CelerisAI · May 2026
Millions of people suffer from heart attacks, lung diseases, and breathing conditions that share the same hidden root cause. CelerisAI has identified three molecular switches that control this crisis — and is using AI to design a single medicine to address all three at once.
Every second, somewhere in the world, a person's body runs out of oxygen in a place it shouldn't. A blood vessel gets blocked. Airways narrow. Lung tissue scars. And when that happens, something unexpected follows — the body doesn't just weaken. It turns on itself.
The immune system, designed to protect us, goes into overdrive. It floods the body with inflammatory signals — far more than are needed — and in doing so, destroys healthy tissue in organs that had nothing wrong with them. A heart attack damages the kidneys. A lung condition destroys the heart. The original problem triggers a chain reaction that medicine, until now, has had no precise way to stop.
This is the shared root of four of the world's most devastating diseases — and it is the problem CelerisAI was built to solve.
When an artery is blocked, the tissue it feeds instantly runs out of oxygen. Cells die, organs fail, and a wave of inflammation spreads far beyond the original blockage — damaging the heart, kidneys, and blood vessels simultaneously.
Damaged airways can't deliver enough oxygen. The body lives in a state of constant low-grade oxygen starvation that relentlessly inflames and destroys lung tissue — with no treatment to stop it progressing.
Scar tissue slowly replaces healthy lung, making it harder to breathe with every passing month. The inflammation driven by this oxygen shortage accelerates the very scarring that's choking the patient. A lung transplant is often the only option — and suitable donors are desperately scarce.
Blood vessels in the lungs thicken and narrow under constant inflammatory pressure, forcing the heart to work impossibly hard until it fails. Patients gasp for breath doing ordinary tasks. There is no cure.
No approved medicine currently stops the underlying inflammatory cascade driving all four conditions.
The pandemic gave us an unforgettable lesson in how oxygen starvation and runaway inflammation work together. The virus itself wasn't the primary killer in severe COVID-19 cases. What killed people was the chain reaction it started — the immune system's catastrophic overreaction to the oxygen crisis the virus caused.
Patients whose lungs couldn't absorb enough oxygen sent distress signals that triggered a "cytokine storm" — a massive inflammatory explosion that attacked not just the lungs, but the heart, kidneys, liver, and blood vessels simultaneously. Many patients who survived the initial viral infection died weeks later from this inflammatory aftermath, not from the virus itself.
The steroid drug dexamethasone — which dampens inflammation broadly — reduced severe COVID-19 deaths by 35%. This confirmed what scientists had long suspected: inflammation, not the original trigger, was doing most of the killing. But broad steroids come with serious side effects. What was needed was a precise, targeted way to stop the inflammatory cascade without suppressing the entire immune system. That is exactly what CelerisAI is building.
When oxygen runs out inside a cell, three molecular systems are thrown into disarray — and all three need to be brought back under control at the same time. Think of them as three switches that all get flipped the wrong way when the body is under oxygen stress.
Deep inside immune cells sits a molecular alarm system called NLRP3. In a healthy body, it fires when there's a genuine threat, calls in the immune system, and then switches off. But in ischemia, COPD, fibrosis, and PAH, this alarm gets stuck in the "on" position — constantly broadcasting an emergency signal that the immune system responds to, over and over, long after the original crisis has passed.
The result is a continuous flood of inflammatory chemicals that destroy healthy tissue in the heart, lungs, kidneys, and blood vessels. It is the primary engine of the damage that kills people — not the blocked artery, not the scarred lung tissue, but this relentless, misfiring alarm. Switch off NLRP3, and you stop the fire at its source.
NLRP3 is also responsible for a particularly violent form of cell death — where cells essentially explode, releasing their contents into surrounding tissue and spreading the alarm signal even further. This self-amplifying destruction is why these diseases progress even when the original trigger has been partially treated.
Every cell in your body has a built-in oxygen detector — a protein called HIF-1α. When oxygen levels drop, HIF-1α activates and acts like an emergency commander, switching on dozens of survival programmes: growing new blood vessels, changing how the cell makes energy, producing more red blood cells. In the short term, this is genuinely protective and essential.
The problem is that in chronic oxygen-related disease, HIF-1α never switches off. It stays permanently active, and the "emergency programmes" it keeps running eventually start causing damage — thickening blood vessel walls, driving the scarring process in the lungs, and amplifying the very inflammatory signals that NLRP3 is already flooding the body with.
CelerisAI doesn't aim to simply block HIF-1α — that would also knock out its protective effects. The goal is to calibrate it: keep the helpful emergency responses running while preventing the chronic overactivation that drives disease. This level of precision is only achievable through AI-designed molecular engineering.
Here is perhaps the most important thing to understand about these diseases: the body already has the tools to protect itself. There is a molecular system — controlled by a protein called Nrf2 — that, when fully active, switches on over 200 protective and anti-inflammatory proteins simultaneously. It is the cell's most comprehensive natural defence programme.
One of the most important proteins Nrf2 activates is called HO-1 (Heme Oxygenase-1). When HO-1 is produced, it creates three protective byproducts — including a tiny amount of carbon monoxide. At the microscopic quantities the body produces naturally, carbon monoxide is not toxic. It is a powerful signalling molecule that actively reduces inflammation, prevents unnecessary cell death, and neutralises the chemical damage caused by the inflammatory process.
In ischemia, COPD, fibrosis, and PAH, the Nrf2 system is suppressed — switched off at precisely the moment the body needs it most. Over 200 protective proteins that could be defending the cell are never produced. CelerisAI's molecule is designed to switch this system back on, restoring the body's own most sophisticated natural shield — from the inside.
The body already knows how to protect itself. It has the tools. In these diseases, those tools get switched off. Our job — with AI — is to switch them back on.
— Dr. Chintan Raval, CEO & Founder, CelerisAI
Designing a single medicine that precisely controls all three of these switches simultaneously — without unwanted side effects — is one of the hardest problems in drug discovery. The traditional approach of testing thousands of molecules in a laboratory, one by one, would take decades and billions of pounds. The vast majority of candidates would fail.
CelerisAI uses artificial intelligence and machine learning to do this computationally. Our AI models can screen millions of potential molecules virtually, predicting how each one would interact with all three targets at once, identifying toxic candidates before any lab work begins, and generating novel molecular designs that no human chemist would have thought to try. The result: faster discovery, dramatically lower costs, and a much higher probability of finding a molecule that actually works.
AI compresses years of laboratory screening into months of computation — getting promising molecules to the clinic far sooner.
By eliminating failing molecules computationally — before expensive physical testing — AI dramatically reduces the cost of finding what works.
AI toxicity prediction is up to six times more accurate than conventional screening — catching dangerous candidates before they reach patients.
The diseases CelerisAI is targeting — ischemia, COPD, pulmonary fibrosis, and PAH — affect over 500 million people globally. They are among the leading causes of death and hospitalisation worldwide. And for all of them, the current standard of care manages symptoms without addressing the cellular crisis driving the disease forward.
We believe that is not good enough. The body's own molecular machinery holds the key to stopping the destruction — and AI gives us the precision to unlock it. That is what CelerisAI is building.