UAE’s healthcare revolution: AI, robots, data redefine medicine
In procedures like endoscopy, where a camera is inserted into the body, AI modules help detect cancers or anomalies in real time
08 September, 2025
TT
16
Artificial intelligence is no longer a futuristic vision in healthcare, it’s happening now, and it’s making surgeries safer, faster, and more precise.
This is according to top experts speaking at the WHX Tech event, which is taking place from September 8-10 in Dubai.
Sandip Kumar, chief digital and innovation officer at King’s College Hospital London, UAE, told Gulf Business how AI is already being deployed in the region’s operating rooms. “In procedures like endoscopy, where a camera is inserted into the body, AI modules help detect cancers or anomalies in real time,” he said.
Read more-How WHX Tech’s Connections Programme is revolutionising healthcare investment
AI systems can scan a 15-minute endoscopic video and highlight potential areas of concern within seconds, enabling surgeons to act quickly and focus their attention where it’s most needed.
Beyond endoscopy, AI also plays a background but crucial role in robotic-assisted surgeries. These systems continuously monitor procedures in real time, detecting anomalies and ensuring quality standards are met. “It acts like a second set of eyes, supporting the surgical team,” Kumar explained.
In Australia, AI is now a standard component in stroke care. Automated systems can swiftly identify brain hemorrhages, pinpoint their location, and even assist robotic platforms in navigating to the affected site with surgical precision. “Every second counts in stroke cases. We used AI extensively for rapid intervention during my time at a stroke center in Australia,” Kumar added.
Simple tools, big impact
AI’s application extends beyond complex interventions. It’s also solving basic but critical challenges in the operating theater.
“Globally, there’s a consistent issue, surgeries often begin without all the required specialised tools,” Kumar noted. General theater kits, rather than procedure-specific ones, are commonly used. The result? Instruments go missing mid-surgery, delaying critical operations.
AI is addressing this with computer vision systems that validate instruments before procedures begin. These technologies ensure sterility, verify tools, and match surgical kits to specific procedures. “This might seem basic, but it’s essential. Getting this right means fewer delays and better patient outcomes,” Kumar emphasised.
The UAE’s digital infrastructure: Ready for AI surge
The UAE has already built a robust foundation for healthcare innovation. From the Abu Dhabi Health Information Exchange (Malaffi) to mandates for unified electronic medical records, the digital groundwork is solid.
According to Kumar, it’s now time for the country to move from infrastructure building to application. “The next phase is execution. We need to expose this data meaningfully to create innovations that are truly patient-centric.”
At King’s College Hospital in Dubai, that shift is already underway. “Since I joined, we’ve launched over 15 AI initiatives designed to directly impact patient outcomes,” Kumar said.
One standout use case is in breast radiology. Patients undergoing mammograms at King’s are assessed by both a seasoned radiologist and an AI system called Lunit AI. The tool serves as a diagnostic co-pilot, scanning images to detect subtle patterns that might escape the human eye.
“We’re also deploying similar technologies in pathology,” Kumar added. “We analyse blood tests and medical histories using AI to detect patterns that could predict disease, even before symptoms appear.”
From lab results to layman’s terms: AI in patient communication
Another area showing promise is ambient listening technology, designed to summarise doctor-patient consultations in real time. These AI systems listen passively during appointments and generate summaries, both for the clinician and the patient.
“Often, patients forget half of what their doctor tells them before they even reach the parking lot,” Kumar said. “This technology bridges that gap.”
Importantly, the system translates medical jargon into plain language. A patient might receive a recap like: “You saw Dr A, the consultation was about X. Your recommendations were A, B, and C. Here’s what that means for you.”
This transparency ensures clarity and helps patients follow through with care plans more effectively.
Bias in AI: A global problem with local solutions
Despite its promise, AI in healthcare carries a significant risk: bias.
“AI can be the great equaliser, but only if it’s developed and applied equitably,” said Dr Harvey Castro, MD, an advisor on AI and healthcare to the Singapore Government. He made the statement during a session at the WHX Tech event.
“If you train a model on one population and use it on another, you’re asking for misdiagnosis,” he warned.
To counter this, Dr Castro strongly advocates for region-specific AI training. “Models should be trained by ZIP code, by community, by region. Different hospitals serve different demographics. The one-size-fits-all model simply doesn’t work in healthcare.”
This need is particularly urgent in multicultural nations like the UAE, where diverse populations access care. Without localised AI training, outcomes risk becoming skewed and potentially dangerous.
Robots and empathy: Augmenting, not replacing, humans
Dr Castro also highlighted the synergy between robotics and human empathy as critical to the future of medicine. “We’re not replacing doctors with robots, we’re augmenting them,” he explained.
In practice, this could look like smart glasses feeding subtle cues to a physician during a patient interaction: “Check the patient’s left ear,” or “Don’t forget to ask about family history.” These prompts, driven by real-time AI analysis, enhance human judgment.
Repetitive tasks, charting, measuring, logistics, can be offloaded to machines. “This frees clinicians to focus on what really matters: presence, empathy, and nuanced decision-making.”
Some areas of healthcare are already seeing fully autonomous AI applications. In Singapore, for instance, medication kiosks refill prescriptions post-teleconsultation, without any human intervention.
“For stable, low-risk cases like routine diabetes meds, this kind of automation is safe and highly efficient,” Dr Castro noted.
However, both Castro and Kumar stress that human oversight remains essential for complex, emotional, or high-stakes decisions. “We still need doctors. We just need them with better tools,” said Castro.
The need for regulation: A ‘nutrition label’ for AI
As AI becomes embedded in clinical workflows, governance is becoming the most critical issue. Today, even general-purpose tools like ChatGPT are being used in medical settings, despite lacking certification for such use.
“That’s dangerous,” Dr Castro warned. “We need transparency, like a nutrition label for AI.”
Just as consumers review ingredients and calorie counts, healthcare providers need to understand what an AI model was trained on, where it’s accurate, and where it could be risky.
“If it was trained on 40-year-old white males, we shouldn’t expect it to work flawlessly on 20-year-old women or a South Asian population,” he said. “Countries must invest in training their own models with their own data. It’s expensive, yes. But it’s the only way to ensure safety and fairness.”
A future that’s human and AI, not human vs AI
Artificial intelligence is poised to become healthcare’s most powerful tool, but only if it’s deployed ethically, regulated responsibly, and integrated humanely.
From surgical suites in Dubai to telemedicine kiosks in Singapore, the message is clear: AI isn’t about replacing doctors. It’s about giving them superpowers.
As Sandip Kumar put it, “Now is the time to move from building infrastructure to delivering true, patient-focused innovation. The tools are here. Let’s use them the right way.”
