Accurate fracture detection is vital as misdiagnoses can lead to prolonged treatment, costly complications, and poorer long-term outcomes.
To address this, Synapxe and the National Healthcare Group (NHG) will deploy a bone trauma interpretation artificial intelligence (AI) model at Woodlands Health's emergency department. The AI model will help doctors reduce missed fractures, shorten turnaround times for patient disposition, and lower patient recall rates.
Beyond that, the AI model can spot effusion, or excess fluid in a joint, and lipohemarthrosis, which is a mix of fat and blood in a joint. Both findings can signal other injuries, such as torn ligaments.
AI is also being adopted in other areas of Singapore’s healthcare system. The National Tuberculosis Screening Centre will be leveraging a chest X-ray AI model later this year to improve the accuracy and speed of tuberculosis (TB) diagnosis.
Developed by Lunit, the AI model analyses digital chest radiographs to highlight abnormalities (including TB) through heatmaps and abnormality scores. This capability helps clinicians prioritise urgent cases, reduce turnaround times, lower recall rates, and cut down on unnecessary evaluations.
The chest X-ray AI tool will be rolled out progressively across Singapore’s public healthcare system and is expected to be a national capability by the end of 2026.
See also: Foundation Healthcare eyes SGX listing in regional push
“At Synapxe, we see AI as a bridge from data to decision, from complexity to clarity, and from systems to people. Most importantly, we are laser-focused on what matters most, which is the human experience in healthcare,” says Ngiam Siew Ying, CEO of Synapxe, at the organisation’s AI Accelerate Conference 2025.
Both AI implementations are on AimSG, the AI medical imaging platform for public healthcare. They build on the successful deployment of AI for chest X-rays at Changi General Hospital and NHG Geylang Polyclinic last year, which helped clinicians prioritise cases more quickly.
Setting the foundation
See also: Seveno Capital backs human healthspan with US$70 mil fund
To support the rapid deployment of AI models to improve patient care, Synapxe and the Ministry of Health launched a national, cloud-based analytics platform called HEALIX (or Health Empowerment through Advanced Learning and Intelligence eXchange) in May 2024.
“HEALIX enables secure sharing of up-to-date and anonymised clinical and socio-economic data, to build and train AI models for healthcare. We can think of HEALIX as the engine or factory to drive AI applications. The infrastructural platform is ready and public healthcare institutions have been successfully onboarded,” says Minister for Health Ong Ye Kung at the same event.
To further strengthen HEALIX, Synapxe has signed separate Memoranda of Understanding (MOUs) with Databricks and Google Cloud.
Synapxe is currently using Databricks’ Data Intelligence Platform to support analytical insights, enhanced security, and data democratisation. The new MOU aims to accelerate AI adoption and workforce transformation across Singapore’s public healthcare sector. Initiatives include encouraging experimentation and implementation of AI use cases, co-developing predictive care solutions, and integrating Databricks’ AI capabilities into HEALIX.
Databricks will also support the HEALIX Data and AI Academy over the next three years, helping public healthcare professionals earn Databricks certifications.
Under the agreement with Google Cloud, Synapxe will train and certify more than 300 healthcare technology professionals, including data engineers, data scientists, and data analysts, from across the public healthcare system.
Synapxe will also tap into Google Cloud’s interoperable data and AI services, such as BigQuery, Vertex AI, and Agent2Agent Protocol, to enhance both HEALIX and Tandem, its generative AI platform.
To stay ahead of the latest tech trends, click here for DigitalEdge Section
Personalised healthcare
To enhance access to public healthcare services, Singapore will unify HealthHub, Health Buddy, NHG Cares, and the NUHS App into a single platform by 2027. The move is expected to enable more personalised care, with AI playing a central role.
As part of this effort, Synapxe will launch an AI-powered conversational assistant on the HealthHub website later this year. Known as HealthHub AI, the chatbot will provide personalised recommendations based on users’ health profiles. It will support both speech and text input, operate in four languages, and interact in a natural, conversational style.
Synapxe has also signed a Master Collaboration Agreement with OpenAI to explore how its multi-agent framework can streamline routine healthcare tasks. Using OpenAI’s Agents software development kit, a prototype has been developed to support transactional services such as appointment bookings and responses to general healthcare inquiries. All chat interactions remain private and are not used to train OpenAI models.
“If the pilot proves successful, there’s a lot more we can explore together. Imagine chronic care management, timely follow-up reminders, or AI-generated summaries of a doctor’s guidance. These kinds of tools can help make health information more understandable and actionable,” says Oliver Jay, managing director, International at OpenAI.
During a fireside chat at the event, he offered practical advice to healthcare institutions looking to adopt AI. He urged healthcare providers to first understand the current capabilities of the technology, then focus on clearly defined problems.
“If you’re a hospital or medical provider, think about where the bottlenecks are—it’s different for everyone. For some, it’s long wait times; for others, error rates or delays in documentation. Start small [with AI]. I’d challenge every single provider—there’s at least one area AI can solve very accurately today. Unlock that one area, learn what it takes, and then scale up.”
He also emphasised the importance of deploying AI safely. “There’s a tremendous opportunity and prior guides on how to make AI safe—how to deploy it in a way that includes guardrails and safeguards. If you understand how it works, you’ll feel more comfortable.”
Looking ahead, he believes AI will shift from being reactive to proactively supporting patients and providers. :AI agents that can handle multi-step tasks, understand accents and dialects, integrate into existing healthcare systems, and provide personalised insights. By simplifying parts of the healthcare journey—like making it easier to book a jab or understand a diagnosis—we believe it will build confidence and momentum to solve for more.”
