As Tech Trickles In, Medicine Is About To Hit Warp Speed
The SkinVision app can keep track of a skin mole over time to calculate the risk that it’s melanoma. (Credit: SkinVision)
This incredible abundance of apps, sensors, and information has already kicked off a major shift away from traditional medical practices.
“Basically, what we’re seeing is the digitization of human beings,” says Dr. Eric Topol, a cardiologist and the director of the Scripps Translational Science Institute. “All these new tools give you the ability to basically quantify and digitize the medical essence of each human being. And since patients are generating most of this data themselves, because their smartphones are medicalized, then they take center stage instead of the doctor. And with smart algorithms to help them interpret their data, they can, if they want, become emancipated from the closed-off world of traditional health care.”
Looking to the future, Topol believes that smartphones will radically transform the role that human physicians play in the health care system. “These tools can reduce our use of doctors, cut costs, speed up the pace of care, and give more power to patients,” he explains. “As more medical data is generated by patients and processed by computers, much of medicine’s diagnostic and monitoring aspects will shift away from physicians. The patient will begin to take charge, turning to doctors chiefly for treatment, guidance, wisdom, and experience. These doctors won’t write orders; they’ll offer advice.”
Medicine, meet computer science
Computers have a long history in the field of medicine. Hospitals have been using them to track medical records and monitor patients since the 1950s, but computational medicine — that is, using computer models and sophisticated software to figure out how disease develops — has only been around for a relatively short amount of time. It wasn’t until the past decade or so, when computers became drastically more powerful and accessible, that the field of computational medicine really started to take off.
Dr. Raimond Winslow, director of the Johns Hopkins University Institute for Computational Medicine, which was founded in 2005, says that in recent years, “the field has exploded. There’s a whole new community of people being trained in math, computer science, and engineering — and they’re also being cross-trained in biology. This allows them to bring a whole new perspective to medical diagnosis and treatment.”
Now, instead of just puzzling over complex medical questions with our limited human brainpower, we’ve begun enlisting the help of machines to analyze vast amounts of data, recognize patterns, and make predictions that no human doctor could even fathom.
“Looking at disease through the lens of traditional biology is like trying to assemble a very complex jigsaw puzzle with a huge number of pieces,” Winslow explains. “The result can be a very incomplete picture. Computational medicine can help you see how the pieces of the puzzle fit together to give a more holistic picture. We may never have all of the missing pieces, but we’ll wind up with a much clearer view of what causes disease and how to treat it.”
In a relatively short amount of time, computational medicine has been used to accomplish some pretty incredible things — such as pinpointing the gene and protein markers of colorectal cancer, ovarian cancer, and a number of cardiovascular diseases.
Lately, the field has even begun to branch out beyond disease modeling. As our computational powers have expanded over the years, the ways in which scientists are using these powers have expanded as well. Scientists are now using technologies like deep-learning algorithms and artificial intelligence to mine information from sources that are otherwise useless or inaccessible.
Take Dr. Gunnar Rätcsh of Memorial Sloan Kettering Cancer Center, for example. He and his team recently used computation to unravel the mysteries of cancer in a totally unorthodox way. Rather than building a model of the disease to understand it on a biological level, Rätcsh and his team built an artificially intelligent software program capable of reading and understanding hundreds of millions of doctor’s notes. By comparing these notes and analyzing relationships between patient symptoms, medical histories, doctors’ observations, and different courses of treatment, the program was able to find connections and associations that human doctors might not have noticed.
“The human mind is limited,” Rätsch explains, “hence you need to use statistics and computer science.”
Computational science will open new ways to fight old problems, like the metastasis of cancer. (Credit: Memorial Sloan Kettering)
And Ratsch isn’t the only one thinking outside the box. With powerful new computers, tons of new data, and a myriad of clever new approaches, researchers are cooking up completely different ways to approach complex medical problems.
For instance, researchers recently developed a machine-learning algorithm that tracks the spread of disease by sifting through Twitter for geotagged tweets about being sick. By analyzing this data, epidemiologists can more accurately predict where viruses like influenza are likely to spread, which helps health officials deploy vaccines more effectively.
In a different study, researchers trained an artificial neural network to recognize patterns in MRI scans, which ultimately resulted in a system that could not only detect the presence of Alzheimer’s, but also predict when the disease was likely to appear in an otherwise healthy patient.
We also have algorithms that can diagnose depression and anxiety by analyzing patterns in your speech, and even predict the spread of Ebola by analyzing the migratory activity of infected bats. And the list goes on. These are just a few examples of a larger trend. Computing has invaded dozens of different medical professions at this point, and it will continue to spread its fingers until it’s reached every corner of medical research and practice.
Any discussion of the most significant advances that have occurred over the past 10 years would be woefully incomplete without a mention of CRISPR-Cas9. This single technique is unquestionably one of the biggest achievements of our time, and will have a profound effect on the future of medicine.
For the uninitiated, CRISPR-Cas9 is a genome-editing technique that allows scientists to edit genes with unprecedented precision, efficiency, and flexibility. It was developed in 2012, and has since swept through the field of biology like wildfire.
The acronym CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats. That probably doesn’t mean much to you unless you’re a biologist, but in a nutshell, it refers to an adaptive immune system that microbes use to defend themselves against invading viruses by recording and targeting their DNA sequences. Some years ago, scientists realized this technique could be repurposed into a simple and reliable technique for editing — in living cells, no less — the genome of just about any organism.
Now to be fair, CRISPR isn’t the first genome-editing tool that’s ever been created. Earlier, scientists could edit genes with processes like TALENS and zinc finger nucleases. These previous techniques, however, don’t hold a candle to the simplicity of CRISPR. Both require that scientists build custom proteins for each DNA target — a process that takes far more time and effort than the relatively simple RNA programming that CRISPR uses.
“We could do all this genetic engineering stuff before,” explains Josiah Zayner, biohacker and biologist, “but previous things that people used, like zinc finger nucleases and TALENS, had to be engineered on a protein level. So if you wanted to engineer something for a certain gene, it would take you like six months to engineer the proteins to bind the DNA. With CRISPR, if I want to do a new CRISPR experiment, I could go online, go to one of these DNA synthesis companies, order 100 different things, and tomorrow I could be doing my experiments. So it went from six months down to, well — some of these companies ship overnight now — so not only can you do 100 times as much research, you can do it 100 times faster than before.”
Put simply, CRISPR has cut down some of the biggest hurdles standing in front of DNA researchers all over the world. The floodgates are now open, and anybody can do gene editing.
In the decade leading up to the development of the CRISPR-Cas9 technique, CRISPR was mentioned in scientific publications just 200 times. That numbered tripled in 2014 alone, and we’re seeing no signs of slowing down anytime soon.
In the past two years alone, researchers have successfully used CRISPR to engineer crops that are immune to certain fungal diseases, eradicate HIV-1 from infected mouse cells, and even perform full-scale genome engineering.
And this is just the beginning. As I write these words, the first ever gene-editing trials in humans are actually underway. In August, a group of Chinese researchers will attempt to treat a patient with cancer by injecting the person with cells modified using the CRISPR-Cas9 method. More specifically, the team plans to take white blood cells from patients with a certain type of lung cancer, edit those cells so that they attack cancer, and then reintroduce them back into the patient’s body. If all goes as planned, the engineered cells will hunt and kill the cancer cells and the patient will make a full recovery.
A litany of successful animal trials suggest that CRISPR has huge potential in the treatment of human disease. But arguably the biggest strength of CRISPR isn’t that it’s so simple and effective — it’s that the technique has become so accessible that everyone can use it.
Right now, thanks to a biotech supply startup in California, anybody with $140 can get their hands on a do-it-yourself CRISPR kit and start performing basic gene-editing experiments right on the kitchen counter. Zayner, the company’s founder, hopes that putting these tools in the hands of citizen scientists will boost our collective knowledge of DNA in a huge way.
“There are so many people out there with all this knowledge and skill and creativity and abilities that aren’t being utilized,” Zayner said. “I read somewhere that there’s over 7 million hobbyist computer programmers in the world right now — which is crazy when you consider that in 1970 there were barely enough to fill a garage. But when it comes to genetic engineering and DNA, we’ve been working on this stuff for longer, or at least as long as computers have been around, yet there’s probably only a few thousand hobbyist scientists doing experiments. That’s what I want to change. Where would our medical world be if there were 7 million hobbyist biologists?”
Regenerative medicine grows up
In 1981, two U.K. scientists made a massive breakthrough. For the first time ever, they managed to grow embryonic stem cells in a lab. Stem cells — the cellular putty from which all tissues of the body are made — have a nearly endless list of potential medical applications, and ever since their discovery, scientists have been singing their praises. For years, we’ve been told that stem cell research will usher in a future where we’ll be able to regrow tissues, organs, and even full limbs. But while we’ve long known their potential, it wasn’t until recently that we figured out how to truly use stem cells to our collective advantage.
The thing is, we hit a few roadblocks along the way. After mouse stem cells were first cultivated in 1981, it took another 18 years for scientists to successfully isolate human embryonic stem cells and grow them in a lab. When this finally happened, it was universally accepted as a monumental achievement — but this new technology wasn’t met with open arms by regulators.
In 2001, the Bush administration placed crippling limits on funding for human stem cell research in the U.S., on the grounds that creating stem cells required the destruction of a human embryo (debates about abortion and where life does or doesn’t begin were very high-profile at the time). This didn’t stop progress from happening in other parts of the world. In 2006, a Japanese scientist by the name of Shinya Yamanaka developed a way to make embryonic-like cells from adult cells — thereby avoiding the need to destroy an embryo in order to make usable, versatile stem cells.