How to improve remote diagnosis using technology?

A talk I gave at IDWeek 2014. I am making the case for the smartphone (mobile, connectivity, sensors, and computing) as the ultimate enabler of remote diagnostics. Use cases, examples of products, and the (possible) future of remote diagnosis.

Tagged , ,

The consumerization of medical devices

A talk I gave at the MEDevice San Diego conference. Consumer medical devices as beautiful devices connected to smartphones with a focus on wellness instead of disease blurring the line between tracking and diagnosis. Devices that consumers want to buy and use as opposed to their clunky, clinical counterparts. I argue that the consumerization of medical devices is the second wave after consumer medical information such as WebMD and Everyday Health in the empowerment of the “patient”. Let’s see when the term “patient” completely disappears 😉

Tagged , , ,

Business model innovation in drug development


Transparency Life Sciences

The right drug/intervention at the right dose at the right time for the right patient is the central dogma of personalized medicine. This implies that treating the median of a population is no longer adequate. Stratifying patients based on genotypic, phenotypic or other markers is key in delivering personalized medicine.

Further, several large pharmaceutical company executives including the CEO of Novartis literally stated on CNNMoney that the current business model of developing blockbuster drugs is dying and that pharma needs to rethink this model. This means that more drugs will be developed targeting smaller patient groups. Reducing costs for conducting clinical trials will be critical to make this model sustainable and payable.

So, in order to deliver targeted therapies and test new therapeutic concepts we need to disrupt the clinical drug development model. An example of an ICT-enabled innovation is the patient-led clinical trial concept that has been tested in several settings, and got popularized by a  landmark study on ALS on the PatientsLikeMe platform.

But, the disruptive innovation may well come from a company called Transparency Life Sciences (TLS). TLS is a drug development company that embraces open innovation in clinical development. The company calls for transparency in clinical trials to improve study design, quality and outcomes while drastically reducing costs. In a blogpost on Xconomy, I discovered that the FDA recently cleared a Investigational New Drug application to assess lisinopril as an adjunct treatment for multiple sclerosis, using crowdsourcing! The business model of TLS uses input from researchers, patients and drug developers to better design clinical trials and conduct these trials with unmatched productivity. This radical (for pharma, this is considered radical ;)) concept  aims to cut clinical development costs by 50 percent initially, along with designing better trials and producing better outcomes. And yes, this is transparent. If you don’t believe me, check the company’s progress on their Lisinopril drug development project.

A great talk on innovation in drug development including the TLS model:

‘Big data’ in healthcare, more than just clinical

For a concept as seemingly simple as an information system that stores and manages data, the electronic medical record (EMR) has experienced a remarkably difficult road to adoption. Many sectors, from banking to retail, embraced big data already a long time ago but, traditionally, the healthcare industry lagged behind. Reasons for this slow adoption include providers’ resistance to change, underinvestment in information technology and privacy concerns. Benefits of adopting EMRs are numerous and nicely laid out in the following info graphic.

Screen shot 2013-05-04 at 19.39.29

But things are about to change, according to a report of McKinsey & Company, a new view of big data in the healthcare industry is reaching a tipping point. After a decade of digitizing medical records an era of open information in healthcare is now under way. In 2005, only about 30% of hospitals and office-based-physicians used basic EMRs. This figure rose to respectively 75% and more than 50% in 2011. Furthermore, around 45% of US hospitals are now participating in local or regional health-information exchanges. Adding to clinical data there exists claims and cost data, pharmaceutical R&D data and patient behavior and sentiment data, that holds great potential for value creation but remains upon today largely unclaimed.

Screen shot 2013-05-04 at 20.21.05

Integrating these pools of information is required to transform healthcare today into a holistic, patient-centered medicine. A medicine that resembles the level of service of companies such as Amazon, Apple and Facebook, and characterized by  Eric Topol‘s C’s (customized, collaborative, crowdsourced, connected and cloud computed).

You’ve got mail. It’s your doctor.

E-mail has been hailed as one of the most disruptive innovations from the past century. Listening to a talk on Big Think from George Halverson, CEO at Kaiser Permanente made me realize how powerful this tool is in health care. He explained how a secured messaging network dramatically impacted the way patients participate in healthcare. With regards to partcipatory medicine, a part of 4P medicine, coined by Leroy Hood, Kaiser has done two things. First, every patient in the health plan can download and access his electronic medical record (EMR) and second, Kaiser allows to send an email to your doctor through a secured messaging environment. You can watch the video here. (not able to embed video from BigThink)

According to, having electronic access to your medical record can help manage your health. 4 in 5 Americans who have access to their electonic health records use it. Access allows a patient to make sure the information is correct and complete, allows for keeping track of important health information and medicines and dosages. More astonishing, a whopping 2 out of 3 people would consider switching to a physician who offeres access to medical records through a secure internet connection.


So, what happened to the messaging system? in 2009 the CEO of Kaiser predicted that in two years from then 40% of primary care visits would be e-visits. Looking at the uptake of the email system, 1 million emails the first year, 3 million the second year and 6 million in 2009, we can conclude that patients love e-visits. This phenomenon fits in the trend of patient-centric care. It is all about making a patient an autonomous, empowered and participating consumer.

One more thing: even including doctor’s notes in the electronic health record has a positive impact. The Wall Street Journal blogged about an initiative called OpenNotes. OpenNotes is an effort to improve doctor-patient communication by letting patients know everything their doctor has to say about them, during and after a visit. It appeared that using OpenNotes, up to 87% of participating patients reported that the platform made them feel more in control of their care and more adherent to medications. Only few people were confused, worried or offended.

Personal genetic information and privacy


Mike Merrill, 35, works in tech support in Portland, Oregon. In 2008 he offered 100,000 shares of himself to investors.
Photo by Chris Buck

DNA is our blueprint. It directs any process in our body and influences who we are, how we think, behave and how we look. Today a post on Technology Review stated that “Privacy experts have identified participants in the Personal Genome Project using “de-identified” data. Reading through the post I learned that they combined a voting list with the data entry on the Personal Genomes Project site in the following way (see figure). I couldn’t help myself thinking that an expert wasn’t really needed to figure this out, but this post made me reflect on what privacy could mean in big-data medicine.

PGP cracked

It strikes me that privacy is huge for government (HiPAA) and FDA but that most people post their whole lives including the good, the bad and the ugly on Facebook or similar platforms. With profiles on LinkedIn, Facebook, Google+ and several blogs we truly created online persona’s with detailed personal information. So, what would be the effect of having your genetic data, publicly available? Would it be shameful to let the world know you have a predisposition to develop certain diseases? After all, everybody has errors and these genetic signatures are just probabilities. Nobody is perfect, right? Or, lies the danger of unprotected genetic information in increased risks of identity theft?

I guess the future will tell but if we take a look at Mike Merrill, who solds shares of himself to investors, and publicly follows what they are saying, things could end up in a mess. For now, you might as well keep your genetic info for yourself.

Why are people so lazy? Why wouldn’t they?


We all know that watching your weight, your diet, making sure you get enough  exercise is a daunting task that even a whole army of behavioral economists, health insurance people and physicians can’t make people do.
A myriad of apps exist to track stuff (sleep, weight, diet,…) and  change behavior but uptake of these apps isn’t that great. After all, do you really want an app in which you have to put in the numbers yourself? Probably, if you do like to do that, you could be the type of person that wouldn’t need it. So, either you are a health freak or a tech adept if you use a Fitbit, a Nike Fuelband or other tracking tools. On top of that, 7 in 10 Americans track their health  but most of them do not even use technology to do so.
So, something is wrong here. My guess is, its too much effort to enter data and/or too much money to buy a device designed to track your health. After all, cash and convenience are king, no? 
Enter the wearable electronics or “epidermal electronics”.
You just put one of these on and it tracks vitals for two weeks. Still seems a bit too far out for me. 
Possibly, real adoption and continuous monitoring of health status could come from the sports clothing industry. Everybody wears clothes and it already support several functions, so why not make it smart?

Doctor Watson will see you now…or is it Dr. Spock?

About a year ago, I was stating that a personalized medicine could not be realized without some artificial intelligence (AI) tools (see below). Just because of the sheer complexity of multi-dimensional data about every thinkable aspect of our body. As we are ushering in the ‘omics’ era of Medicine, AI is moving along. 
In February, IBM announced that is putting Watson to work in healthcare, together with its partners Memorial Sloan Kettering and WellPoint. You can view a demo of Watson in oncology here. In order to help out the oncologist, Watson, combines natural language processing, machine learning, and hypothesis generation and evaluation. Pretty nifty, I must admit. We can expect that such an applications will become ubiquitous during the coming years. 
What if we take this approach one step further and we try to “computer” diagnose ourselves? No need for a doctor with a tool such as Symcat. Symcat calculates the probability of disease based on thousands of patients records. Plug in your symptoms and take it for a spin next time you are feeling sick. 
However, now that we kicked out the doctor, how are we going to examine ourselves? With a tricorder of course! Just like in Star Trek! 
If you think this is too far out. Think again, many companies are working on this. There is even an X-prize to develop a tricorder. This is how it could look like:

Omics and privacy?

The age of “omes” is here. It began with the genome, continued with the proteome and several other “omes” followed.  As we are starting to make sense of the “omes”, concerns about privacy arise. For instance, your genome will reveal much about you: a plethora of risk factors for various diseases, your ancestral past, your phenotype (eye color, hair color…) even whether you like Brussels sprouts and what kind of earwax you have. 
All this information, stored in several repositories would make it in theory possible to re-identify you. As a patient or consumer you would want to get an answer on questions about health, diet, sports regimen, wellness…
In order to be able to get these answers you would need to have your data stored in the cloud and you would have to allow different services to access it and examine it. Or you would have it an your iPod and send it to different services for analysis. This would create opportunities for data duplication and dissemination.Further, social networks already decimated the privacy of personal information and will continue to challenge this privacy even when it concerns sensitive data.
But, do we have to protect our blueprint? As a matter of fact, nobody actually knows. Some people even advocate making it public. For example, Lone Frank, author of My Beautiful Genome, asks herself whether our genomes actually really belong to us, or belong in the public space. 

Another example is the Personal Genome Project who asks volunteers to make their personal, medical and genetic information public, “to advance our understanding of genetic and environmental contributions to human traits and to improve our ability to diagnose, treat, and prevent illness”

Finally the degree of privacy protection by commercial direct to consumer and other sequencing enterprises varies. The company 23andMe allows consumers to choose whether they wish to disclose personal information. However, the company does give you the ability to connect with other individuals who have 23&ndMe accounts. 23andMe also announces that it might enter into commercial or other partnerships for access to its databases. deCODEme offers tiered levels of visibility, but does not grant access to third parties in the absence of explicit consumer authorization. Navigenics is transparent: “If you elect to contribute your genetic information to science through the Navigenics service, you allow us to share Your Genetic Data and Your Phenotype Information with not-for-profit organizations who perform genetic or medical research”.
Just think about this: if your genomic data would be stored, analyzed and disseminated while using all sorts of services. This information would add to the data several companies already gather from you. So, besides your digital fingerprint you create while surfing the web, pieces of your genetic data might get attached to your profile. This would create an ever more detailed digitized “self” version of you on the web. Whether this is a good thing, I am afraid only the future will tell…

From sequencing to services

According to Andreas Sundquist founder of California-based DNAnexus, about 20,000 full genomes have already been sequenced worldwide and this will rise to a million in a few years from now. All that genomic data will soon amount to more than an Exabyte of data.

Just to give you an idea how big an Exabyte is, 2 Exabyte was the total volume of information generated in 1999 and 5 Exabyte contained all words ever spoken by human beings at that time. In 2009 all Internet data accounted for 500 Exabyte, equivalent to a stack of books stretching from Earth to Pluto 10 times.

Possibly in 10 years, everyone in the developed world will have their genome sequenced when they’re born. Since sequencing is no longer a bottleneck, storage and analysis of DNA data is. DNAnexus develops tools that analyze genomic data and the company uses Amazon’s and Google’s cloud services to store your data in a cloud-based account.

Taking it a step further, Illumina announced that it opened an app store for genomics. The app store, called BaseSpace apps, will offer a marketplace for informatics tools that enable researchers to analyze cloud stored genomic data generated by Illumina DNA sequencing systems. Like on existing app stores for mobile devices, Illumina expects to get a cut of the software sales.
The rapidly declining cost of sequencing is making DNA decoding a commodity service, yet a premium will be placed on technology that enables researchers to rapidly and efficiently analyze and interpret genomic data. These initiatives show that the structures are being built; business models are being set up to anticipate the coming “Tsunami” of data.
You might oppose that this phenomenon is still in an early stage and focused primarily on research. There exist still al lot of issues to be resolved. Is your genomic data really yours? What about privacy? However, the success of companies such as deCODEme, 23andMe and Navigenics show that consumer genetics is growing. Also in medicine there exists a push towards integrating the “omics”data with clinical data to realize personalized medicine. 
Oh and just to get started. Play around with Illumina’s app MyGenome® for iPad for 99 cents. The app provides a glimpse of what could become a clinical tool for use by physicians with their patients to improve understanding and communication of genetic data.