Pages

Monday, January 20, 2014

Computers + a few docs can manage "an entire city", and other cybernetic miracles

The computing aboard the Apollo spacecraft, that provided computation and electronic interfaces for guidance, navigation, and control of the spacecraft that went to the moon, was primitive by today's standards.   From Wikipedia (http://en.wikipedia.org/wiki/Apollo_Guidance_Computer):

The Apollo flight computer was the first to use integrated circuits (ICs). While the Block I version used 4,100 ICs, each containing a single 3-input NOR gate, the later Block II version (used in the crewed flights) used 2,800 ICs, each with dual 3-input NOR gates.  [Today's CPUs alone have millions of logic gates and can have billions of transistors, http://en.wikipedia.org/wiki/Transistor_count - ed.]  The ICs, from Fairchild Semiconductor, were implemented using resistor-transistor logic (RTL) in a flat-pack. They were connected via wire wrap, and the wiring was then embedded in cast epoxy plastic.

... The computer had 2048 words [2K - ed.] of erasable magnetic core memory and 36 kilowords of read-only core rope memory. Both had cycle times of 11.72 micro-seconds. The memory word length was 16 bits: 15 bits of data and 1 odd-parity bit. The CPU-internal 16-bit word format was 14 bits of data, 1 overflow bit, and 1 sign bit (ones' complement representation).

Today's computers that you buy in WalMart have several million times more memory and long term storage, and are thousands or millions of times faster.  Very roughly, a cycle time of 11.72 microseconds or 0.00001172 seconds/cycle translates out to about 85,324 cycles/second = 85 KHz = 0.085 MHz.  Today's computers have clock speeds in the GHz (1 GHz = 1,000 MHz = 1,000,000 KHz) and do far more per clock cycle than the Apollo guidance and navigation computers.

Therefore, we should be going to the moon millions of times faster, and millions of times cheaper, requiring millions of less man-hours of work, now than in 1969, no?

This, of course, is bizarre and absurd, but it is the type of wild (il)logic received by Presidents and Congress and other national leaders regarding healthcare computing:

http://www.healthcareitnews.com/news/mu-creates-medical-bridges-nowhere

MU creates 'medical bridges to nowhere'
Clinton Foundation event draws industry heavyweight who would change meaningful use if he could
January 15, 2014

As far as Patrick Soon-Shiong is concerned, the $34 billion health IT and electronic medical record incentive program was a grave misstep for the healthcare industry -- but not necessarily for the reasons one might think.

Soon-Shiong, MD, who serves as chairman and chief executive officer of healthcare IT company NantHealth and heads the Chan Soon-Shiong Family Foundation, said one of the biggest failures that led to the creation of the Meaningful Use EHR Incentive Program was an inherently flawed mindset of how we view the healthcare industry.

"Nobody has looked at healthcare as a systems approach," he said, speaking at the Clinton Foundation's 2014 Health Matters Conference Tuesday.

I don't think that "nobody" has looked at healthcare via a systems approach, but the point is well-taken.

... The meaningful use program, Soon-Shiong opined, incentivized providers with a process issue, not an outcomes issue -- so what big effect does that have on patient care? The effect, he said, could be much bigger.  

All we need is to do a few things with computers and interopearability, and...

... Soon-Shiong then opted to act on his own, privately. He sold two of his companies, Abraxis BioScience and American Pharma Partners, took $1 billion from the sales and starting building an integrated, interoperable clinical operating system. This system, he says, talks to any software.   
The operating system has been up and running for three million people with cancer and used by some 8,000 oncologists for the past three years, said Soon-Shiong. "We've built in a software system that actually takes 10,000 cancer protocols and provides for the doctor in real time the knowledge of which cancer protocol to give to the patient."  

He then went further, inking deals with Verizon and AT&T to build an electronics company that created boxes that could talk to each other -- blood pressure machines, pulse oximeters, you name it. Soon-Shiong and his team wrote to the APIs of 6,000 medical devices made from every different vendor, and went into hospitals with a real-time wireless, biometric connect box. "We are now capturing 3 billion vital signs, real-time, self-populating in the electronic medical record," he explained. Device connectivity is now operational at more than 250 hospitals, according to NantHealth officials, 120 of those being Epic sites.

And now ... Presto!


... Soon-Shiong and his team also have an Internet-based telemedicine device with a four-to-five way video conferencing ability.

"If you look at this from a systems perspective, if you can now manage a patient from a home, a clinic, hospital and through a supercomputer do the genomic analysis in 47 seconds, which we've now accomplished," said Soon-Shiong, "you then have an engineered system for the nation."

Here's where we go off the cliff of Cybernetic Miracles:


... He continued, "You have the ability to create what I call, 'NORADs' of healthcare. You then have the ability to create a building with three cardiologists, 10 oncologists, two pathologists, one pediatrician that can manage an entire city."  


3 cardiologists, 10 oncologists, 2 pathologists, 1 pediatrician + computers can manage an entire city.

Remotely.

Strap yourself in for the next departure from your local spaceport to the moon.  We have the high-speed computing and the Internet, after all ...

For related health IT and moon illogic, see my 2007 presentation to the IEEE Medical Technology Policy Committee "To The Moon In A Hot Air Balloon:  Why Is Clinical IT Difficult?" at http://www.cci.drexel.edu/faculty/ssilverstein/IEEE_12102007.zip.

-- SS

2 comments: