FDA-Cleared40+ Academic University Hospitals in Korea

Stroke AI: End‑to‑End.

Real-Time Detection. Predictive Intelligence.

Fast, accurate, consistent AI triage for acute stroke response and transfer decisions.

[video coming soon]
40+
Academic University Hospitals in Korea
#1
AUC (0.9864)
8
Patents Granted

Hyper Insight AI Detection Solutions

Comprehensive Stroke AI Platform

Hyper Insight™ – ICH

FDA K240353
Intracranial Hemorrhage Detection · NCCT

AI software that triages suspected brain hemorrhage within seconds on non-contrast CT and alerts clinicians to potential findings.

AI Triage
Mobile Alerts
0.9864
AUC
95.45%
Sensitivity
98.47%
Specificity
~16s
Notification

Source: Clinical Study Report, Protocol No. SKAIICH-02, Hyper Insight-ICH FDA Clinical Performance Evaluation, Midstate Medical Center, USA.

Radiological computer-aided triage and notification software. Hyper Insight™ – ICH functions as a notification and triage tool — supporting, not replacing, medical judgment. Final diagnosis and treatment decisions remain the responsibility of qualified medical professionals.

ICH Detection

Medical Insight+™ Brain Hemorrhage

RUOKFDA
AI Hemorrhage Detection
(ICH, SAH, IPH, SDH, EDH)
ROI Visualization
Probability Scores
Volume Calculation
Pediatric
Automatic Report
Mobile Alerts

Not cleared or approved by the US Food and Drug Administration (FDA). This software is not for clinical use in the United States and is provided for research, evaluation, and educational purposes only. The indication for use in pediatric patients (under 18 years of age) is currently under review by the Ministry of Food and Drug Safety (MFDS) and has not yet received regulatory approval. Clinical use in pediatric populations is subject to the outcome of this review.

AI-Assisted Reader Performance Improvement

Patient-level improvement when radiologists use Medical Insight+™ Brain Hemorrhage vs. unassisted reads.

Sensitivity+2.82%
Specificity+1.84%
Accuracy+2.33%
ICC+0.0718

+3.2% accuracy improvement among non-radiologists — the greatest gain across all reader groups.

ICC improved from Good (0.8475) to Excellent (0.9193). All metrics measured at patient level.

Source: Clinical Study Report, Protocol No. SKAIICH-01, Ver. 2.0, Seoul National University Hospital & Ajou University Hospital, Korea. CRIS: KCT0006734.

Top 15 Industry Features Avail.

Automated alerting / real-time notificationsWorklist prioritization / triagePACS integration / automated DICOM routingCloud or hybrid deploymentSecure team communication / collaboration

Industry-standard integration capabilities available with PurpleAI deployments.

Clinical Evidence

Validated by Rigorous Clinical Research

KR

Multi-Center External Validation

(49,841 Patients)
6 hospitals

AI Performance in Patient-wise Analysis

Accuracy0.977
Sensitivity0.944
Specificity0.982
AUC0.992

AI Sensitivity by AIH Subtype

SAH0.954
SDH0.933
EDH0.933
IVH0.994
IPH0.977
Source: Yun TJ, et al. "External validation of deep learning-based intracranial hemorrhage detection." npj Digital Medicine, Springer Nature (2023).
KR

Multi-Center External Validation

(49,841 Patients)
6 hospitals
ScannerAccuracy
GE0.98
Philips0.95–0.99
Siemens0.94–0.98
Toshiba0.99
High accuracy and robustness across a large-scale, multi-institutional dataset.
Source: Yun TJ, et al. "External validation of deep learning-based intracranial hemorrhage detection." npj Digital Medicine, Springer Nature (2023).
KR

Korean Multi-center, retrospective, crossover study

9 reviewers across 3 expertise levels
Without
AI
With
AI
P-
value
Sensitivity94.4%97.2%0.0017
Specificity95.0%96.9%0.0376
Accuracy94.7%97.0%0.0075
Greatest benefit for non-radiologist ED physicians.
US

Performance vs. Peers

Hyper Insight-ICH compared to other similar FDA-cleared AI solutions.
MetricRank
AUC#1
Specificity#2
Speed#2
Sensitivity#3
Internal analysis, FDA database, 2024. Companies anonymized.

Who We Serve

Solutions for Every Emergency Room in Rural and Critical Access Hospitals.

👨‍⚕️

Clinicians

  • Flags hemorrhage in 16 seconds on average. Non-radiologists reach specialist-level accuracy.
  • Non-specialist uplift+3.15% accuracy improvement for non-radiologists.
  • Subspecialty gap coverageAI compensates for absent neuroradiology in rural/off-hours settings.
  • Worklist prioritizationSuspected ICH cases surfaced first — frees attention for highest-risk patients.

Source: Yun TJ, et al. npj Digital Medicine, Springer Nature (2023). p=0.0072.

👥

Care Teams

  • Automated real-time alertsMobile push alerts to your entire care team upon suspected ICH detection.
  • Secure DICOM viewerSmart messaging and team invitations for instant case collaboration from anywhere.
  • Stroke code accelerationSupports faster door-to-imaging and imaging-to-treatment times.
🏥

Hospitals & Administrators

  • Up to 70% cost savings compared to peer solution providers.
  • Reduced dependence on outsourced readsThe primary cost driver for low-volume sites.
  • 12-month zero-cost pilotNo costs for deployment, integration, and licensing. Limited-time offer for US reference building.
  • Cloud-nativeFlexible deployment options.

What Clinicians Say

Trusted by Medical Professionals

It will be very helpful for physicians who lack expertise but must make initial judgments in emergency cases.

Non-Radiology Physician
Korean Academic University Hospital

Successfully detected small findings and correctly identified negatives as negative. The AI software helps clinicians who have difficulty distinguishing beam hardening artifacts.

Neuroradiologist
Korean Academic University Hospital

When detecting small SAH (subarachnoid hemorrhage), the AI assistance helps accurately review many images in a short amount of time.

Radiologist
Korean Academic University Hospital

What impressed me most was the software's ability to distinguish real hemorrhage from look-alikes. Partial volume artifacts — where bone and brain tissue overlap — can easily mimic a small subarachnoid hemorrhage on CT, and normal intracranial calcifications appear hyperdense enough to fool most deep learning algorithms into flagging them as bleeds. These are areas where even experienced neuroradiologists have to slow down and look twice. PurpleAI's system correctly identified both as non-hemorrhage, avoiding false positives that would otherwise erode a clinician's trust in the tool.

Dr. Yoo, MD, PhD
Department of Radiology, Korea University Anam Hospital

The system consistently detected small-volume hemorrhages that are among the hardest findings to catch on non-contrast CT — thin subdural collections along the anterior falx cerebri, trace subarachnoid blood in a single sulcus. These are the cases that get missed on overnight reads or by physicians who don't specialize in neuroimaging. The AI flagged them accurately and localized them with color mapping, giving the reader a clear starting point instead of having to hunt through every slice.

Dr. Kim, MD
Department of Radiology, Eunpyeong St. Mary's Hospital, College of Medicine, The Catholic University of Korea

Testimonials from medical end users at Korean academic university hospitals

See It In Action

Product Demo

ICH DetectedNow

Patient ID: 12847 · NCCT

Urgent Alert2m

Brain aneurysm suspected

Analysis Complete5m

No abnormalities detected

doctor

Cross-Platform

Optimized for all devices

📱
iOS
🤖
Android
team chat
KJ
Dr. Kim

ICH case needs immediate review

2:34 PM
LS

On it. Checking now

2:35 PM
PH
Dr. Park

I'll join the consultation

2:36 PM
Type a message...

Mobile DICOM Viewer

DICOM Access Anytime, Anywhere

Your stroke care team stays connected with real-time mobile notifications and a full-featured DICOM viewer.

Instant Specialist Notification

Auto Alerts for Suspected ICH

Seamless Team Collaboration

Smart Messaging and Invitations for Instant Case Review

Mobile DICOM Viewing

Full Diagnostic-Quality Images Review

Predictive Intelligence

StrokeShieldAI

"Advance from rapid stroke detection toward predictive intelligence that anticipates patient trajectories through multimodal data, transforming acute stroke care from reactive response to proactive decision-making."

StrokeShieldAI Architecture

StrokeShieldAI ArchitectureImaging Data, EHR, Vital Signs flowing into AI Engine producing Result Report and Alert System

Trusted By Leading Institutions

Our Partners

Leading Hospital Partners
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Hospital Partner
Strategic Business Partners

Strategic partners expanding our reach and capabilities

Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Strategic Partner
Advisors
🧑‍⚕️

Joshua N. Goldstein, MD, PhD

Professor, Harvard Medical School · Vice Chair for Faculty Affairs, Dept. of Emergency Medicine, Mass General Brigham · Director, Center for Neurologic Emergencies, MGH. Research focus: hematoma expansion, risk models, early neurological deterioration.

FDA 510(k)KOREA MFDSGMPISO 27001ISO 13485

Get In Touch

Contact Us

We'd love to hear from you

📍

Headquarters

3rd Floor, 18 Teheran-ro 20-gil
Gangnam-gu, Seoul, Korea 06235

✉️

Email

purpleai@purple-ai.co

📞

Phone

+82 10-2923-4270

🌐

Website

purple-ai.co

Distribution Partners (US)

SaveLife.AI · CARPL.ai · VSee

Request a Demo

Terms of Use

Effective Date: [To be determined]

1. Acceptance of Terms

By accessing or using the PurpleAI website and services ("Services"), you agree to be bound by these Terms of Use. If you do not agree to these terms, do not use the Services.

2. Description of Services

PurpleAI provides AI-powered medical imaging analysis software, including Hyper Insight™ and Medical Insight+™ products. These products are intended for use by licensed healthcare professionals and institutions. All products are subject to their respective regulatory clearances and intended use statements.

3. Medical Disclaimer

PurpleAI products are designed to assist — not replace — the clinical judgment of qualified healthcare professionals. AI-generated outputs are advisory in nature and must be reviewed by a licensed clinician before any clinical action is taken. PurpleAI is not responsible for clinical decisions made based on AI outputs.

4. Regulatory Status

Hyper Insight™ – ICH is FDA 510(k) cleared (K240353) for clinical use in the United States. Medical Insight+™ products are KFDA-approved and available for Research Use Only (RUO) in the US market. Products labeled "Pipeline" or "In Development" are not cleared for clinical use in any market.

5. Intellectual Property

All PurpleAI products, documentation, algorithms, and trademarks (including Hyper Insight™, Medical Insight+™, and StrokeShieldAI™) are the exclusive property of Purple AI Inc. Unauthorized reproduction, distribution, or reverse engineering is strictly prohibited.

6. Limitation of Liability

To the maximum extent permitted by applicable law, PurpleAI shall not be liable for any indirect, incidental, special, consequential, or punitive damages arising from your use of the Services.

Privacy Policy

Effective Date: [To be determined]

1. Information We Collect

We collect information you provide directly (name, email, organization, role) when requesting a demo or contacting us. We also collect standard usage data (IP address, browser type, pages visited) through analytics tools.

2. Protected Health Information (PHI)

PurpleAI processes medical imaging data exclusively within HIPAA-compliant, ISO 27001-certified infrastructure. All PHI is encrypted in transit and at rest. We do not sell, share, or use PHI for marketing purposes. BAA agreements are available for all US healthcare customers.

3. Data Retention

Contact form data is retained for business relationship management. Medical imaging data retention follows customer-specific BAA terms and applicable regulations (HIPAA, GDPR where applicable).

4. Third-Party Services

We use cloud infrastructure providers (subject to our security requirements) for hosting and processing. Distribution partners (SaveLife.AI, CARPL.ai, VSee) may process data under their own privacy policies when accessed through their platforms.

5. Your Rights

You may request access to, correction of, or deletion of your personal data by contacting purpleai@purple-ai.co. We respond to all data requests within 30 days.

Security Policy

Effective Date: [To be determined]

1. Certifications & Compliance

PurpleAI maintains the following active certifications:

  • ISO 27001 — Information Security Management System
  • ISO 13485 — Medical Device Quality Management System
  • GMP — Good Manufacturing Practice
  • FDA 510(k) — US Federal regulatory clearance (K240353)
  • Korea MFDS — Korean regulatory approval

2. Data Encryption

All data is encrypted using AES-256 at rest and TLS 1.3 in transit. DICOM images are processed within isolated, encrypted compute environments with no persistent storage beyond the analysis session unless explicitly configured by the customer.

3. Cloud Infrastructure

PurpleAI uses cloud-native deployment with no on-premise hardware required. Infrastructure is hosted in SOC 2-compliant data centers with 99.9% uptime SLA. Multi-region deployment available for enterprise customers.

4. Access Control

Role-based access control (RBAC), multi-factor authentication (MFA), and audit logging are standard across all deployments. Integration with hospital SSO/LDAP systems is supported.

5. Vulnerability Management

We conduct regular penetration testing, dependency scanning, and security audits. Responsible disclosure inquiries can be directed to purpleai@purple-ai.co.

Ready to Transform Your Stroke Workflow?

Join 40+ hospitals already using PurpleAI. Flexible deployment for institutions of all sizes.

Or email us: purpleai@purple-ai.co