Pre-Seed  ·  Medical AI  ·  Los Angeles

Detecting
Biological Collapse
Before It Happens.

An AI system that identifies physiological disorder in ICU patients up to 10 hours before cardiac arrest — validated on 1,310 confirmed cases from MIMIC-IV.

0.70 AUC Score
10 hrs Median Detection Window
1,310 Validated Patient Cases
1 Provisional Patent Filed
Scroll

Hospitals are reacting.
We need to predict.

Existing early warning systems like MEWS and NEWS2 are rule-based threshold tools. They wait for a value to breach a limit. By then — it's often too late.

The real signal is not the breach. It's the disorder building toward it.

Late Detection

Current systems detect deterioration after significant physiological decline has already occurred

🔔

False Alarm Burden

Threshold-based alarms generate high false positive rates causing alarm fatigue in clinical staff

📊

No Disorder Awareness

No existing system measures biological signal order — the structural coherence that collapses before catastrophic events

🏥

Reactive Medicine

Hundreds of thousands of in-hospital cardiac arrests annually — most preceded by detectable physiological signals

Biological Signal Order.
A New Measurable Property.

Unlike systems that monitor absolute values, Regenesis-OS measures the structural organization of biological signals over time. As a system approaches collapse — disorder emerges.

01
📡

Signal Acquisition

Continuous vital sign data ingested from existing hospital monitoring infrastructure. No new hardware required.

ECGHRBPSpO2RR
02
🗺

Field Representation

Raw signals transformed into spatiotemporal signal fields capturing relationships across time, space, and biological subsystems.

03
📐

Order Metrics Engine

Custom metrics quantify deviation from baseline signal order — entropy, variance, gradient energy, anisotropy, cross-system coupling.

04
🧠

Confidence-Aware Classification

Temporal sequences analyzed to detect and classify five distinct disruption modes with associated confidence values.

05
🔄

Adaptive Control

Closed-loop system selects interventions based on detected pattern and confidence — modulating intensity to biological response.

Five Disruption Modes Detected

Regenesis-OS classifies the type of biological disorder — not just its presence

Diffuse Shock
Gradient Collapse
Localized Spike
Noise Flood
Repetitive Micro-Damage

Built on the World's Most
Rigorous Critical Care Data.

Dataset
MIMIC-IV v3.1
PhysioNet critical care database — the gold standard for ICU research
Model Performance
AUC 0.70
5-fold stratified cross-validation
Detection Window
10 Hours
Median time before cardiac arrest event
Patient Cases
1,310
Confirmed ICU cardiac arrest cases

Signal Sources

Heart Rate
Non-Invasive BP Systolic
Arterial BP Systolic
Signal Entropy
Signal Variance
Provisional Patent Filed

USPTO — Systems and Methods for Detecting and Restoring Biological Signal Order Using Adaptive Closed-Loop Control

Cardiac Arrest Is
Only The Beginning.

The same disorder signal that precedes cardiac arrest precedes every major physiological collapse event. Regenesis-OS is a platform.

Phase 1 — Now

ICU Cardiac Arrest Detection

Deploy in ICU environments as a software layer on existing monitoring infrastructure. Clinical validation. FDA pathway.

$2B–$5B addressable market
Phase 2

Universal Collapse Prediction

Expand to sepsis, respiratory failure, renal failure, hepatic collapse. One platform predicting all major in-hospital deterioration events.

$50B–$100B addressable market
Phase 3

Chronic Disease Early Warning

Wearable integration detects pre-Alzheimer, pre-cancerous, and metabolic disorder years before clinical diagnosis.

$200B–$500B addressable market
Phase 4

Biological Operating System

Continuous whole-body biological stability monitoring. The intelligence layer inside every hospital bed and home device on earth.

$1T+ platform opportunity

Built by a Signal
Detective.

Justin C. Radford-Berry spent seven years as an independent quantitative researcher — finding anomalous patterns in complex, noisy, real-time data streams.

He noticed something: the methodology for detecting market disorder and the methodology for detecting biological disorder are structurally identical. Both are entropy detection problems in time series data.

He turned that lens on ICU telemetry data — and found a signal nobody had measured before.

🎓 BA Environmental Business — University of Redlands
📊 7+ Years Independent Quantitative Research
Provisional Patent Filed — USPTO 2024
🏥 MIMIC-IV Research Access Secured
📍 Los Angeles, California
"

Biological systems don't fail suddenly. They warn you. Rising entropy. Increasing variance. Loss of signal organization. The disorder precedes the collapse — every single time. We built the system that listens.

— Justin C. Radford-Berry, Founder & CEO

Our Mission

Detect biological collapse before it happens — giving clinicians the time they need to intervene, and giving patients the chance they deserve.

Join Us in Redefining
Critical Care.

$1.5M
Pre-Seed Round
Valuation Cap $8M
Instrument SAFE
Minimum Check $50,000
Runway 18 Months
Request Materials

Use of Funds

ML Engineering 40%
Dataset Expansion 25%
Clinical Partnerships 20%
Regulatory & Legal 10%
Operations 5%

Next Milestones

Q1
Dataset expanded to 3,000+ patients
Q2
AUC ≥ 0.80 achieved with expanded feature set
Q3
Clinical collaboration with ICU partner established
Q4
Real-time monitoring prototype deployed in pilot

Let's Talk.

Whether you're an investor, clinician, researcher, or potential partner — we want to hear from you.