Archive for the ‘Uncategorized’ Category

Raw Notes #HealthAPI #SXSW with .@andreimpop .@jitin .@lisabari .@JoshCMandel

March 15, 2016

Raw notes, mostly unedited yet.

Description from SXSW website:

APIs have reshaped the web into an interconnected network of products and services that provide a seamless user experience and have enabled SaaS companies to take over the consumer and enterprise world. However, healthcare famously suffers from a matrix of siloed, dated technologies.

But where are the healthcare APIs? They’re coming. A generation of startups are taking on the enormous task of building simple, portable APIs for health data. These companies are building solutions that will finally bring about the interconnected health system that we are all desperately waiting for, and could hold the keys to cracking open the health IT market.

Why I attended: Because I am interested in the movement to make Wellness more usable.

What I learned (Post-Op): I had never heard of FIHR and learning about this opened my eyes to the fact that very smart people are working on making standards so that data can be easily exchanged. The first question asked during Q&A tapped into my inquiry this year at SXSW which is this: is it time to completely disrupt the idea of “doctor.”

The very simple phrase the questioner, who I think said he is the “CMO” (?) from the American Society of OBGYN is that doctors don’t actually think in data, they think in stories. Therefore there is a need to really create great data visualization of the patient story.

Conclusions I’m drawing: Big need for UX, content strategy (people who figure out how to tab and label groups of data) in this industry.

The raw notes:

Discussion of EHR (Electronic Health Records)

Making things usable.
“Meaningful use.”

The policy question – a very conservative industry, regulation comes in, then what happens with a product roadmap?

What does it mean for the guy with the computer to start winning? The people in *billing* started to use computers to “win” by streamlining billing (.@JoshCMandel).
So now it is how do we compensate people for giving “value,” e.g., high-quality care.

Vendors are expressing frustration that they need to check the boxes. And if it becomes compensation for box checking or you don’t get paid, the focus becomes box-checking.

The good sort of evil. When Governments light the beacon to get industry to go and do the right thing in the first place.

Commonwell – http://www.commonwellalliance.org/

A still evolving app. Started as a set of EHRs. Growing evil that was taking to a wrong place, good set of technologies helping them to inter-operate. Patient portals and app platforms joined. Where will providers really go for care to take better care of their patients.

“FHIR” is an acronym – Fast Healthcare Interoperability Resources – open-standard – being developed by Health Level 7. Like modern web standards. The goal is to describe the most common kind of Health Care data and how they are used and displayed. They define data models and using REST api, 3rd party apps can use to connect. Can pay to be a member. Creative Commons 0 license.

“Technically, FHIR is designed for the web; the resources are based on simple XML or JSON structures, with an http-based RESTful protocol where each resource has predictable URL. Where possible, open internet standards are used for data representation.”

PIMs – Health Industry Tradeshow. Was in Las Vegas. Lot of buzz around standards. How can we use these standards to plug in apps and services to run seamlessly.

Need open standards to allow for extensibility.

OLAF to authorize apps to connect.
FHIR to structure data
EPIC
CERNER
Medihealth

Hosting sandboxes and working together to try to resolve differences. If 12 different ways to converge API, how do we stay must be this way or optional.

Very valuable to be able to share data across network. On the horizontal side a lot of hope. Need for vendors to come together.

One EHR versus another EHR are not that different.

Human API: Look at underlying semantic model of the data. What’s under the structure (or semi-structure)? You can have the adaptor, but you must look at how these things fit together. Where the universe of things is known, here are all the known medications and dosages, for example. Have to do a lot more interpretation of the data when unstructured.

Mandel: “Interoperability” has lost its meeting so they created “substitutability” – meaning you can swap out old one and switch in new one. You need something much bigger than FHIR. Have to layer a few things on top of FHIR. 1) Security protocols and consistent across every system. 2) Terminology that are used to describe the data. To use successfully need to lock those things down. 3) If you want to lock into EHR, you need User Experience Glue, so you’re not just starting an isolated session.

Steve Hasley – CMO of American College of OBGYN. Has 30K docs that don’t think in data, they think in STORIES, in notes, physical history. Their challenge is getting data in.

Very exciting session.

Mandel: Need to tell the relevant part of the story in a clear way. Being able to build a viz tool on lab results, that’s something cool I can do with an app. When it comes to putting data back into the EHR that is easiest. An API like FIHR defines data in two directions. When we host a sandbox server, can do both things. In terms of real vendor implementations, start with read mostly, or read only.

Asnaani: First thing people wanted for Commonwell was Give me the Notes.

Designer for IBM had two questions about patient experience.
Challenges for privacy on healthcare end. Both privacy and security questions.

Pop answered: Have to be very transparent with end user. Need to let them know what is going from where to where. When you build that into your platform you are enabling consumers to vote with their data.

#sxsw @medialab Nanotech and Future of Bioengineering Live Notesle

March 16, 2015

Dr. Bob Langer presented; was supposed to be George Church.

Moderator Joey Ito (MIT Media Lab)

Convergence of engineering and medicine: Using materials to deliver drugs. First: Angiogenesis inhibitors.

Example of how the concepts of people of what can and cannot be done actually stop invention. Dr. Langer hadn’t heard you “could not do” what he has done.

Started with explanation of metasis of cancer and need to stop that process to stop cancer. Used polymer to deliver drug into tumor.

Key thing in science is reproducing results.

1976 published paper on polymers.

FDA has added Angiogensis therapy to list of approved cancer treatments.

NOVA video clip (KQED9)

Slide with Nanochip with microwells for drugs to be contained. Cover the drug with bio-inert, but that open with remote control.

Can take a chip, and communicate over an approved FCC and FDA chip, communicate with via a “Medical Implant Communications Service Band.” This helps with ADHERENCE.

Clinical trial done in Denmark.

Talked about treatment of using this for Osteoporosis. Huge patient compliance issue. Must be given at the right rate.

In Denmark did a small test and had excellent results [My comment: only 8 patients.]

[Dallas Buyer’s Club comes to mind in some of the sessions I am in this morning.]

Talked about ability to implant in “third world” women to control and able to plan families. “Personal Fertility Control System.” [This worries me as seems ripe for abuse.]

One of the things you learn is design. Make it a “design problem.”

Principle of therapy. Line surgical cavity with BCNU-polymer. BCNU half life in vivo = 12 minutes. Polmer protects BCNU from degradation. Explose only the cells you want to BCNU.

Money: Write proposals for grants to Fed Gov’t. But proposals were failing.

Showed a timeline showing the proposal bouncing around from 1981 to 1996, overcoming every “approach that will not work.”

Talked about how his post-docs and students all became heads of things, and leaders — whereas the reviewers did not become leaders. 🙂

Stents: causing recinosis (smooth muscle cells start proliferating.) Solve is to coat with a polymer. Elizar Edelman, professor at Harvard, did lot of work on this technology.

Example of liver failure and doing implantation with biodegradable polymer scaffold.

Speculates in 40 years plastic surgeons will be able to install a new nose.

Can use polymers to replace skin for burn victims, new bones, etc.

FDA approved for burn victims and diabetic patients.

Replacement spinal cords have been tested on mice. Implant plus stem cells makes replace better.

Showed control versus treated animals in laboratory settings. [Difficult to watch because on one hand, useful results. On other, cruel to those particular animals.]

In human trials, main success has been no adverse events.

Can do work at the interface of engineering of biology and medicine, to relieve suffering and prolong life.

Q&A with Joey Ito:

Internet start-ups are better posed to solve problems because no one told them they can’t do it.

Need to know the science? Depends on what you do in the company. It’s a team effort, doing different aspects of the work, very well. Role for all kinds of people in terms of the start-up culture.

Joey thinks of things connected together into a system. Our whole body is a system. System outside and drug compliance via social media. How important things like community and systems are.

Dr. Langer: My goal is to take concepts that seem impossible, but he feels can change the world. And move it to where it seems feasible.

How to fix the FDA? Bob Langer: FDA is very well-intentioned but under-staffed. They are not in a position to do things they’d like to do scientifically. If they make a slight mistake it echoes loudly, [so makes them a bit risk-averse.] Also the political changes affect. Having good role models is important.

Can bioengineering be used in entertainment? Polymer libraries. Some are used in hair care.

#sxsw #decodingme Decoding Our Bodies: New Era of Citizen Health Live Notes @sam_debrouwer @lindaavey @jessicarichman @DrGreene @wearecurious

March 16, 2015

Looking to doctors for diagnosis… In Greek means to know apart.

Interactive health.

Greene: asked panelists to tell personal story of what motivated them to do their work.

Avey: founder of worlds largest database. Had a friend who took an allergy med that one day woke up with egg yoke- colored eyes. Severe reactions to drugs. Her motivation to start 23 and me was this.

Sam DeBrouwer: Scanadu co-founder. Her son was ill. Limitations to what she could process and what doctors could do. She realized others in same situation. And that smart phones ubiquitous, allows us to connect. New sensors on market. Potential of machine learning. Design: attention to good design is improving. Scanadu product allows us to measure own bodies.

Jessica Richman: wants to fix broken systems. Big Data was once called computational social science. Citizen science is about being heard and not ignored. Biome is about measuring poo of bacteria. 

Dr. Alan Greene: Chief Med Officer Scanadu — levels of authoritarianism and participate tin. For level of engagement, increase of engagement meant better outcomes. (Study name I missed).

Richman: almost all medical practice based on Petri dish, but we are covered in bacteria. Has the largest micro biome database in world. Turning in direction of citizen empowerment. More dignified.

DeBrouwer: we don’t have data on go. Dignity: we have been treated as average, but we are unique. I am a number. A very unique number.

Dr. Greene: “Dia” implying uniqueness. 

[my comment: do I care about privacy if sharing my data will help me get better?]

Avey: about “we Are Curious” when you look at a disease diagnosis, lot of grey area. How appropriate is a diagnosis. How appropriate is the diagnosis. Do you share with people who have it in common.

Q: Danish system does not incentivize sharing this info with doctors. 

Richman: doctors are scared of liability, trained to follow protocol. So thinking of ways to make ubiome 

Q: 23 and me restricted from discussing health, only can do ancestry

Richman: 23 and me pioneered citizen powered health

DeBouwer: we have been excluded from our own health for thousands of years. This will give us all potential to learn. Triage can be faster, more automated. [comment: makes me think of John Havens and need to make ethical decisons]

Richman: instead of doctor being center of knowledge repository, patients will be at center. Experts will have a place. Just that there is too much of a dichotomy. What will happen is patients will learn more about themselves and then consult experts. Legitimate concerns about who should have info and at what pace.

Avey: picking up diagnosis earlier.

Dr. Greene: early info on diagnosis. Study: people who got results of tests earlier actually felt better and made fewer unnecessary calls.

Society for Participatory Medicine: has list of practitioners

– idea for list of data source

Q: what would drive the larger change?

Richman: insurance about costs; doctors care about utility more; Pharma cares about FDA; FDA cares to protect people from menaces. As you tweak, have to think about stakeholders. You think about all these stakeholders, we have not thought about patients.

DeBrouwer: thinking about price versus value.

Q: tools on horizon for cross correlation?

Avey: looking at analytics for We Are Curious, working with Steven Wolfram

How to understand what is meaningful. Hidden data around us every day like CO2 in house affecting sleep.

Q: how does this relate to public health (Ernesto asked, someone they knew)

Quantified Self

Richman: up pipeline from public health. [ocurred to me that ACA more amazing than I’d considered, because brave about tech, fearless about failing]

DeBrouwer: ecosystem is being built right now.

Avey: what’s cool about US is exactly that we are a melting pot. We can learn and apply because of diversity.

Q: functional doctors not taking insurance

Q: discharges? Health loop start-up

Q: breast cancer survivor — researchers lined up for her — Richman, we do collaborate

#sxsw @walterbrouwer @erictopol @johnnosta Raw Live Notes Is Big Data the New Wonder Drug

March 15, 2015

Live notes

The Patient Will See You Now, book by Dr. Eric Topol.

Scale of data collection.

Apple’s HealthKit and rearchkit — in a day over 10k people enrolled in a day.

Main mission of consumer is another more educated conversation with doctor. And willing to pay for data interpretation.

Too much data? Topol says becomes hoarded data.

Data is a conglomerate of noise and signal, deBrouwer. By collecting data on ourselves, we increase our signal. 

Wisdom of body. Homeostasis. New way for individual to have control. Topol

DeBrouwer: data can heal. Digital homeostasis. Beyond the individual.



Mobile phone medical record is what we want. Has to be automatic.

Big data is a wasteland and bottleneck is.

(Note: wordpress app issues. Not saving notes).

[Ironic to see McDonald logo behind them–and Miller Lite.]

Home runs?

Dr. Eric Topol: cutting across all medicine. Folks who can process huge swaths of data are working in other industries.

Big Data the next microscope/telescope

Behaviorial change: when people see data, it changes their behavior short term. 

DeBrouwer: everyone doing it helps with quitting smoking.

Clinicians not going away.