the time it usually takes to see a dermatologist, the skin issue a patient has at home may not be the same as the skin issue they present during a consultation. Flare-ups and rashes don’t wait for a specialist. They fluctuate and may be gone when a patient finally goes in for a consultation, only to return later. The human brain can also be all too prone to relativising the past, meaning a patient might not be able to recall the state of their condition some weeks earlier accurately.

What patients need is a tool they can integrate into their everyday lives – something to bridge the gap between consultations with quality data from which doctors can get a better overview of the skin condition that goes beyond the immediate contact with the patient. This is where Imagine comes in. By empowering patients with a systematic means of monitoring their skin disease, it allows both patients and doctors the ability to track and see changes that are often gradual, correlated to symptoms and triggers. This is powerful because people living with chronic skin conditions now have a single ecosystem for their information.

From the physician’s point of view, Imagine enables a better evaluation of patients based on a more holistic overview of their condition. Instead of relying on the patient’s own recollection of the progression of their skin condition, Imagine allows for more objective and aligned comparisons, and helps patients to better structure and share information about their disease and overall well-being. This, in turn, makes the consultation process more efficient, by establishing a platform where doctors and patients have the relevant imaging and patient data gathered in one place.

The impact of this is already manifesting itself among patients. Our initial findings indicate that when people use Imagine, there’s an uplift in their well-being, simply from not having to worry so much about their condition because of the personal insights they gain from the app. Imagine takes on a number of burdens such as remembering what the skin looked like at a specific point in time, how the symptoms have changed and when to take photos to document the condition. This means that the patient, from a more informed standpoint, can focus on maintaining a lifestyle to improve and take back control of their skin disease.

 

From tracking to AI-driven diagnostics

Though the first phase of Imagine was to create an effective tracking tool for patients, our ambitions have always been much bigger. That’s because tracking alone won’t tell you what triggers your symptoms, or how certain interventions are working. That still requires some detective work. And so, to take the burden of analysis off the patients, we’ve since used our data to work towards developing machine learning algorithms, that will eventually be able to accurately diagnose skin diseases, first time around.

Due to the complexity of dermatology with the existence of more than 2,000 different diseases, there’s currently up to 50% misdiagnosis rate for dermatological conditions among general practitioners during the first consultation. In contrast, for the pictures users of the Imagine app have been sending in, our AI-powered diagnostic tool was capable of diagnosing psoriasis with 91% accuracy, which is on par with experienced dermatologists remote-diagnosing based on the photos.

As this technology becomes more sophisticated, we foresee a future where Imagine works in unison with doctors to accurately diagnose the patient that steps into the consultation room. This will not only enable doctors to provide better patient outcomes early on, but also to make better use of their time, by removing the burden of unnecessary follow-up consultations due to initial misdiagnosis or by avoiding that patients end in a limbo of treatments that don’t work.

Following Imagine’s success with psoriasis, we’re now working to improve our diagnostic accuracy with other skin diseases including prevalent chronic conditions such as atopic dermatitis, rosacea, and acne but also mimickers that are important to distinguish.

 

What does the future hold?

In the short-term, Imagine will support doctors not only to diagnose patients but also to triage prior to a consultation taking place. The effects of this will be two-fold: Firstly, rather than having to wait several weeks (or months) for an appointment, we foresee a future where textbook cases of skin diseases are immediately diagnosed and funnelled into a data-driven care plan. Secondly, it will help to optimise the available time within a consultation where dermatologists can work with the predicted diagnosis data in the Imagine app, and spend more time on ensuring that the patient understands the implications of their disease and the importance of treatment adherence. Quality of life of the patient thereby becomes a more central area of focus for the physician to evaluate not only the progression of the disease but also patient well-being – an area which requires time.

In the long-term, we hope to establish a platform to connect patients and doctors in a more seamless way. This would facilitate quicker feedback loops with physicians to determine whether a treatment is actually working but also allow doctors to postpone, cancel or replace redundant follow-up visits for well-treated patients.

 

Paving the way to inclusive healthcare

The goal of Imagine is ambitious. We see the future of the Imagine technology as having the potential to increase the accuracy of diagnosing skin diseases and predict treatment outcomes. Ultimately, our goal is to implement Imagine to reduce the cost of chronic diseases, while improving patient care for all people, irrespective of their socioeconomic background or access to healthcare. Central to our work is the ambition to empower patients to become active participants in their healthcare plan rather than passive healthcare recipients – and to make expert-level dermatological disease management available for everyone. The future’s bright, if you dare to Imagine.

By Adrian Young-san Roessler, Head of Imagine, Aleksander Eiken, Medical Advisor, & Mélanie Bourlioux, Product Manager of Imagine

 

About Imagine

Imagine is a mobile app that helps patients stay on top of their skin disease. By tracking and documenting their condition, they get a more detailed understanding of their skin and symptom changes, treatment effectiveness and potential triggers.

For physicians, the resulting documentation can provide a more complete overview as a basis for the next follow-up consultation.

Imagine is currently breaking new ground within skin diagnosis powered by artificial intelligence. Through the smartphone camera, the sophisticated AI-algorithms currently diagnose psoriasis with an accuracy of 91%. This is on par with experienced dermatologists.

 

RECENT BLOGPOSTS