Solutions co-developed with Ministry of Health Malaysia
Solutions such as AskCPG are positioned around national clinical workflows and Malaysia clinical practice guidelines, with MOH-linked co-development and governance context.
Resources - Research & White Papers
Clinical-AI research, regulatory trust and external validation.
A curated evidence library covering Qmed Asia recognition from Google Research and Intel, MOH-linked development, ISO certifications, Malaysia Digital status, NVIDIA ecosystem work and peer-reviewed technical research.
Evidence Library
Source-backed evidence for clinical AI credibility.
Qmed Asia's evidence base spans public-sector clinical collaboration, international standards, global technology partners and peer-reviewed technical research.
Recognized By
Google Research and Intel
Research Venues
IEEE Xplore, arXiv and eLife
Trust Signals
MOH, ISO, Malaysia Digital and NVIDIA
Regulatory trust
Recognized and trusted by healthcare regulators.
Qmed Asia combines public-sector clinical collaboration with the quality, security and digital-economy credentials expected in healthcare environments.
Certified ISO 13485 medical device quality management
Demonstrates disciplined design, development and quality management practices expected for medical device and clinical technology environments.
Certified ISO 27001 information security controls
Reinforces data security governance, access controls, audit readiness and privacy-minded deployment for healthcare environments.
Malaysia Digital accredited
Recognized under Malaysia Digital status, the successor framework to MSC status for digital-economy companies in Malaysia.
External recognition
Named in official Google and Intel sources.
Qmed Asia's clinical AI work has been referenced by Google Research and profiled by Intel for real-world healthcare deployment.
Google Research / Health AI Developer Foundations
GoogleHealth AI
Qmed Asia cited as a MedGemma 1.5 deployment example
Google Research explicitly names Qmed Asia's askCPG as an example of MedGemma adapted for real-world clinical guideline access. The article notes askCPG helps clinicians query Malaysia's 150+ clinical practice guidelines through a conversational interface.
Official Google Research article names Qmed Asia and askCPG
References Malaysia's 150+ clinical practice guidelines
Notes Ministry of Health Malaysia feedback on day-to-day clinical decision support
Generative AIClinical Guidelines
Open Google Research mentionIntel Customer Spotlight
intel.
Intel profiles Qmed Asia for privacy-preserving cancer diagnostics
Intel's customer spotlight profiles Qmed Asia's collaboration with JelloX on MetaLite, a federated learning solution for 3D medical imaging and cancer diagnostics. The deployment uses Intel Xeon processors, OpenVINO and SGX to support secure, multi-site AI workflows without moving patient data.
Intel customer spotlight dedicated to Qmed Asia and MetaLite
Highlights federated learning, Intel Xeon, OpenVINO and SGX
Reports workflow efficiency and diagnostic cost improvements from the deployment
Federated LearningData Privacy
Open Intel spotlight
AI ecosystem collaboration
Strengthening medical AI with NVIDIA ecosystem support.
Qmed Asia works with the NVIDIA ecosystem on model development, applied AI training and university talent programs that support Malaysia's clinical AI pipeline.
Co-develop AI models with NVIDIA engineering teams
Graduated from the NVIDIA Inception Program
Train future AI talent through joint university programs
Selected as NVIDIA's local healthtech partner for ML and AI training with local universities
Peer-reviewed validation
Technical rigor across clinical NLP and medical imaging.
Qmed Asia's engineering team contributes to efficient medical summarisation, fundus image quality assessment and explainability work for medical LLMs, with public publication records available through DOI, arXiv and journal pages.
IEEE
IEEE Xplore
Enhancing Medical Summarisation with Parameter Efficient Fine Tuning on Local CPUs
Peer-reviewed conference work on adapting medical summarisation models with parameter-efficient fine tuning for constrained local CPU environments.
DOI: 10.1109/ICECCE63537.2024.10823619
arXiv
arXiv
FundaQ-8: A Clinically-Inspired Scoring Framework for Automated Fundus Image Quality Assessment
Introduces an expert-validated eight-parameter framework for fundus image quality scoring, with a ResNet18 regression model and EyeQ validation.
arXiv:2506.20303
eLife
eLife
Critique of Impure Reason: Unveiling the reasoning behaviour of medical large language models
Peer-reviewed review defining reasoning behaviour for medical LLMs and surveying evaluation methods for explainability, safety and clinical trust.
DOI: 10.7554/eLife.106187