Building the bridge between AI and biotech

AI is a very powerful toolkit, it is the merge of mathematics, neuroscience, statistics, and many other fields that have built powerful methods to analyze information and obtain models that can predict, classify and segment many problems in our everyday life. As the era of deep learning keeps appearing in our cellphones and TVs it has always seemed weird to me that we are close to having self-driving automobiles, yet we face the common flu as we did 12,000 years ago.

This was the very first issue that got me into biomedical engineering, 3 years later after learning the fundamentals of computer science, robotics, and physiology; I hope to be able to answer the following question as well as possible: where does AI fit in to the field of health?

The first thing my professors, colleagues, and basically anyone who dedicates their life to health care will get angry about is if I said that we treat diseases the same way we’ve done since the caveman times. And they’re completely right!

FMRI, CT scans, vaccines, drugs, DNA sampling, and sanitary conditions everywhere have improved; I’d be mad if I said that we have the same conditions as the patients of Dr. Snow back in London in the second half of the 20th century. We live longer and healthier (to a fair regard) and we even appear younger than our ancestors.

Nowadays diagnostics are mainly powered by human experts, but it is also common to see AI algorithms now taking place on the board’s decision over complex cases; CNNs, transformers, and SVMs have taken over FMRI and X-ray imaging to predict tumors, fractures, calculate bone age, and many other amazing tasks with the precision of an expert and the efficiency of a machine. Other powerful algorithms are taking care of analyzing DNA and trying thousands of new combinations of molecules for treating disease; smart pooling is being used to predict the susceptibility of a population to a new disease. In short, high-level medicine involves and accepts AI and surely will accelerate the creation of a disease-free world someday.

But progress also comes with unexpected maladies, the so-called “21st-century diseases” have taken over the diet, exercise routines, and overall life quality of us clever humans living in the future: Diabetes, chronic heart disease, accelerated aging, lung diseases such as emphysema and pneumothorax, obesity and many more are our daily life concerns; What about all these diseases? How come we don’t have tools to prevent them?

Often when trying to explain the complex dynamics of the human body, one comes to the analogy of a car, you’d be surprised how well it works! Professor David Sinclair wrote in his book Lifespanthat we are the equivalent of driving a car without a dashboard: it would seem unreasonable that you buy a brand-new car in which you can’t see how much fuel you have or if your motor has any issues.

Well, what about humans? we seem to go through our daily life without knowing our heart rate, cholesterol and blood sugar levels, our DNA landscape, blood pressure, and many other important values. We are accustomed to caring about diseases when pain is no longer controllable instead of seeking to prevent and mitigate risks. Biometrics and biotechnology are a wide field that hasn’t yet become as important as they should be around the world, especially in third-world countries in LATAM and other parts of the Global South.

Now we come to the task at hand, and what I hope would shape the world of tomorrow in regards to how we choose to live our lives: We must upgrade our current embedded systems.

Nowadays (and still with plenty of effort) you can get a smartwatch or wristband that can measure the basics of heart rate, calorie consumption, oxygen levels, and pressure; good work indeed but we have some issues to cover.