In Digital Health, the remote monitoring of patients and the monitoring of chronic diseases is one of the frontlines with the greatest interest and activity. Diabetes is undoubtedly arousing the most innovative and entrepreneurial activity worldwide.
Real-time monitoring of the patient’s glucose values, assistance to the patient so that his lifestyle allows him to live with his disease optimally, and ability to detect alerts that allow preventive actions, are some of the goals of these initiatives in Digital Health around Diabetes.
In Spain we have several examples focused on the improvement of diabetic patient care, and today we are pleased to have one of the most prominent.
We are accompanied by José Carlos Montesinos, CTO and Co-Founder at Medicsen, one of the first companies that have joined Nubentos to scale their Artificial Intelligence API for Diabetes.
Q: Hi José Carlos, thanks for accepting our interview. Tell us a little about yourself, your career and your role in Medicsen.
A: I studied the Degree in Telecommunications Technology and Services Engineering at the Polytechnic University of Madrid and subsequently studied the Master in Biomedical Engineering in the same faculty.
During my studies I did internships in electronics departments of the same university and in the master I could enjoy a stay at the Infanta Leonor Hospital, so that my world has always been between technology and medicine, since they have always been fields of great interest to me.
Medicsen was born from an idea that Eduardo Jorgensen, our CEO, had to improve the lives of people living with diabetes, since the technology they use to control their disease is far from the level they could reach based on to the existing technology today.
We met 4 lifelong friends to see if the idea was viable, when I was still finishing my degree studies, and from there we started working on it.
As a CTO, my role is to manage everything related to the technology we use in the company, from planning and development, to its subsequent implementation and review, taking care that everything is aligned with the company’s general strategy.
Q: What is Medicsen?
A: In Medicsen we are developing an intelligent system for the management of chronic disease treatment, starting with diabetes.
This all started when Eduardo was in the office and one of his patients, an 11 year old girl, refused to use the insulin pump she needed for her diabetes, which left her in shock.
He began to investigate and realized that, even with current technology, patients continued to have a low quality of life because the treatments are uncomfortable and not very intuitive.
Because of this, only 1 in 3 people achieve the goals of their treatment to avoid complications and, with more than 400 million people with diabetes in the world, this generates huge costs both economically and socially.
So we decided to solve these problems by designing what we call a non-invasive artificial pancreas, which is composed of software (an application and a predictive learning algorithm), and a physical device, which we have called an intelligent patch, that administers the drug non-invasively, that is, without needles.
We have designed the system so that both parts, software and hardware, can function completely independently of each other.
Q: What can your software contribute to the health ecosystem?
A: The objective of our software is to provide new useful tools for better management and control of chronic diseases.
Q: Many innovations and startups are emerging around diabetes. What would you highlight as the differential value of your solution?
A: It is true that there are many companies that are focused on helping people living with diabetes.
Our differential value is that we address the entire cycle that involves a chronic disease, such as diabetes.
Some companies offer only applications that act as a daily diary of the disease, with general advice about it. We customize the content based on the user’s individual profile.
Others focus only on glucose or medication data. We additionally emphasize diet and exercise, as they are crucial factors for proper disease management.
Others offer to predict hypo and hyperglycemia in the near future, but they are really based on trends. We use our machine learning algorithm, which gives us high accuracy, using all the aforementioned data.
And, finally, we close the whole cycle with a non-invasive patch of drug administration, without needles, controlled by our algorithm for the control and calculation of the dose in real time, something that no other company can offer for the moment in the market.
Q: Could you explain to us exactly how the Artificial Intelligence API that you have developed for the diabetic patient works?
A: Behind our API is our machine learning algorithm for predicting future glucose values.
We use the values of the different parameters that reach the API: glucose, food intake and insulin administered, to predict the future values of the glycemic curve of the user to which these parameters belong.
These values are broken down by hours and returned to the user of our API.
Q: You have just published the Medicsen API in Nubentos. What kind of consumers can be interested in using your API?
A: This API can be useful for diabetes research, since we provide a tool that allows you to know more precisely how certain changes in the habits and medication of a person living with diabetes can affect the variations of their curve of glucose.
Likewise, it can help these researchers or some companies with the training and development of insulin bolus calculation algorithms.
It can also serve as a complement to applications or support systems for patients with diabetes, or to customize the content of these applications and systems based on the current and future state of their users.
These are some of the possibilities offered by our API, but we are sure that consumers will be able to get more out of this tool.
Q: As you know, consuming your API through Nubentos allows your customers to use the free tools included in our platform to manage their developments and optimize their costs. What importance does it have for you to offer these additional features to your customers through Nubentos?
A: Nubentos allows us to provide our customers with the additional functionalities that an API must offer, apart from the main service, such as these management tools, because, seen from the client’s point of view, it provides greater comfort and control in the use of the API, reducing complexity in initial deployment and subsequent management tasks.
In addition, this helps us a lot, allowing us to focus our development on the core of the API, without worrying about such details that may be more secondary to us.
Q: What are your plans for future versions of the Medicsen API?
A: We have many ideas for the next versions we will deploy. Some of the main ones are:
- Add sport as an additional parameter of the predictive algorithm.
- Development of an endpoint that returns an activity or exercise recommendation, based on the current and future state of the user.
- Development of another endpoint that offers personalized diet and exercise plans based on the analysis of the users profile and their objectives in the medium and long term.
- Offer a bolus calculator that uses machine learning algorithms based on user parameters and their future status.
Q: How would you rate your experience while publishing the Medicsen API in Nubentos?
A: I think it can only be described as positive.
From the moment we contacted with Nubentos for the first time, all the Nubentos professionals I have worked with have been guiding and helping us throughout the process until the moment of the API publication.
In addition, the Nubentos platform is really easy to use, allowing us to have the API published in their store in a matter of minutes.
Q: What goals do you expect to achieve with this collaboration?
A: With our collaboration with Nubentos, we expect that our tools will reach a wide sector in the health ecosystem, as well as offer a better service to our potential clients, thanks to the Nubentos platform.
Q: Finally, we would like you to address to those companies that are considering publishing their API through Nubentos. What would you tell them?
A: I would tell them to do a test by deploying their API using the Nubentos platform, so they can see how simple and fast it is to have the opportunity to monetize the APIs they have developed, always having the assurance that there is a team behind to ensure that the infrastructure through which API users interact always works properly.
And additionally, with reports and statistics that allow you to evaluate the use that the API is having and with the help of Nubentos to promote the service.
Q: Thank you again José Carlos for your time and a pleasure to have shared with you this interesting talk.
A: Thank you. The pleasure was all mine.
It is clear that Artificial Intelligence has much to contribute in the monitoring and improvement of the quality of life of diabetic patients. Medicsen is leading that way and in Nubentos we are proud that they have chosen our platform to scale and enhance their solution.
The Medicsen API is published and available in our API Store, where you will find all the information, resources and tools to test the API, integrate it into your developments and put it in the hands of your end users, with total control of your development costs.
Thank you for reading and we’ll be back soon with more interesting interviews.
Stay tuned and subscribe to keep informed!