The Next Generation of Preventative Medicine
Turning Sick-Care Into Healthcare
--
Imagine a world where people don’t have to get sick to get cured. Where people have the peace of mind to never worry that disease might catch them off guard. A world where people don’t have to have their lives shattered by a single visit to the doctor. A world where people know exactly what is going on in their body and can take action early and effectively.
Unfortunately, such a world does not exist… yet
Healthcare As It Stands
Here’s a hard truth. Healthcare since its inception up till today has not really been healthcare. It would more accurately be described as sick-care 😷. Other than encouraging people to exercise lots and eat healthy, we don’t prevent diseases from occurring before the patient feels symptoms.
Instead, what we do is we wait for people to get sick — and then we wait for them to feel sick enough to get help. Then, and only then do we have a doctor try to diagnose the disease based on the symptoms which the patient is describing.
We are always one step behind our enemy.
This is not to mention the fact that different people show different symptoms, different diseases show the same symptoms, and people often avoid visits to the doctor altogether.
- 80% of Americans admit in delaying or forgoing preventative care
- 25% of doctor’s appointments are missed due to transportation
All this leads to is a very inefficient “healthcare” system that is focused on diagnosis and dependent on a symptomatic-reporting basis. In the US, diagnosis alone accounted for $250 billion dollars per year — but not only is diagnosis costly but in many cases, it comes much too late.
The Importance of Early Diagnosis
To illustrate just how important early diagnosis is, up to 90% of patients who are diagnosed with cancer at an early stage survive. Those who are diagnosed at a later stage only have a 10% chance to live. Those are not very good odds when you couple that with the hard fact that almost half of cancer patients are diagnosed too late. The truth is, most people are diagnosed in the later stages, where symptoms are more apparent, more serious, and they are thus more willing to seek help.
It is exponentially harder to find cancer in its early stages where one feels nothing is wrong with them and subtle biomarkers are the only hints of the deadly disease. The same goes for many other afflictions, which don’t appear serious early on but quickly worsen as a result of late diagnoses.
Our healthcare system is flawed, archaic really — and this issue will continue to affect humanity if nothing is done. So what can we do about it? How do we make our healthcare system focused on prevention and early diagnosis rather than waiting for the onset of symptoms and then resorting to treatment?
Enter our moonshot project Iris, which will be capable of providing 24/7 health diagnostics on the human body.
Next-Gen Diagnosis At The Nano-Scale
Iris works by using silicon nanowires attached to a stent to pick up on the telltale biomarkers of lethal diseases long before symptoms occur and the pathogen progresses to a later stage. These nanowires are the perfect size as they are on the nano-scale and are thus capable of being implanted in the body without complications. Moreover, they are capable of giving real-time biological information about a system.
How do nanowires detect specific biomarkers?
Think of a key 🔑and a lock 🔒; each key has a special shape and fits only into one lock. Biomarkers work the same way. Each biomarker has a specific shape, so to detect them, a bioreceptor uniquely moulded to fit that biomarker is attached on the ends of the nanowires (locks).
As the biomarkers near the bioreceptors, they bond, sending an electrical signal through the nanowires and generating a current. The higher the ampere, the higher the concentration of the biomarker in the blood — and the more likely it is for a person to be at risk of that particular disease.
But wait — you may be thinking — what happens if all the locks have keys in them? Doesn’t that make the sensors useless if the concentration seems oversaturated all the time?
Having these locks be reusable is essential for allowing these nanowires to function effectively in the body for extended periods of time. Luckily, the body has a natural way of taking keys out of locks: red blood cells.
Learning From Nature
Red blood cells are specially shaped to attach to oxygen. Every time an oxygen atom attaches onto a red blood cell, the shape of the red blood cell slightly changes in a way to attract even more oxygen molecules until it reaches a point of saturation. Once saturated, the red blood cell changes shape to only want to accept carbon dioxide molecules. Then, when the CO2 bonds to the saturated red blood cell, it releases all the oxygen.
By modelling the shape of the receptors to act like red blood cells, the nanowires can release the biomarkers when it bonds onto a common particle in the blood (for example, the amino acid alanine) — thus rendering them reusable.
TLDR: Silicon nanowires are tiny devices that use bioreceptors to detect specific biomarkers.
In the Iris system, these nanowires are attached to a stent which is implanted in the thoracoacromial artery — an artery with low blood pressure in the shoulder. Traditionally, stents have been tiny tubes that medical professionals have inserted into blocked passageways to keep them open. Stents restore the flow of blood in the area and can remain in the body safely essentially forever. They are a tried and tested medical device that doctors trust.
However, the ability to add silicon nanowires to a stent is a novel idea that has not yet been explored and which we think has enormous potential for the in vivo early detection of disease. A hollowed-out stent would provide the much-needed protection for the nanowire electronics from corrosive bodily fluids while acting as a static location for nanowire sensing.
Detect Diseases In Real-Time
With multiple silicon nanowires in the bloodstream, we will be able to detect many diseases, even at an early stage. That’s again because of the 🔑 ability of the nanowires to detect biomarkers — measurable substances in an organism whose presence is indicative of a particular disease or infection.
For instance, Basic Fibroblast Growth Factor and Insulin-like Growth Factor are both universally recognized biomarkers for increased risk of cancer. Using nanowires on a stent, we will be able to detect the increased presence of these biomarkers in blood, warning patients when they are at particular risk of developing cancer.
The same goes for other endocrine diseases because they have specific hormones related to them. These hormones are tested in a lab setting and identified as the biomarkers for those specific diseases. Thus, when our nanowire sensor detects that the levels of that hormone are exceptionally high in the body, it can warn the user and advise them to visit their doctor.
Even for conditions like depression, which are not easily diagnosed in laboratory settings, nanowires can help. Traditional methods ask for specific symptoms of depression like daily moods and behaviours. But now neurotransmitters such as serotonin, dopamine, or norepinephrine which regulate feelings of happiness can also be detected.
Using LSTM Networks for Diagnosis Predictions
Coupling the biometric sensor information with demographic data and previous medical history allows a Long Short-Term Memory (LSTM) network to make informed predictions on the risk levels for different types of diseases. This particular recurrent neural network type will allow an AI algorithm to not just use the real-time sensor data but take in feedback from its history, allowing for greater temporal resolution.
For example, based on your family’s history of hypertension and the recent decrease in your blood’s level of soluble tumour necrosis factor-like weak inducer of apoptosis (sTWEAK), the LSTM might recommend that you see your doctor, giving you a risk level of 54% for a potential abdominal aortic aneurysm (AAA). Using the feedback it obtains from the true diagnosis your doctor gives you after screening, it will self-improve by adjusting its weights, becoming more and more accurate with its predictions over time.
Furthermore, keep in mind our nanowire is sensing and transmitting data in real-time. By working hand in hand with doctors using cloud-based AI and constant monitoring of in-body diagnostics, Iris is able to provide a never-before-seen, 24/7 comprehensive health picture for the risks of different diseases.
But How Is It Going To Be Powered?
Small sensors capable of being implanted in the human body have been around for a while, but they have all been roadblocked, largely as a result of the necessity to power them. Batteries are a terrible solution for any device within the body because they must be periodically replaced and are usually also very bulky and big. Yet traditional approaches for wireless communication have been unable to power miniature sensors beyond superficial depths.
Iris uses a novel radio frequency networking technique that precludes the use of batteries. Developed by researchers at MIT and Harvard Medical School, in-vivo networking (IVN) is a system that can communicate with miniature sensors implanted deep in tissues without the need to know the exact location of the sensors.
IVN able to communicate deep inside tissues where traditional approaches have failed by having their signals attenuate. It is able to do this using a unique beamforming technique, which uses harmless radio frequency (RF) waves in the sub-GHz range (around 900 MHz).
The system relies on an array of antennas that emit radio waves of slightly different frequencies. As the radio waves travel, they overlap and combine constructively. This allows the sensors to power up, sense, and communicate all without the need to be replaced (no batteries!).
The system has been tested with sensors no bigger than a grain of rice and trials have successfully communicated with sensors 10cm deep in fluids, as well as with battery-free tags placed in a central organ of a swine. IVN is a truly amazing system that offers a much-needed power source for the realistic operation of in-body biometric sensors.
A Worry-Free Future
With Iris, our vision is to have an RF generating band that would both communicate and power up the nanowire sensors within the thoracoacromial stent. It would be able to aggregate the sensor data and relay it wirelessly to smartphones, where an application would be ready to receive it.
Using a centralized computing system, the AI would then analyze the data, providing suggestions and giving warnings when needed. Through a smartphone, you would be able to monitor your own health on a 24/7 basis.
No longer will people have to get sick to get treated. No longer will families need to grieve because the doctors found the tumour a little too late. Symptom-based diagnosis will be viewed as archaic, inefficient, and a thing of the past.
With Iris, people will live longer, healthier, and happier lives.
Challenges to The Iris Vision
There are two main challenges that need to be tackled before the Iris electrochemical sensors can be implanted in the human body and deliver full body diagnostics in real-time.
1. Price of Silicon Nanowire Sensors
First is the price of nanowire sensors. Right now, each nanowire sensor costs around $~16 USD. That might not seem like a lot if it weren’t for the fact that each individual sensor can only detect one biomarker. For the Iris system to be effective in providing real-time internal diagnostics, it must be able to detect a whole host of biomarkers.
The bright side is that the nanowires are small enough that thousands of them can fit in a stent (although practically, even less than 100 nanowires might be needed). We are optimistic that the price of nanowires will continue to drop as more research is done in the area.
2. Size of RF Generators + Communication Speed
As mentioned before, IVN requires multiple RF generators to produce enough power for the sensors. In the original paper for IVN, the researchers at MIT used USRPs (Universal Software Radio Peripherals). These are large radios that require a fast connection so they were connected via ethernet to a computer.
As you can imagine, this is not ideal. We cannot be lugging around a bundle of wires and radios if we want to have convenience while still having in-body diagnostics. Current RF generators have reached a point where they are only millimetres in size and weigh only 150g. Iris’ vision is to eventually scale them down enough so that they can be embedded in a wearable band.
The second problem is the connectivity speed. The Iris band must be completely independent of any wires so that the user has freedom of movement. We anticipate that the development and widespread implementation of 5G networks will alleviate this issue.
Key Takeaways
- The modern healthcare system is symptom-based and people only go to the doctor after they feel sick
- Many people are diagnosed in later stages of a disease where it is harder to cure, more costly, and more painful
- Iris provides a solution by using silicon nanowires powered by radio waves to detect biomarkers in the bloodstream
- This information is aggregated via in-vivo networking to a wearable band and then relayed to a smartphone where cloud-based AI analyzes the data and gives a recommendation when necessary
- Doctors and patients alike will be able to take proactive action early, when it is most important and prevent diseases from becoming a bigger threat within the body
