Endometriosis diagnosis today sucks

20 min readSep 23, 2021

“the doctor said I was making my pains up so I could miss school — it’s taken me 11 years to finally get a diagnosis”

“just got told I had heavy periods”

“when I go to A&E, they think that I might be a drug addict and that I was making it [my pains] up”

“the pain is like barbed wire wrapped around your insides and someone’s pulling it while at the same time an animal is trying to eat its way through you.”

There’s a disease that affects 1 in 10 women (190 million)where it feels like “someone’s getting a knife and just whipping it around on the inside” however, it does not have a cure and the average person gets diagnosed 8 years LATE.

Endometriosis.

Nathalie Zender, my teacher, and her endometriosis story represents that of millions, and we’re here to build a different storyline for those that follow.

Here’s her story.

She got her first period at 13. Her symptoms included severe cramps, bloating, heavy flow, body aches, nausea, fainting, vomiting, fever, loss of appetite, intense mood swings, depression, insomnia, and social anxiety.

At age 14, she says, “my mom finally had enough and took me to a doctor to see if my menstrual cycle and side effects were normal. The doctor immediately prescribed me birth control to reduce the symptoms without mentioning any other options. It felt as if my suffering was insignificant and birth control was the most convenient solution for the doc.”

“Between the ages of 14–28 (now), I have tried 7 different birth control pills because of their negative side effects. Some would relieve my cramps at the expense of my mental health. It seemed as though I had to sacrifice one thing for another. Until I finally found one that seemed to offer me the best balance between the mental and physical health of the pill: Zinnia.”

Fast forward to now, at the age of 28, she was diagnosed with Deep Vein Thrombosis (DVT) which was likely caused by the prolonged use of oral birth contraceptives.

Most blood clots caused by birth controls develop in the leg. She was one of the few whose blood clot formed in between my left clavicle and axilla. She was considered a high-risk patient because the blood clot could move to her brain, heart, or lungs in a moment’s notice, resulting in immediate death.

Nathalie continues, “all of this could have been avoided had my doctors or gynecologists accurately diagnosed me with endometriosis in my early teen years, or at the very least warned me that it is not advised for females to be on birth control longer than 5 years due to increased risk of blood clots and infertility.”

Now, thanks to this misdiagnosis, she needs to get a “rib resection” to alleviate the compression caused by the blood clot in her left clavicle. This is an invasive cardiovascular surgery that could have been prevented had she’d been educated about the risks of the prolonged use of hormonal birth control pills as well as the dangers of endometriosis.

Because her endometriosis was diagnosed 14 years late, she’s much more likely to be infertile compared to the average person.

“ I find it ironic living in a world where women are pressured to birth children, yet society doesn’t educate us enough about the female reproductive system because it’s taboo.” — Nathalie

please proceed with caution as there are graphic pictures that might trigger some people. graphic pictures end before the “ what causes endometriosis” part.

What exactly is Endometriosis?

Endometriosis happens when the endometrium, the tissue that usually lines the inside of the uterus, grows outside it and attach to other parts of your body. This tissue acts like regular uterine tissue does during your period: It will break apart and bleed at the end of the cycle. But this blood has nowhere to go. Surrounding areas may become inflamed or swollen, potentially resulting in scar tissue and lesions. This is why it’s most noticeable and main symptoms are experienced during menstruation — stomach pain, nausea, cramps, etc.

The growth of endometriosis is dependant on estradiol, an oestrogen steroid hormone, which stimulates prostaglandins. The thief responsible for period cramps and pains.

This affects one in ten menstruators — 200 million womxn have endometriosis. Despite this, in the United States alone, where endo affects 7.6 million people, 13x more than the number of people cancer kills (<600,000), it receives 714x less research money than cancer (~5 billion) per year.

What causes Endometriosis?

Some in the past have had different theories about what causes Endometriosis. One of the most hashed out and talked about theories is the Retrograde Menstruation Theory:

This theory proposes that endometriosis happens due to the flow of endometrial cells/menstrual debris in the fluid that goes reversed AKA the opposite way it’s supposed to go via the fallopian tubes into the pelvic cavity during menstruation ( thus its nickname, retrograde ). Wait… how ?

So every month during one’s period, the lining of the uterus sheds, comes out through the cervix and the vagina. But in everyone, a little bit of it flows backward, up into the fallopian tubes and up to the abdominal cavity, and can settle. Once it settles it can grow into the ovaries, uterus, bowel, etc.

But recently this theory has proven to have some weaknesses if we only think that Endometriosis is ONLY revolves around this theory.

Endometriosis has been classically defined as a gynecological disease, predominantly a pelvic disease. Not until recently (March of 2021) did scientists make groundbreaking news on the disorder

“this description ( paragraph above ) is outdated and no longer reflects the true scope and manifestations of the disease”, https://doi.org/10.1016/S0140-6736(21)00389-5

Endometriosis is now a disease that affects the whole body, a systemic disease. Not just a uterus thing anymore. How?

Endometriosis affects metabolism in the liver + adipose tissue -> systemic inflation -> alters gene expression in 🧠-> pain sensitisation + mood disorders

https://doi.org/10.1016/S0140-6736(21)00389-5

Because proinflammatory cytokines travel around the body, this makes it not just a pelvis-uterus thing. It causes inflammation around the body and some data has shown that it can migrate to other organs. From your spleen to giving your brain stronger pain perception. The stem-cell trafficking that happens in someone who has Endometriosis alters gene expression throughout the body. (1)

There’s MORE Endo Does Than Make You Feel Like “someone’s whipping a knife around in your insides?”

The major symptom of endometriosis is pain and cramps during one’s period that affects your daily routines. However, it’s not just the pain:

Those who struggle with anxiety or depression due to dealing with the symptoms however being unaware of the diagnosis. Medical treatments and mental health care can help, however suppressing the consequences, does not solve the problem.
Raising the risk of ovarian cancer or another cancer called endometriosis-associated adenocarcinoma.
about 40% of those who have trouble getting pregnant have endometriosis. It makes it harder to get pregnant if the tissue growing outside your uterus causes scarring, which can affect your fallopian tubes and keep an egg and sperm from meeting. It can also stop a fertilized egg from implanting in the lining of your uterus. surgery can remove the extra tissue, which may make it easier to get pregnant. an alternative option: assisted reproductive technology (such as in-vitro fertilization) to help you conceive.
Endometriosis reduces the quality of life, accounting for 60% of loss of productivity at work. From work absence, loss in social and sexual aspects of life due to the pain
Those with surgically confirmed endometriosis had a statistically significant greater risk of cardiovascular disease than those without endometriosis. This is due to the concentrations of reactive oxygen due to proinflammatory cytokines.
Neurological association with higher percentages of anxiety and depression, fatigue, hypersensitization to pain. This is due to the changes in electrophysiological activity & gene expression in the brain.
Lower than average BMI due to metabolic changes due to the miR-342–3p and let-7b gene.
Bowel problems. From bloating to sensitive bowels, up unto IBS (Irritable Bowel Syndrome). Inflam

Altogether, the economic burden caused by Endometriosis exceeds est. US$22 billion in the USA alone.

Endometriosis Diagnosis SUCKS!

Gender Pain Gap

Scientists investigating gender differences in pain have found that not only do women report more pain throughout the course of their lifetime, they also experience it in more bodily areas, more often and for longer duration when compared to men, which is likely why the pains are neglected.

With the normalization of pains and the period stigma still very prevalent in our society, they have difficulty expressing their pain, their main symptom.

Historically, women’s health research has suffered largely because men have been making decisions about which conditions to give limited research funding to. people don’t talk about things like period pain, which HERE is obviously a major symptom of the condition.”

Low recognition of endo at the general practitioner (GP) level

Up to 20% of those with endometriosis have concurrent chronic pain conditions, including irritable bowel syndrome, interstitial cystitis/painful bladder syndrome, fibromyalgia, and migraines.

Prior to attributing pelvic pain to endometriosis, bowel, bladder, psychiatric, and musculoskeletal etiologies should be ruled out.

Given that endometriosis may be a diagnosis of exclusion, the diagnosis of endometriosis in someone with pelvic pain is often delayed and stretches over several years.

If these symptoms are dismissed as primary dysmenorrhea, which it may be for some people, then a doctor won’t investigate further and this contributes to the delay in diagnosis and timely management of symptomatic endometriosis normalization of pain.

Stigma

Painful periods have been normalized for far too long and there aren’t enough tools to catch the disease early. The fact that doctors may take woman’s pain less seriously than men’s, which can lead to a late diagnosis or no diagnosis at all. Women end up “getting used” to our pain and stop bringing it up with doctors and OBGYNs. Growing up, most women are told that our pain was normal and that we just had to “suck it up”, so people stopped talking about it transparently.

“embarrassing”, “stigmatized”, “tolerate it”

The prime concern is the uncertainty of normal versus abnormal symptoms and physician-centered causes such as normalization of patient symptoms

Silent Endometriosis

Even though its main symptom is period pains, silent endometriosis is still s thing. This occurs when a patient has no overt, typical symptoms, so healthcare professionals don’t think to check for it. Research published in the Journal of Assisted Reproduction and Genetics suggested that 20 to 25 percent of endometriosis patients are asymptomatic. Making it even harder to diagnose even though whole-body inflammation do not stop for asymptomatic patients.

Misdiagnosis

Endometriosis symptoms are nonspecific and overlap with other gynecologic and gastrointestinal diseases, leading to long diagnostic delays, or are misdiagnosed with diseases like dysmenorrhea and irritable bowel syndrome.

In a prospective study that the World Endometriosis Research Foundation (WERF) conducted, we found that 50% of people that were referred with bowel issues to a gastroenterologist actually had endometriosis and not bowel issues. Those with painful symptoms need to see a gynecologist and it’s not about seeing any gynecologist

😖Diagnosis today is inefficient & unreliable

Pelvic exam

During a pelvic exam, your doctor manually feels (palpates) areas in your pelvis for abnormalities, such as cysts on your reproductive organs or scars behind your uterus.

In several studies, visual diagnosis of endometriosis has been demonstrated to be unreliable. Only 54–67% of suspected endometriotic lesions are confirmed histologically, and 18% of patients clinically suspected to have endometriosis have no evidence of endometriosis on pathology. Indeed, a 2004 metaanalysis which assumed a 20% prevalence of endometriosis found that “a positive finding on laparoscopy will be incorrect in half of the cases.”

👎🏻 Often it’s not possible to feel small areas of endometriosis unless they’ve caused a cyst to form.

2. Ultrasound

This test uses high-frequency sound waves to create images of the inside of your body. To capture the images, a device called a transducer is either pressed against your abdomen or inserted into your vagina (transvaginal ultrasound). Both types of ultrasound may be done to get the best view of the reproductive organs.

👍 cheaper
👎🏻User-dependent + A standard ultrasound imaging test won’t definitively tell your doctor whether you have endometriosis, but it can identify cysts associated with endometriosis (endometriomas).

3. Magnetic resonance imaging (MRI)

An MRI is an exam that uses a magnetic field and radio waves to create detailed images of the organs and tissues within your body. detailed information about the location and size of endometrial implants.
👍 more accurate than an ultrasound

👎🏻 inefficient , expensive ( $400 to $10500 ), not accessible

4. Laparoscopy ⚱️

The “gold standard” for endo🥇 Surgery for a procedure that allows the surgeon to view inside your abdomen (laparoscopy). While you’re under general anesthesia, your surgeon makes a tiny incision near your navel and inserts a slender viewing instrument (laparoscope), looking for signs of endometrial tissue outside the uterus.

Laparoscopy can provide information about the location, extent, and size of the endometrial implants. Your surgeon may take a tissue sample (biopsy) for further testing. Often, with proper surgical planning, your surgeon can fully treat endometriosis during the laparoscopy so that you need only one surgery.

👍🏼only way to confirm a diagnosis of endometriosis. The most common surgery is called laparoscopy.

👎🏻 invasive, leaves scares, people are scared which makes them reluctant to do the procedure, expensive

In several studies, visual diagnosis of endometriosis has been demonstrated to be unreliable. Only 54–67% of suspected endometriotic lesions are confirmed histologically, and 18% of patients clinically suspected to have endometriosis have no evidence of endometriosis on pathology. Indeed, a 2004 metaanalysis that assumed a 20% prevalence of endometriosis found that “a positive finding on laparoscopy will be incorrect in half of the cases.”

The Ideal Future, the Moonshot 🌙

Diagnosis in a manner that’s easy, affordable, and non-invasice.

We’re developing a solution that will make endometriosis diagnosis 70080x faster & 320x cheaper.

Introducing ENDOLOOP.

ENDOLOOP is a moonshot startup idea we created. Our goal is to end the loop of people suffering from endometriosis unknowingly by revolutionizing diagnosis.

Our testing method is non-invasive, rapid, and affordable.

If a patient experiences the symptoms of endometriosis, they simply travel to their closest point of care — this could be any primary care doctor or pharmacy.

A doctor, nurse, or pharmacist would extract a few drops of blood with a finger prick test, and the patient will receive a positive or negative test result for endometriosis within one hour!

MicroRNA Biomarkers

Scientists have recently begun to investigate the role of microRNAs (miRNA) as biomarkers for a host of diseases, especially cancers.

To understand the role of miRNAs, remember how genes translate to biological functions: when translating our genetic code (DNA) into a cell’s biological functions, the DNA is transcribed into messenger RNAs (mRNAs) which are converted to proteins by ribosomes.

MicroRNAs are short, single-stranded, RNA molecules that help cells degrade or block the translation of mRNAs, effectively stopping the production of certain proteins. This can block or alter cellular functions and lead to some of the symptoms of endometriosis.

Promising new research has indicated that levels of various miRNAs, when taken together, can serve as extremely accurate biomarkers for endometriosis. The cytokines causing the inflammation increase and decrease the concentrations of these certain miRNAs we will detect. These miRNAs stand as mechanisms that induce/inhibit this inflammation.

miR-196a, miR-29c, and miR-9 are overexpressed in those with endometriosis -> decreased concentrations of PR-B and its target genes = mechanism for progesterone resistance
127,129 elevated concentrations of miR-451c cause cell proliferation and survival.
127 Additionally, a study found that increased concentrations of serum miR-125–5p and decreased concentrations of let-7b led to increased production of inflammatory cytokines in those with endometriosis

Our proposed 3 step process :

Finger prick for blood sample

Instead of requiring an extensive, expensive surgery (laparoscopy) to diagnose endometriosis, testing blood obtained via a fingerprick device provides an effortless and widely accessible method of retrieving blood for analysis. This does not require a doctor or specialist: any pharmacist or nurse is perfectly capable of retrieving the blood in this way.

Any blood is acceptable for miRNA diagnosis because these cells are in circulation. They impact not only the uterus.

Microfluidic chip to detect miRNA presence

Left: Schematic of a microfluidic chip (μPAD) from blood sample to colorimetric output of miRNA level. Right: Visualization of blood separation process in microchannels of the μPAD.

A core element of EndoLoop’s technology consists of microfluidic chips: an emerging technology that can translate a tiny blood sample into detailed insights to make a diagnosis.

Commonly known as “labs on a chip,” microfluidic chips are small devices with micro-channels that can manipulate liquids and test them for certain substances.

In this case, it can separate a liquid sample into its components and detect the presence of biomarkers.

EndoLoop specifically uses microfluidic paper-based analytical devices, or µPADs, because they are low-cost and easily disposable after use. At high production quantities, we expect each paper microfluidic device to have a materials and manufacturing cost between $1 and $3.

The microfluidic chips perform two essential functions in the test for endometriosis: separating the plasma from the blood sample, and detecting the presence of target miRNAs.

Without microfluidics, separating a blood sample into the red blood cells and the surrounding liquid, or plasma, requires spinning a large blood sample at high speeds with expensive equipment. Paper-based microfluidic devices can perform rapid separation of blood from plasma with a very small sample.

Similarly, microfluidics overcomes the current challenges with the detection of miRNA levels. Currently used mechanisms for testing for miRNAs are complex and expensive, requiring technologies like qPCR (similar technology to what is used for COVID-19 testing) and microarrays (which require specialized equipment).

The microfluidic chip, on the other hand, makes the detection of miRNA levels quite simple. Each indicator has an enzyme with a miRNA template that binds to the target miRNA if it is present in the plasma, catalyzing a chemical reaction that changes color.

Higher concentrations of each miRNA will result in a deeper color in its respective indicator, creating a colorimetric detector for the levels of each endometriosis-associated miRNA (marked 1–6 in the diagram).

Once the microfluidic chip completes the analysis of the blood sample, the person administering the test would use EndoLoop’s app and take a picture of the microfluidic chip.

This app would be available on smartphones or computers, and would not depend on internet access.

The software would automatically translate the colorimetric information from the chip into a numerical value for the concentration of each miRNA (this would be resilient to different lighting conditions and cameras).

Once the microfluidic chip completes the analysis of the blood sample, the person administering the test would use EndoLoop’s app and take a picture of the microfluidic chip.

This app would be available on smartphones or computers, and would not depend on internet access.

This information is then put into a random forest algorithm, which classifies the patient as either positive or negative for endometriosis ✅

We use a random forest algorithm because all the biomarkers assessed were strongly associated with disease status, thus we can rank all the variables rather than obtaining nonzero coefficients from other types of regression models.

The gap between ENDOLOOP’s vision and today’s reality.

miRNA Biomarker research

Current research on miRNA biomarkers for endometriosis place the potential accuracy at up to 94–100% when the concentrations of three-six miRNA are analyzed with machine learning algorithms. However, there is no consensus on which biomarker(s) to rely on for the endometriosis diagnosis, since most research to date has been relatively small-scale and not included the full range of potential endometriosis patients; for instance, the efficacy of these biomarkers has not been investigated for pre-symptomatic endometriosis patients. The success of EndoLoop relies on a large-scale trial to definitively determine the most reliable suite of miRNA biomarkers for use in our testing and ensuring that the accuracy for this is extremely high even at early stages of endometriosis.

Artificial intelligence analysis of microfluidic output

The use of image processing to analyze the colorimetric output of microfluidic chips has been validated by recent research. However, further investigation is required to ensure that this is accurate enough for the machine learning algorithm to precisely predict whether a patient has endometriosis.

Endometriosis education

Our diagnosis system is only effective if people know to get tested for endometriosis and what symptoms to look out for. Once we finalize and validate the diagnostic technology, we plan to engage doctors around the world in a wide-scale endometriosis education campaign to ensure that testing for endometriosis through our method becomes commonplace.

Market Opportunity

The fertility market is expected to reach $36B+ by 2026, and the endometriosis market alone is estimated to reach $2.4B+ by 2026.

Though the cost of the paper microfluidic device itself is $1 to $3, and the disposable lancet costs a few cents, there are numerous other costs that will go into building a company like EndoLoop, including the development of the AI algorithm, clinical trials, distribution, and much more.

Additionally, the test provides an extraordinary value because the laparoscopy to receive an endometriosis diagnosis would cost between $5000 and $9000. $15–30 range is the standard cost for simple in-hospital / doctor’s office tests (e.g. urine analysis, blood count. For these reasons, we choose to price EndoLoop’s test at $25 per test. We want it to affordable and easy to access.

We anticipate that an initial expansion to the United States could achieve $1.2M in annual revenue, which could increase to $14M soon after as the test becomes more prevalent and spreads to other countries.

Bringing EndoLoop To Market | R&D Perspective

Large-scale data collection

The first step to making Endoloop a reality is gathering large amounts of mciroRNA data from patients with and without endometriosis.

We would run microRNA tests through standard lab procedure (qPCR rather than microfluidics) and obtain the levels from patients for ~15–20 candidate miRNAs that could potentially serve as the most accurate biomarkers.

With this data (need on the order of 5k-20k data points, with equal numbers of patients with and without endometriosis), we can determine which miRNA levels correlate most closely with endometriosis.

After deciding on 4–8 microRNAs that together can provide the best diagnosis, we input the data collected (80/20 split into train and validation) into the random forest classifier as training data and ensure that the sensitivity and specificity each exceed 90% on both the training and validation data.

2. Colorimetric validation
The second step to making EndoLoop a reality is ensuring that the colorimetric detection of miRNAs on the microfluidic chip provide levels of microRNAs that are accurate enough for the ML algorithm to make the correct diagnosis.

This will involve testing the same blood samples with:

qPCR (gold standard) for a baseline result
Colorimetric + Spectroscopy (ensure the precision of colorimetric reaction on microfluidic chip)
Colorimetric + phone camera (with computer vision processing) to ensure that a smartphone camera can correctly obtain accurate miRNA levels, adjusting for different cameras + lighting conditions

Essentially, R&D work will be required to fine-tune the microfluidic reaction sensitivity and the image processing pipeline to ensure that we have accurate enough data to work with.

3. Clinical trials
After the full pipeline including microfluidic chip manufacturing, image processing, and AI algorithm has been developed, the EndoLoop technology needs to be tested on a large scale to ensure its efficacy.

An important question this trial could answer is whether this allows for a diagnosis at the early stages of the disease before a trial participant had even received a diagnosis via a standard method used today (laparoscopy, ultrasound, etc).

As we solve the cost and accessibility of diagnostic tools for endometriosis, no one wouldn’t have to suffer for 8 years anymore thinking that their pains are ‘just period stuff’ .

But our job isn’t done. We need to start validating those suffering between us. We need to raise awareness so that more people will be aware and more funding will be put in order to solve this underresearched disease and hopefully, find a cure.

This paper was made in collaboration with Liesl Anggijono, Adit Shah, Oscar Petrov, and Sriya Chintalapalli for the TKS Moonshot Hackathon 2021.

We ARE currently NOT A REAL COMPANY. This is simply an idea we worked on for a hackathon project.

Withal, I decided to extend the research individually post-hackathon and maximize our findings that didn’t reach their full potential due to time constraints.

I personally want to bring up that we’ve used the word women twenty times and woman three times in our original hackathon submission. We understand and want to emphasize that we recognize endometriosis symptoms and endometriosis happens to those who do not identify as a woman. ( But ones still with uteruses ). I have tried to minimize the use of the word ( reduced women to 6x and woman to 1x ) to be inclusive and not exclude those who do not identify as a woman but still experience endometriosis. But I can’t seem to find a gender-neutral word for some terms tied with statistics that would still be justifiable to the data I’m presenting. So if you have any suggestions, please email me at liesl.anggijono@gmail.com. Thank you.

Although scientific papers have been cited as links throughout this medium article, I want to highlight and thank papers that were crucial in our project and led to the major findings of this project :

(1) Taylor, H. S., Kotlyar, A. M., & Flores, V. A. (2021). Endometriosis is a chronic systemic disease: clinical challenges and novel innovations. The Lancet, 397(10276), 839–852.

(2 ) Moustafa, S., Burn, M., Mamillapalli, R., Nematian, S., Flores, V., & Taylor, H. S. (2020). Accurate diagnosis of endometriosis using serum microRNAs. American Journal of Obstetrics and Gynecology, 223(4), 557.e1–557.e11.

(3) Cosar, E., Mamillapalli, R., Ersoy, G. S., Cho, S., Seifer, B., & Taylor, H. S. (2016). Serum microRNAs as diagnostic markers of endometriosis: a comprehensive array-based analysis. Fertility and Sterility, 106(2), 402–409.

(4) Fakhri, N., Abarghoei, S., Dadmehr, M., Hosseini, M., Sabahi, H., & Ganjali, M. R. (2020). Paper based colorimetric detection of miRNA-21 using Ag/Pt nanoclusters. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 227, 117529.

(5) Cai, X., Zhang, H., Yu, X., & Wang, W. (2020). A microfluidic paper-based laser-induced fluorescence sensor based on duplex-specific nuclease amplification for selective and sensitive detection of miRNAs in cancer cells. Talanta, 216, 120996.