Roman Chernyshev, the leader of Healthcare & Life Sciences practice, shares his view on medical wearables, “smart artificial limbs”, conservatism of the industry and Big Data in medicine.
- Who are the leaders in the medtech market - start-ups or corporations?
- It's quite a difficult question to answer. Medicine is very slowly adapting to new technologies due to the inevitable conservatism. There is huge potential here, a collection of new ideas and technologies promising to turn everything upside down and modify medicine, however this updating will take a lot of time. All innovative ideas tend to pass through a series of long stages. From origin in startups and academic institutions where they are formed, but have no great influence on the industry, to acknowledgement of their potential by big players, who have the resources for a change of a visual environment and transfer of innovations to the mainstream.
I will provide an example. Several years ago there was a set of startups which created solutions for telemedicine platforms with the help of which people could get medical advice remotely. It was an incredibly popular trend, however widespread adoption of it didn't happen. Now the situation is gradually changing, the industry is beginning to accept the rules of the game. Just a month ago one of the world's largest insurance companies turned to DataArt with an idea to develop such a project, a solution for telemedicine. And they, with their resources, with their extensive penetration into the industry and their serious influence on the industry, have a chance to revolutionize the rules. But just a few years ago they barely had an interest in this issue.
My guess is that in five years the big companies will take over. But new niches will emerge and will be occupied by new startups. It is the process that moves the industry forward.
- The gathering of huge volumes of data about patients around the world enables systems that operate with big data to find new dependencies and correlations. For example, the gene of FoxP2 which is responsible for language, was discovered in this way. Are there any more stories like this? What can we expect from the industry in the future? What is the current role of wearable devices?
-Potentially, it is a groundbreaking area in medicine. All the conversations about personalized medicine in many respects are about it. Everything concerning the software development of wearables belongs to startups now, though big companies are gaining interest in wearables too. Not so long ago we communicated with a large company which is engaged in making artificial limbs. They are beginning to understand the potential wearable devices have in their business. The adoption of sensors in their products allows for collecting detailed statistics about the way they are used, where they are used, whether the patient follows the recommendations of their doctor. This information is irreplaceable.
Wearables are capable of collecting of a great variety of data. And, for example, with correlation of the data of a specific person with a historical data file, it will become possible to predict many deviations even before their emergence. When it starts working in full, we will have predictive medicine instead of reactive medicine solving already obvious problems. Medicine has a chance to literally become healthcare as opposed to medical treatment that describes the current condition of the field more precisely.
Alas, all these fine prospects are hindered by conservatism in the field. It is necessary to have the opportunity to collect this data, in order to create solutions in the field of big data, and it is essential to have access to data for comparison and analysis of it. This field is very problematic. Everything regarding personal information is very rigidly regulated. It is necessary to combine data from a great variety of sources to receive a full picture of it. Currently it is quite difficult for companies to get access to data that belongs to other companies. Especially, if one of the services is located in America, the second one in Europe, and the third one in Australia. In this case it is necessary to meet the requirements of three different regulators, and that is practically impossible to do.
— And what are the main risks? Privacy? Excessive dependence of solutions on the data which systems obtain automatically or even provide solutions automatically? External penetration into such systems?
- The roots of conservatism of the industry are in medicine ethics. The main rule of all medicine is "do no harm", and medical technologies are no exception. Every new technology bears big risk due to their interaction with a living person. Any technology won't be adopted, till we don't have certainty that it works correctly in every scenario. The adoption of new technologies in medicine takes the same approach as the adoption of new drugs. It takes long time to enter the market: first we need to be convinced that the technology isn't dangerous, to conduct clinical trials in due form (there are a fair amount of those trials), to be convinced that the specific technology in a specific case is beneficial. It takes a lot of time and money. That is why startups don't drive the medtech market, they lack the resources.
— And what markets are more open now? Perhaps, at first such technologies will appear somewhere in Japan or South Korea, and will other markets get involved then? Are there any examples?
— We work mainly with the American and European markets, my knowledge about other markets is second hand. We may notice that emerging markets act as a platform for many new solutions, just because they are less regulated and more relaxed. Large companies tend to choose India or Brazil for the introduction of innovative breakthrough solutions. Actually, we observed a similar situation with cloud services in developed markets about ten years ago. At that time everyone used to say that cloud solutions were the future, that it was necessary to transfer services to clouds. However, the emergence of fears connected with safety was inevitable: it was difficult to define the responsible agency for the safety of data. Nevertheless, the market has significantly evolved over the last 5 years, legal subtleties and infrastructure have settled, and these issues have ceased to be a problem. Market players should become mature and comprehend the current situation, legal mechanisms should emerge and start regulating technological processes.
— Let's return to wearable devices. How far have medical wearables, not toy fitness bracelets, but real devices like cardiographs-halters, mobile sensors of epileptic seizures, glucometers become a part of every-day life? What shall we expect from them?
— Glucometers, thermometers and other similar devices on the whole are toys. If we have a better look at industrial meetings (which stuff is brought in, what things are discussed), it is possible to notice that this area is still embryonic. The majority of startups are now focused on the production of devices whose full potential isn't used, this resembles situation with fitness trackers. Many companies can produce such devices, but only a few of them can built the infrastructure required for their use. To be more precise, none of them can build it. The main potential for development is located here. There are also more serious opportunities for the application of wearable-technologies. I will tell you about artificial limbs later.
There is also another problem. Wearable devices will remain toys till doctors don't become able to accept the data obtainable from wearable devices. Currently it is just impossible. We can create a very cool wearable cardiograph which will be able to collect data for long term. But, having come to a doctor your user will hear something like this: "Everything is cool, but you should make a regular cardiogram". This is because the doctor is unable to accept the data from the wearable device with the current situation.
—There are devices which are able to read the brain signals responsible for basic actions. In the entertainment sphere it is already being used, it is possible to move a ball by force of thought and so on. Is anything similar used in medicine?
— I can recall only various devices for disabled people helping paralyzed people to use computers. It works by means of the analysis of brain signals or tracking eyes. Similar things help Stephen Hawking to write books and perform in front of audience. All in all, we can see that this direction is developing, but it is not a large market, therefore there is no basis for its further development unlike the market for consumer devices.
— And what real problems are solved by medtech? What kind of problems do clients come to DataArt with?
— They usually come for technologies which already have a place in the market. These technologies are electronic medical records or services for medical business: practice management systems, systems for laboratories. We worked with similar systems for Charles River Laboratories. We are developing in the life sciences area rather actively with pharma-clinical trials and other similar interesting things. Next year we are going to make this field a priority for ourselves. It has enough room for the introduction of technologies and many of our experiments in this field are rather in demand. I am not talking about global industry changes or systems for the invention of new drugs, though we have such projects too. The main work is connected with the automation of already existing processes which used to be carried out without use of new technologies.
— It looks like building a foundation for further development.
— And so it is to a considerable extent. We automate processes which used to be manual. It is a necessary requirement for the further introduction of big data and data flows from wearables and many other things.
— Speaking of that, in the long-term, do you think ResearchKit for the new iOS is a good attempt to create a big platform for the collection of medical data about a person? Did they manage to make anything meaningful?
— Actually, it is not a brand new idea. At different times (and a long time ago) many companies were engaged in similar businesses. Among those companies were Microsoft with their HealthVault, which still exists. It is a centralized storehouse where health data is stored.There was Google with their now closed Google Health. Apple hasn't invented anything fundamental new, they've just integrated it with their mobile platform.
How good is it really? Definitely good. Sooner or later such platforms will gain momentum. We need to collect data for data analysis and making of predictions. However, it is difficult to say whether Apple or other company will pioneer it. But experiments in this field won't stop for a long time, till a stable foundation isn't created.
— At present, Google with their Life Sciences are being engaged in medicine. But as it seems, they more interested in serious fundamental research rather than the consumer market.
— Google is good at providing revolutionary technologies, although they don't always succeed with the commercialization of those technologies. Google laboratories provide things which set new trends in the industry. They have launched their new algorithms for deep learning and let their code be in free access.
With the use of Google Health they tried to make something a little atypical for them, a mass product. It didn't have any fundamental know-how, any academic science for its basis. All these platforms are created due to their capability to raise a lot of money for those who are in control of health data. Population analysis will become possible due to such platforms, and it is the main condition needed to make a step towards predictive medicine.
Now Google is changing their priorities. They are trying to do things which they are good at, revolutionary breakthrough technologies.
— I've recently heard of an interesting project that links together pharmacology and texts analysis. The system scours PubMed and analyzes the latest publications in detail, collecting information about trends. Have you heard of other interesting projects like that?
— Yes, there have been some interesting projects. One of them analyzed academic publications over many years and compared them with understudy data in order to exclude nonviable variants. Speech analysis on the whole has high potential everywhere where technologies can be applicable, especially in medicine. For instance, in the civilized world people are trying to free up doctors' time for patients by letting them spend less time on a computer. To do so, they utilize services for transcribing doctors' voice notes and turning them into digital texts. However, those services haven't been automated yet. And this is when speech analysis comes into play. Voice management is the interface of the future. All these beautiful screens with buttons and even touch-interfaces will die off. When the time comes that a person is able to communicate freely with a machine by voice the traditional user interface will have come to an end.
— And among the things which have already become reality, what projects have you been taken aback by?
— It's one I've already mentioned — artificial limbs with sensors. It was the first really understandable and absolutely helpful use of wearable-devices in medicine, but not in the fitness/wellness-realm where the most of players are concentrated now. As for me, it was the moment when I realized the value of wearables besides their entertainment value. A second case concerns DataArt. We helped one large company reduce a cycle of testing new chemical compounds by replacing the tests that were performed in a "wet" laboratory (a laboratory where test tubes are located) with digital tests. Parsing a large quantity of data about connections through the system and comparing them with a large volume of data that was obtained earlier, we could predict which connections will turn out to be non-viable. The system eliminates a huge number of impractical options which otherwise had to be tested manually. Only viable compositions are sent to the laboratory where they are physically tested. It considerably reduces the costs spent on development.
— What can we expect from medtech in 2016?
— It seems to me, that telemedicine will really start progressing next year. Small companies have already achieved certain success in this area, and major players are starting to pay attention to the field. A few years ago it was difficult to foretell that the major force here will be insurance companies. But now it seems obvious, because telemedicine allows them to considerably reduce expenses. I can't say for sure whether the year of 2016 will become a year for change, after all, such transformations can't be made quickly. But I have no doubts that this trend will become a major one and might gain momentum in the future.