An Eye for AI

Eye Scan

An AI becomes intelligent after it has learned a lot. For example, let’s consider the weather. If a meteorologist considers a number of factors including the size and shape of clouds, temperature, dew point, wind direction and speed, and barometric pressure, he or she can forecast the weather for the next hour will be X. With a different set of conditions, the forecasted weather would be Y. If you applied machine learning software to thousands of conditions and resulting forecasts, an AI could be trained to forecast the weather. As more and more data with sets of conditions and actual weather are accumulated and submitted to machine learning, the accuracy of the forecasts would get better. With enough data and machine learning, the weather AI may produce better forecasts than meteorologists.

Similarly, machine learning can be applied to medical data and enable an AI to learn how to diagnose a medical condition. AI systems are learning to diagnose disease across a wide range of medical conditions, and gradually they are becoming as accurate as human doctors. A good example of AI diagnosing is occurring in London. A collaboration is underway there between researchers from Google’s DeepMind subsidiary, University College London, and Moorfields Private, the private patient division of Moorfields Eye Hospital NHS Foundation Trust.[i] The researchers are using deep learning to create an AI which can identify more than 50 common eye diseases based on thousands of 3D scans. With a single scan of a patient’s eye, the AI can recommend a specific treatment. While the research is still in the early stage, not ready for clinical use, the results to date are very promising. The Verge quoted Dr. Pearse Keane, a consultant ophthalmologist at Moorfields who was involved in the research, saying,

The number of eye scans we’re performing is growing at a pace much faster than human experts are able to interpret them. There is a risk that this may cause delays in the diagnosis and treatment of sight-threatening diseases. If we can diagnose and treat eye conditions early, it gives us the best chance of saving people’s sight. With further research it could lead to greater consistency and quality of care for patients with eye problems in the future.[i]

The software the researchers developed uses algorithms which can identify common patterns in data from 3D scans of patients’ eyes. The scans are made using a technology called optical coherence tomography (OCT). The researchers submitted data from nearly 15,000 OCT scans from 7,500 patients. In addition to the data from the scans, the researchers fed the software diagnoses made by Moorfields doctors. Based on what the AI software learned from the data, it is able to develop a diagnosis a new scan. The Verge reported the AI’s diagnoses were 94% accurate when compared to the diagnoses made by a panel of eight doctors.[ii]

[i] James Vincent, “Deepmind’s AI Can Detect over 50 Eye Diseases as Accurately as a Doctor,”  The Verge (2018),
[ii] Ibid.

Tagged with: , , , , ,

Drug Pricing Transparency

Healthcare Cost

Last March, I quoted the new Health and Human Services (HHS) Secretary, Alex Azar, as saying, “Change is possible, change is necessary, and change is coming.” I hope so. Change is desperately needed. It remains to be seen if the federal government can make the changes needed and make them fast enough. The biggest barrier is Congress. HHS tried to simplify drug insurance last year, and Congress blocked the simplification because it would have been negative for the health insurance industry, one of the powerful lobbies that fuels Congress with donations. The most important thing Azar said as part of the recent address was, “We’re unafraid of disrupting existing arrangements just because they are [controlled by] powerful special interests.”

Medical science has advanced rapidly, but innovation in healthcare payment and delivery systems has been slow. While Blockbuster put local video shops out of business and then Netflix put Blockbuster out of business, the healthcare system of care delivery and administration changed very little. The cost of Medicare has grown from $400 billion in 2001 to trillions, but quality has not grown at the same rate.

One of the priorities Azar outlined for HHS was making healthcare drug pricing more transparent, and this week he took steps to make the promise a reality. While USDA regulations require companies to disclose possible side-effects in TV advertisements, the administration has now stipulated pharmaceutical manufacturers must disclose their list prices in the ad copy as well. The pharmaceutical lobby will argue vehemently against such disclosure, even claiming first amendment rights. In other words, they will argue their right to hide prices should supersede consumer rights to understand the pricing. Pharma would like to keep the pricing impossible to understand. Today, patients and purchasers pay very different amounts for the same medications. They use coupons and rebates to lower the advertised price for patients while simultaneously raising what they charge insurers. The insurers then pass those costs back to consumers through higher premiums. While Congress allows U.S. consumers to pay top dollar prices, pharmaceutical companies offer the same drugs to wealthy European countries at much lower prices. Those countries negotiate aggressively. The U.S. Medicare system is forbidden by Congress from negotiating.

Azar has a list of other actions he plans to take to lower drug prices. The president has said drug prices will drop “Really, Really Substantially”.  People go bankrupt or even die because they can’t afford the price of needed drugs. To date, no politician has been willing to stand up to the pharmaceutical lobby. I hope the Secretary and the President have the courage to continue the fight.

Read much more about healthcare and the pharmaceutical industry in Health Attitude: Unraveling and Solving the Complexities of Healthcare.

Tagged with: , , , , ,

Robot for Your Teeth

Reflecting on the future of healthcare during my two weeks out of the country, I conclude two things. Electronic health records, the ability to get speedy access to your health information, the ancient processes for getting medications, the frustrations of getting and paying for medications, and the backward methods of communications are all a train wreck. I continue to believe Amazon, Apple, and others are going to shake up this side of healthcare soon. The other thing I conclude is medical research and development of new surgical and pharmacological treatments are moving forward at a breakneck pace. In ten years, healthcare will be quite different than today.

On a lighter note, a company called Chiiz has developed a sonic-powered automatic toothbrush it claims to clean all your teeth in just 30 seconds. The mouthpiece-styled automatic toothbrush may make the twice daily teeth-cleaning routine a quick and effortless job. The device is basically a mouthguard fitted with an assortment of bristles designed to come into contact with every part of all 28 of your teeth. A small motor using sonic technology is inserted in the middle of the device. The motor generates sonic vibrations which Chiiz claims are equivalent to 25,000 strokes per minute. Before use, you add some toothpaste mousse which cleans plaque stuck between teeth. Simply rinse the device after the 30-second cleaning.

The startup company is raising money through and hopes to be ready to ship product before end of the year. You can read the full story and scroll down to a video of the Chiiz device in action by clicking here.

Tagged with: , , ,

Be Back Soon

I look forward to getting home and writing about healthcare, technology, Internet voting, robots and artificial intelligence. For now, we are off the coast of Santorini in the Greek Islands. I have heard about the islands over the years, but never had much interest in them. Being here now, I can see why so many like to go there. The topography and beauty are amazing. The bay where we are floating was formerly the center of a huge volcano. The water is a beautiful deep blue, and the depth of the bay is 1,300 feet. Being so deep, the ship is unable to anchor and, as a result, it must continuously engage the forward and reverse propulsion of the engines supplemented by the port and starboard side thrusters. Otherwise, we would smash into the shore or one of the other ships in the bay. The view is amazing, and I wanted to share a picture, but the bandwidth is so slow, I have been unable to upload the picture. When I get back, I will share some albums from the trip. We started with a couple of days in Venice and then cruised to Rovinj and Dubrovnik in Croatia. Then, on to Greece: Corfu, Katakolon (Olympia), Nafplion, and Santorini. Tomorrow we will be in Rhodes and then end up in Athens. Be back soon.

Tagged with: , , , , , , , ,

Tracking Cancer Cells With a Nano Cocktail

Nano Robot

I have no scientific basis to substantiate this, but it seems that advances in healthcare during the next 10 years will surpass what has occurred during the past 100 years. Consumer devices such as the Fitbit are the tip of the iceberg that will lead to massive amounts of epidemiologic data being collected by consumers that will benefit them individually and the population at large. One of the many areas of technological advancement is occurring with nanotechnology.

I remember attending a conference 15 or so years ago where a presenter said the day would come when we would be able to drink a “nanobot cocktail”. The nano robots would then traverse our bodies and make “corrections or replacements” to cells they found to be defective. To most, it seemed unbelievable. Fast-forward to 2013 when engineers at the University of California, San Diego had invented a “nanosponge” capable of safely removing dangerous toxins from the bloodstream – including toxins produced by MRSA, E coli, and even poisonous snakes and bees. The nanosponges have already been studied in mice and have been found able to neutralize poor-forming toxins that destroy cells by poking holes in their cell membranes. The nanosponges look like red blood cells, and therefore serve as red blood cell decoys which can collect the toxins. The nanosponges absorb damaging toxins and divert them away from their cellular targets. After a half-life of 40 hours in the researchers’ experiments in mice, the liver safely metabolized both the nanosponges and the sequestered toxins. The liver incurred no discernible damage. This is not science fiction. In fact the researchers have a goal to translate their work into approved therapies. This would be a welcomed breakthrough in the treatment of MRSA, a dangerous and antibiotic-resistant bacteria which has become prevalent in hospitals.

Before continuing, lets reflect for a moment on the word nano. Simply put, nano means “one billionth”. Nano is normally used in connection with with the metric system. For example, the nano prefix denotes a factor of 0.000000001 meters. A typical nano robot ranges in size from 1,000 to 10,000 nano meters. To put this in perspective, a human hair is approximately 80,000- 100,000 nanometers wide.

The latest nano invention goes a step further. A team of researchers at Ecole Polytechnique Federale de Lausanne in the French-speaking part of Switzerland are proposing a nano cocktail. Patients may soon be able to track their illness simply by drinking a solution containing millions of tiny electronic sensors disguised as bacteria. After the microscopic sensors have been ingested and enter the blood stream, they would locate and attach themselves to diseased tissue in the patient’s body and send out a continuous stream of diagnostic data via telemetry. A potential use is to track cancer cell metabolism which could be valuable information for oncologists. The researchers do not believe there would be any side effects. The nano bots would be removed from the body either when a tumor is removed or, if therethere is no diseased tissue, through the patient’s method of passing things out of their bodies. 

Source: Tracking Cancer Cells With ‘Drinkable’ Electronic Sensors

Tagged with: , , ,
Page 2 of 386
1 2 3 4 386