Helio — Increasing lupus diagnostic speed, accuracy, and accessibility
Harnessing the power of Nanosensors and Machine Learning to package the various Lupus biomarker tests into one, increasing the speed and accuracy of Systemic Lupus Erythematosus (SLE) diagnostic.
(Written by Isabelle Lau, Marzooq A, & Ashley Mo)
80 years ago, the world froze in shock. The first case of Cystic Fibrosis (CF) was recorded, and little did we know that this small abnormal gene would cause thousands of deaths to date.
The good news is, we can currently get ahead of CF by detecting it early on and treating its symptoms. So now the 70,000+ living with CF are expected to live a normal lifespan. And this is all thanks to medical advancements in diagnostics.
Unfortunately, this is not the case with Lupus.
Lupus is an autoimmune disease where you immune system is unable to distinguish harmful foreign invaders from healthy cells. You body ultimately produces ANA antibodies which either form antibody-antigen complexes with nearby nuclear antigens and lodge themselves in organs or blood vessels causing inflammation, or latch themselves onto the receptors of innocent cells to destroy them.
This can cause a large range of complications to start happening to your body the most notable being kidney failure, blood disorders, and inflammation of the lungs. And since there’s currently no cure for Lupus, we treat for symptoms instead.
Lupus has earned the infamous nickname of “the disease of 1,000 faces’’ because it has the potential to simulate numerous other illnesses and it presents itself differently on everyone. Which means common symptoms of Lupus are also common symptoms of other diseases.
Lupus diagnosis rates are only reaching 50%.
Note that this percentage could be higher because there are likely to be more people unknowingly living with Lupus. Which means doctors are trying to solve the wrong problem by giving them wrong treatment.
Although Lupus can be drawn back to a combination of environmental and genetic factors, the root cause is still unknown today. It’s highly possible that there isn’t even a root cause to be found because the way that people get lupus is often a unique case every time.
There are over 1,000 known lupus gene susceptibilities that have various levels of strength, and the gene susceptibilities alone are not harmful nor do they guarantee an onset of lupus. They only come into play when a person’s DNA becomes so damaged (typically through sun exposure or certain drugs/medication) that certain cells undergo apoptosis (self explosion of the cell to prevent it from causing harm) and the gene susceptibilities cause nearby immune cells recognize the exploded DNA as external foreigners (and thus begin produced these ANA antibodies). Or sometimes they are noticeable when the body’s clearance system isn’t as quick at sweeping away damaged DNA remains lurking around in the body.
And this is why lupus diagnosis is so tricky.
If we can’t uncover the root cause, we have to rely on a combination very unspecific symptoms to make a final statement. According to the Lupus Research Alliance, at least 4 out of 11 common lupus symptoms must be noticed, and they must be so obvious to the point where patients are suffering painful lives.
Not only does it take time for symptoms to be visible, but various unspecific biomarkers must also be tested as well including ANA antibodies, blood disorders (lack of red & white blood cells and platelets) other proteins as a result of inflammation, etc. This often leads potential lupus patients down rabbit holes of a seemingly never ending series of tests upon tests.
And this causes multiple problems:
- Time consuming: they need one be taken one by one and the process often requires overnight incubation in the labs (it can take up to a weak for patients to hear back)
- High costs: One antibody/antigen/protein test can cost up to $60 USD each (so over 10 is at least $600).
- Need to be done in person: because the tests often need access to labs, patients need to book in person appointment with rheumatologists, which can often take up to several months
And the tests themselves are open to human subjectivity and thus error as most of them are observed with the naked eye.
So at the end of the day, all of these consequences resulting from the very tedious, inaccurate, and costly methods we use to track lupus biomarkers only elongate patient suffering by delaying them from receiving proper treatment to suppress their painful symptoms. On average, it can take nearly six years for a person to be diagnosed with lupus.
Introducing Helio: The world’s fastest and most accurate lupus diagnostic test.
Helio is inventing a next generation electrochemical biosensor to detect all of Lupus’s biomarkers, and uses a machine learning classifier to determine if the patient has lupus — all within just a few minutes.
Detecting 7+ lupus biomarkers
- Various ANA antibodies
- A deficiency of any one or more of these blood cells (red, white, or platelets)
- Protein biomarkers during inflammation (C-reactive proteins, cytokines)
- Abnormal levels of specific proteins due to damaged kidney functions (creatinine, urea nitrogen in the blood)
- Nuclear antigens from damaged cells that underwent apoptosis (an increased amount can tell us if they have gene susceptibilities that cause them to have poor clearance)
- DNA sequences (done with a blood sample)
Deep dive: Electrochemical nanosensors
Electricity is made through the transfer of electrons to create kinetic energy. Electrochemistry takes advantage of this function and uses oxidation-reduction reactions to create this movement of electrons from one material to another.
Using electrochemistry is by far one of the most sensitive ways to build an immunosensors as they can detect the individual transfer of electrons at the atomic level.
To make the nanosensor we need four specific components:
- A biorecognition element: This will be an antibody which the ANA antibodies/antigens/proteins will bind onto
- A working electrode: This is the surface which the biorecognition elements are attached onto and is a conductor through which electricity enters/leaves
- Counter electrode: Used to make a connection to the working electrode allowing a current to flow
- Reference electrode: Is used to determine the exact potential difference between working and counter electrodes
As the each of the biomarkers comes into contact with the biorecognition element, a different oxidization-reduction reaction takes place, and thus a different number of electrons are transferred depending on the specific biomarker. Ultimately, we are able to quantify those electrical signals. This is how we can distinguish between what we are measuring as the electrical current varies depending on the binding agent.
*Note: the DNA sequencing will involved a different process known as capillary gel electrophoresis, but we are still working to integrate this part into the same nanosensor*
Deep dive: Using supervised machine learning to make a diagnosis
Further along the line, Helio introduces a logistic regression machine learning model to pursue the diagnostic process.
It is primarily a statistical analysis method used to predict a data value based on prior observations of a data set. Based on historical data about earlier outcomes involving the same input criteria, it then scores new cases on their probability of falling into a particular outcome category.
It is one of the simplest ML algorithms that is already being applied to diabetes prediction and cancer detection.
The various electrochemical responses shown via a graph will be taken and fed through the algorithm as an input. The output will produce a final percentage stating the likelihood that the patient has lupus (we don’t want to make it a “yes” or “no” response as there are still outer symptoms that need to be considered when making the diagnosis).
The model will be trained with data that we collect with both from people who are confirmed to have lupus, those who are perfectly healthy, and some who may be in between. Through supervised learning, the model will learn to recognize the results of patients who mostly likely have lupus and we will use an additional random assortment of datasets to further validate its accuracy after training.
A faster proper diagnosis isn’t far away.
Without the same root cause for every patient, combining all the biomarker tests into one and using a machine learning algorithm to make a diagnosis is our best shot at cutting the time, cost, and inaccuracy and removing as much human subjectivity as possible. Our goal is to turn those months of testing into just a few minutes, bringing the correct treatment faster to those who truly need it.
Taking the misdiagnosis rate of Lupus from 50% to 0% is ambitious, but at Helio we believe that it is possible.
A misdiagnosis shouldn’t be what stops someone from living their best life.
As the world gets closer to a cure for Lupus, we are doing our part, starting with a proper diagnosis of Lupus. Our solution is currently under validation, but we known this technology is something our team is determined to bring to reality.
Because healthcare is about living a healthy life, regardless of what race, gender, sexuality, religion, disorder or disease. Not about late treatments, an endless series of tests, and apologetic doctors.
If you would like to hear more about Helio’s research, ask questions or leave comments, feel free to contact us. We would love to hear from you!
Ashley: www.linkedin.com/in/ashley-mo-a044781b5
