Using AI to Improve Quality of Life For Diabetic Patients
Global Summit Creator
April 28, 2017
The world of startups is constantly moving and evolving. With the exponential growth of deep learning research and technologies in recent years, innovative new companies are often funded, acquired and transformed from startups to industry leaders extremely quickly.
MedicSen is a startup focused on developing non-invasive treatments for diabetes, utilising connected devices, machine learning algorithms and a cloud platform to revolutionise diabetic patient's lives. By applying artificial intelligence and sensor tech, the algorithm can predict future glucose levels and risky events and, according to that, give the patient medical advice and instructions for the amount of insulin they need and at what time. Since we first met the MedicSen team at the 2016 Deep Learning in Healthcare Summit in London, their team has grown and their mission has evolved. I spoke to the CEO Eduardo Jorgensen to learn more about their startup journey over the last year, and to find out what we can expect next.
What has been essential to the success of MedicSen so far?
A team committed to the idea of improving diabetic patient's way of living.
What is the biggest challenge you've faced?
For sure it has been the development of an algorithm that not only predicts the glycemic state but also learns from the patient's condition in order to customize the prognosis and advice to each person. Doing it with lean startup principles and a lack of data has been a real challenge to overcome.
We've seen the MedicSen team grow over the last year, how have new team members shaped your goals?
The new team members contribute with their own way of thinking about how to improve the living of people with diabetes. With big readjustments, especially in the technological team, we have been capable of reaching working prototypes. This wouldn't be possible without the wonderful team that we have.
Each new member, from their own personal scope and discipline of knowledge, contributes to the rest of the team to culminate together in the creation of a quality product that improves diabetic patient's day to day lives.
Can you share a success story from a patient who used MedicSen?
There is not much to tell so far, but I remember one patient telling me that he couldn't believe that an accurate prediction for his blood glucose was possible. He talked about being able to organize his day in advance, and how that would release him of additional stress and worry.
What areas of healthcare can you see being shaped by artificial intelligence in the future?
All those that need an exhaustive follow-up of the patient and in which the continuous administration of a drug is necessary in personalized doses (mostly any disease, but with different levels of benefit). All those areas that entail a control and administration of the drugs by the patient himself, will improve quality of life by freeing him of the burden of being constantly focused on drug calculations and intake.
What other technological advancements outside of AI are aiding your work at MedicSen?
Remote servers in the cloud for real time data processing, the introduction of sensors in the shape of wearable technology and new techniques for materials manufacturing will not only decrease costs but also make our business possible, because people don't like to make decisions or input data, they want things to just happen and not make decisions if they can get rid of them. Healthcare will very soon involve personalized telemonitoring and treatments, stay tuned for the big revolution.
Can you tell us about the patent-pending medical device you are developing?
It is a non-invasive drug delivery device, so it does not carry needles. In small tests we have been able to demonstrate that the drugs cross the skin with this system, and soon we will begin with the safety tests in the laboratory, at the close of the current round of financing. It is shaped as a smartpatch, comfortable and modular in order to ease dose loading. The future of drug delivery happens through the skin and without needles.
What are you working on next?
We are just about to launch the beta testing of the app in medicsen.net/1000XLaDiabetes; we are looking for 1000 patients to help improving the software and soon we will start with the safety tests of the smartpatch: non-invasive administration of insulin through the skin.
Why is it important to bring together experts in AI and healthcare at meetings and events?
The future of medical practice and health technology will be closely linked to the development of autonomous and intelligent units for the treatment of patients, therefore multi-disciplinary teams will be needed to solve the biggest healthcare problems.
How do you find MedicSen's services are used differently by patients and healthcare providers? How has this shaped the way you work?
The caregiver can access the information in real time or post-events, up to their choosing. Our aim is to ease interaction with the patient, allowing them to make contact whenever they prefer. They also can add data to the clinical record with the click of a button trough the caregiver dashboard. We believe that without acceptance from the healthcare community, the solution will not work.
Do you have any advice for startups entering your field?
Do not be afraid to take risks, the entrepreneurial world is frustrating sometimes, but also very rewarding, because you know that your project can change the lives of many people in a radical way. How has collaboration with international partners aided the success of MedicSen?
It has helped the internationalization of the company and our products, as well as giving access to markets and investors that would otherwise be impossible to achieve at this time.
How can we expect AI and chatbots to change the way we approach our health in the future?
AI and chatbots will improve the quality of the healthcare delivery in the XXIst century. AI is helpful as it can learn from the patient individually and give personalized treatments, and chatbots allow for an easy and convenient interactions with technology. We will certainly see many chatbot-like solutions in the coming years.
Learn more about the techniques and applications of AI in the healthcare sector at the Deep Learning in Healthcare Summit in Boston on the 25-26 May, taking place alongside the annual Deep Learning Summit. Confirmed speakers include Junshui Ma, Senior Principal Scientist, Merck; Nick Furlotte, Senior Scientist, 23andMe; Muyinatu Bell, Assistant Professor, John Hopkins University; Saman Parvaneh, Senior Research Scientist, Philips Research; David Plans, CEO, BioBeats; and Fabian Schmich, Data Scientist, Roche. View more details here.
Opinions expressed in this interview may not represent the views of RE•WORK. As a result some opinions may even go against the views of RE•WORK but are posted in order to encourage debate and well-rounded knowledge sharing, and to allow alternate views to be presented to our community.