Bio-Conscious builds clinically-validated AI that reads continuous glucose data to predict, prioritize, and prevent metabolic disease — years before symptoms appear.
Proof, not promises
As part of a $17.3M PacifiCan investment across eight B.C. technology companies, Bio-Conscious Technologies received $1.5 million to commercialize Endobits — its AI platform that detects medical events at their earliest stage for patients with diabetes — for hospitals across Canada and the United States.
Read the announcement →CGM adoption is exploding, but the data dies in dashboards. Healthcare still reacts to chronic disease instead of preventing it. The unused signal inside glucose data is one of the largest untapped opportunities in metabolic health.
Our defensibility isn't one model; it's a stack. Each level builds on a proprietary understanding of glucose metabolism, compounding into an early-detection engine that extends far beyond diabetes.
The model is the easy part. The moat is everything around it — and most of it takes years and clinical relationships to assemble. We already have.
A generic foundation model can predict the curve. Only forward clinical-outcome labels tied to the glucose signal tell you what that curve meant for the patient — and that labeled, longitudinal dataset takes years of clinical partnership to build, not a quarter of engineering.
Not a research notebook. A prediction engine hardened inside a live clinical product against real-world CGM noise, gaps and sensor switches — the unglamorous work that separates a demo from a deployment.
Proprietary clustering of curve shapes into distinct metabolic phenotypes — the layer that turns a forecast into personalization, with early longevity indications competitors don't have.
We integrate Dexcom, Abbott and Senseonics — including the only implantable CGM. A single device-maker building this in-house builds a walled garden; we are the neutral intelligence layer that spans all of them.
Nearly a decade of R&D and research presented at ADA's Scientific Sessions, born from a study at BC Children's Hospital. In healthcare, trust is earned slowly — and it can't be cloned in a sprint.
Consumer CGM apps stop at the reading. General-purpose models can describe the curve but not what it means. We are built for the part that changes outcomes.
| Capability | Typical CGM apps | General AI models | Bio-Conscious |
|---|---|---|---|
| Prediction horizon | —Reactive, after the fact | ~Curve only | ✓12 hours ahead |
| Sensor coverage | ~Single vendor | ~Varies | ✓Dexcom · Abbott · Senseonics, incl. implantable |
| Outcome labels | ✗None | ✗Curve, not outcomes | ✓Forward clinical-outcome labels |
| Personalization | ~Population averages | ✗Generic | ✓Glucotype phenotyping |
| Clinical validation | ~Limited | ✗Research-stage | ✓Live product, ADA-presented |
| Reimbursement | ~Varies | ✗Not applicable | ✓RPM / CCM aligned |
✓ built-in · ~ partial · ✗ absent — illustrative category comparison
Endobits eliminates data overload for clinics — automating triage, surfacing risk, and generating AI recommendations with seamless CGM and EHR integration. It's live and generating revenue today: the wedge into our larger prevention platform.
Endobits is HIPAA-compliant and runs on top of FDA-cleared sensors from Dexcom, Abbott and Senseonics. It is designed to operate within the established Medicare remote-patient-monitoring and chronic-care-management framework — so the clinical value it creates is billable today, not contingent on a new reimbursement category.
For any CGM platform, the cost of acquiring a patient dwarfs the cost of keeping one — and the patients who churn are the ones who stop seeing value in their data. Endobits turns raw readings into something a patient acts on every day, which is what keeps them on-sensor. The math at platform scale is not subtle.
Illustrative model — adjust the inputs. Annual revenue retained = base × revenue/patient × retention gain. Scenario inputs, not a forecast.
A single average — HbA1c, mean glucose — collapses a living signal into one number, and hides the volatility that actually damages tissue. Bio-Conscious models the whole curve: its shape, timing, variability and trajectory. We cluster patients into distinct glucotypes, then forecast where each curve is heading. That is the shift the field has been waiting for — from measuring the past to modeling the future.
Same number on the chart. Opposite risk in the body. The average is what every clinic sees — the shape is what we model.
There is no single “diabetes.” By clustering curve shapes into glucotypes, our models separate patients who look identical on paper into distinct metabolic phenotypes — each with its own risk profile and its own intervention. This is the engine behind moving care from reactive to predictive to preventive.
The thesis
Glucose is the most-tracked, least-used signal in medicine. We built the half that acts on it.
Dysregulated glucose metabolism drives the core mechanisms of aging itself — from cellular senescence to mitochondrial decline. That makes the 5th vital sign a lever not just for disease, but for healthspan. We built an interactive guide to show exactly how the connections map.
Insulin/IGF-1, mTOR, AMPK and the sirtuins are the glucose- and nutrient-sensing machinery. Chronic hyperglycemia keeps these master switches mis-set.
Endobits lens The hallmark our data reads most directly — it is the core signal we model.
Glucose variability and glucotoxicity flood mitochondria with reactive oxygen species, degrading the cell's energy production over time.
Endobits lens Curve volatility is an early, non-invasive proxy for mitochondrial strain.
Sustained hyperglycemia and advanced glycation end-products (AGEs) push cells into a senescent state — and senescent cells in turn handle glucose worse.
Endobits lens Repeated excursions flag an accelerating senescent load.
High glucose activates NF-κB and a cascade of inflammatory cytokines — the “inflammaging” engine that links metabolism to nearly every age-related disease.
Endobits lens Glycemic instability tracks tightly with metabolic inflammation.
Glucose flux sets the acetyl-CoA and NAD+ pools that rewrite DNA methylation and histone marks — the molecular basis of “metabolic memory.”
Endobits lens Glucotype patterns hint at the direction of epigenetic drift.
Glycation cross-links and misfolds proteins, while overwhelming the chaperone and clearance systems meant to keep the proteome intact.
Endobits lens Cumulative exposure above range compounds glycation burden.
Glucose excess suppresses autophagy through mTOR; the troughs between meals and overnight are when cellular self-cleaning reactivates.
Endobits lens Time-in-trough maps to the body's autophagy windows.
Oxidative stress driven by hyperglycemia damages DNA directly and impairs the repair pathways that protect the genome.
Endobits lens High-variability signatures correlate with oxidative DNA load.
A dysglycemic environment impairs progenitor-cell function and the tissue repair that depends on it — visible in diabetic wound healing.
Endobits lens Chronic exposure is an upstream marker of regenerative decline.
AGE–RAGE signaling and shifted adipokine balance distort the chemical messaging between cells and tissues across the body.
Endobits lens Glycemic state is a systemic input to this signaling network.
The oxidative load from glucose excursions is associated with faster telomere shortening; diabetes consistently tracks with shorter telomeres.
Endobits lens A longitudinal signal we are positioned to study at scale.
Diet and glucose reshape the gut microbiome, which in turn feeds back on glucose metabolism — a bidirectional loop only now being mapped.
Endobits lens CGM is the highest-resolution window into that loop.
hallmarks of aging with a documented link to glucose metabolism. The most-tracked signal in medicine is also one of the most upstream.
A sequenced plan: prove and scale the wedge that already earns, then extend the same engine outward — deliberately, one tier at a time.
A team that has spent nearly a decade turning continuous glucose data into clinical decisions.
“Integrating CGM data with AI like Endobits is a significant advancement in diabetes care — giving clinicians actionable insight in real time.”
Glucose is the most-tracked, least-used signal in medicine. We've spent nine years turning it into a defensible, revenue-generating engine for disease prevention and longevity. And we're just reaching Level 3.