PainCube AI Pain Management Platform
We aim for a world where no pain goes untreated.
Stop asking patients, ‘How much does it hurt?’
We ask the body through objective biosignals, not self-report.
PainCube measures pain on a 0–10 scale from biosignals, predicts onset 30–60 minutes ahead, and provides data-driven analgesic dosing decision support.
01Why PainCube
In clinical practice, 80% of patients do not receive adequate pain management.
No objective way to measure pain
And no precise standard for analgesic dosing
Pain assessed by subjective self-report
Measured against imprecise criteria
Analgesics prescribed by clinician intuition
02How PainCube Works
01
A sensor that attaches at various sites per patient: chest, upper arm, wrist.
02
ICU noise removal; pre-onset pain prediction from biosignal and pain patterns.
03
Dashboard-based pain alerts and dosing support, cutting clinician resource by 80%.
03PainCube Sensor + Data
12+ biosignals and medical records collected across Johns Hopkins (32 beds) and Korea University Anam Hospital (24 beds).
Data source
Wear type
Biosignal data (6+)
Patient-record data (6+)
04TCAtt-PainNetPain Quantification & Prediction AI
Learns pre-onset biosignal patterns
And pain-reduction patterns after a dose
Produces a 0–10 pain index (CPI)
Predicts onset 30–60 min in advance
Secures the golden time to respond
Time for clinicians to prescribe and administer analgesics
05DDCAESignal Denoising AI
Encoder-decoder architecture
Exploits the regularity of pain against irregular noise
Removes noise mistaken for pain
ECMO operation, body motion, and similar artifacts
Extracts pure pain signals
Reduces false alarms and raises prediction accuracy
06CDSSClinical Decision Support System
Alerts 30–60 min before pain onset
Secures the golden time to respond
Data-driven analgesic dosing decisions
Includes drug type, timing, and dose
Real-time patient monitoring dashboard
Supports patient-tailored pain management
07Clinical Impact
Before adoption: 34 min per patient
Pain management time for a single patient
*Based on a 12-hour shift (SCCM PADIS 2018; Gélinas et al., 2006)
After adoption: 7 min per patient
An 80% reduction in pain management time
*Routine rounds, pain assessment and dosing greatly streamlined
Establishes the adoption rationale for hospitals
Backed by time and cost savings
Team
Clinicians and engineers from Johns Hopkins, Korea University Anam Hospital, and StellarCube.
CEO
Professor, AI Application, Hansung University
CTO
22+ years leading service development across AI, healthcare, senior care, gaming, and portals
CSO
17+ years in data & AI strategy; led global R&D projects exceeding KRW 1B
Head of StellarCube AI Medical Center · Prof. of Thoracic Surgery, Korea University
Prof. of Neurosurgery, Johns Hopkins University
Project Manager
4+ years in AI product management and data engineering · ex-Naver Cloud
Korean & English IR available. We'll follow up by email.
