Inside Philips’ AI Strategy To Deliver Better Care At Scale

Shez Partovi is Chief Innovation and Strategy Officer and Chief Business Leader of Enterprise Informatics at Royal Philips, a global health technology company headquartered in the Netherlands. Philips is focused on improving health and well-being through meaningful innovation that combines advanced technology with deep clinical and consumer insight. In 2024, the company generated €18 billion in revenue and employs more than 67,000 people, serving customers in over 100 countries.

An experienced clinical professor, neuroradiologist, global executive and entrepreneur, Partovi brings a unique perspective shaped by decades of leadership in health systems, cloud transformation and artificial intelligence. Under his leadership, Philips is advancing its role as a trusted partner in integrated diagnostics, scaling analytics and driving workflow optimization across the healthcare continuum. Operating at the intersection of diagnostics, informatics and patient care, Partovi is helping to shape the company’s vision of delivering better care for more people, especially in an age marked by clinician shortages, cost pressures and exploding demand.

A Strategy Anchored in Customer Pain Points

Philips’ strategic approach is deceptively simple: invest in solving the customer’s problems, not in promoting the company’s ideas. “The best innovations are painkillers, not vitamins,” Partovi said, emphasizing a shift away from traditional R&D silos toward customer-driven design.

Each of Philips’ divisions, from precision diagnostics to sleep and respiratory care, applies this philosophy through formalized partnerships with health systems. These relationships often involve co-located teams of engineers, product managers and architects embedded directly in clinical environments. “We iterate at the bedside, in the radiology suite, or in the pathology lab,” Partovi explained. “That’s how we build algorithms, software and solutions that deliver meaningful outcomes.”

A standout example is the SmartSpeed AI software developed for Philips’ BlueSeal MRI scanners. Inspired by the needs of Leiden University Medical Center in the Netherlands, which faced a patient backlog, Philips built a neural network that tripled scan speeds and improved image resolution without requiring new hardware. “It’s a software-driven leap in productivity,” Partovi said. “And it came from listening to one customer’s pain point.”

Organizing Innovation for Scale

Innovation at Philips isn’t left to chance. Partovi oversees a tightly structured system that balances decentralization with long-horizon research. Roughly 80 percent of Philips’ R&D resources are embedded within its core business units, ensuring close proximity to customers and rapid feedback cycles. The remaining 20 percent reside in central innovation teams focused on industry-shifting breakthroughs.

A case in point is BlueSeal itself, which uses only seven liters of helium, far less than the industry-standard 1,500. “No one asked for helium-free MRI,” Partovi said. “But we saw sustainability and cost benefits years ago, and now the entire industry is moving in that direction.”

Whether close to market or looking around corners, all Philips innovation shares three traits: it must be

• People and patient centric
• Co-created with partners
• Designed for scale.

“We want to improve the lives of 2.5 billion people annually by 2030, including 400 million from underserved communities,” Partovi underscored. “That only happens if you build with global impact in mind.”

Reimagining Healthcare Software with Enterprise Informatics

Two years ago, Partovi expanded his responsibilities to include Philips’ Healthcare Informatics division. The move was strategic: unlike hardware, which has long product cycles, software must evolve continuously. “If an MRI scanner updates every two years, software should update every six weeks,” Partovi noted.

Philips consolidated its imaging and clinical informatics into a standalone unit to accelerate innovation. The division spans everything from diagnostic viewers for radiologists and pathologists to medical device integration that feeds predictive algorithms. “Inside a hospital, anything with a digital readout [including] ventilators, IV pumps, monitors, can ingest data and apply AI to it,” Partovi said.

These capabilities are already powering predictive solutions. At Massachusetts General Hospital in Boston, Philips is working to forecast respiratory distress in ICU patients, helping clinicians intervene earlier and improve outcomes.

Three Dimensions of AI: Automate, Augment, Agility

Partovi describes Philips’ AI strategy in terms of three functions: automation, augmentation and agility. Each serves to amplify clinical effectiveness and efficiency.

• Automation: In cardiac ultrasound, AI can now identify key frames and calculate heart strain in seconds, tasks that once took 20 minutes of manual effort. “Sonographers didn’t train to do Photoshop,” Partovi quipped. “They trained to care for patients.”
• Augmentation: In Philips’ Azurion image-guided therapy suite, AI fuses real-time imaging to guide catheters through the heart with precision, effectively becoming a co-pilot for interventional cardiologists.
• Agility: Through medical device integration and continuous data analysis, AI enables always-on surveillance systems that can predict and prevent complications, improving patient safety and optimizing clinician response.

Partovi sees AI not as a replacement for caregivers but as a tool that expands capacity. “The most pressing challenge in healthcare is the gap between demand and supply,” he said. “AI helps us close that gap.”

Bridging the Trust Gap in AI

According to the latest Philips Future Health Index, a global survey of 16,000 patients and 2,000 clinicians, a notable gap persists between physician and patient trust in AI. While 79 percent of clinicians believe AI reduces burdens and improves care, only 59 percent of patients agree, and many worry AI might reduce time spent with doctors.

However, trust grows dramatically when humans remain in the loop. “Patients showed 86 percent confidence in AI when a physician was involved,” Partovi said. “The takeaway is clear: design AI to enhance, not replace, the human relationship.”

This insight has become central to Philips’ AI design philosophy, which stresses transparency, fairness and regulatory rigor. “Innovation must be responsible, or it’s not innovation at all,” Partovi emphasized.

Internal Use of Generative AI at Philips

While most attention goes to patient-facing applications, Philips is also leveraging AI internally across the enterprise. In software engineering, generative AI is now responsible for up to 30 percent of new code within some teams. “One developer told me, ‘I feel like I have a new job,’” Partovi recalled.

Customer service is also evolving. Large language models trained on Philips’ proprietary knowledge base now power advanced chat tools used by both customers and internal service teams. “In one team, 10 percent of service interactions are fully handled by agentic AI,” he said.

The result? Faster cycles of innovation and more effective knowledge sharing across a complex, global organization.

Looking Ahead: A New Age of Intelligence

For Partovi, the excitement around AI and generative technology hasn’t dimmed; it’s accelerating. He likens today’s moment to past tectonic shifts: the agricultural revolution, the industrial revolution, the rise of computing.

“We’re now entering the age of intelligence,” he said. “For the first time, you can embed reasoning and action into digital tools. This changes everything.”

From closing care gaps in underserved regions to transforming daily workflows in hospitals, Partovi sees AI as the engine that will finally deliver on healthcare’s promise of access, equity and personalization. “We’ve dreamed about using technology to close the healthcare gap,” he said. “Now we finally have the tools to do it.”

Peter High is President of Metis Strategy, a business and IT advisory firm. He has written three bestselling books, including his latest Getting to Nimble. He also moderates the Technovation podcast series and speaks at conferences around the world. Follow him on Twitter @PeterAHigh.