Graduate’s AI models reshape care for older patients

10 September 2025

Business School, Alumni, Doctoral, Graduation, Staff news, Business and economy, Health and medicine

In his home near Iran's capital Tehran, a young Abtin Maghsoodi received a gift that's left an impression to this day.

It was a Lego Mindstorms set, given to him by his dad when he was around nine years old. The set gave Abtin the chance to create programmable robots out of Lego bricks, and it's how he started tinkering, building, and teaching himself the basics of engineering and coding.

"My parents were amazing at nurturing curiosity; they used to get my brother and me Lego sets and DIY engineering kits, and they would sit with us, not just to help us build, but to really explore how things worked. Looking back, I think that curiosity and love for creating things never really left me."

Now, Dr Maghsoodi, who crossed the stage to receive his PhD from the University of Auckland Business School on 9 September 2025, is at the forefront of utilising artificial intelligence to solve complex problems in the healthcare industry, and further afield.

The inspiration for his thesis, which he started in 2021, was very personal. At the time, Abtin's late grandmother was living with Alzheimer's, and he saw how challenging it was for her to navigate the healthcare system.

"That experience stayed with me and made me want to work on something that could improve care for older patients."

Older people often face complex health issues such as multiple chronic conditions, which make their time in hospital particularly challenging, says Abtin, who created an intelligent tool called REACH (Recognising Episodes of Acute Complexity in Health).

REACH uses machine learning to help hospitals identify complex patients in real-time and prioritise their care. It's being considered for rollout across the Midlands region (Te Manawa Taki) of Health New Zealand.

Abtin's second programme, REACH 2.0, is also now being considered for wider implementation.

"This next version doesn't just classify patients into complex and non-complex; it incorporates early supported discharge, aged residential care placement, and even mortality risk into the prioritisation process," he says.

"Imagine a busy hospital ward where staff know in advance which older patients need more attention, and which can safely follow a standard pathway. That's what my models do: they classify patients into complex and non-complex groups and predict who might be at risk of coming back to the hospital soon after discharge. This means hospitals can prioritise care, reduce delays, and prevent unnecessary readmissions."

Seeing something that started as an idea now being considered for real-world deployment is humbling and motivating, says Abtin, who celebrated his graduation at the Auckland Town Hall with his wife Mojan, family and friends.

"Mojan and I came to New Zealand to pursue our PhDs, and we did it in the middle of the Covid pandemic, through lockdowns, quarantine, and long periods of isolation," he says.

"We shared the entire experience, the highs, the lows, the late nights, and the small victories. Doing a PhD can feel like a marathon, and having someone who understands what you're going through is invaluable."

Abtin's PhD supervisors also played a huge role in supporting him and shaping his research.

"I owe a special thanks to my amazing mentors and supervisors, Professor Valery Pavlov, Professor Paul Rouse, Professor Cameron Graham Walker, and Professor Matthew Parsons. Their support, insights, and encouragement were instrumental in turning an idea into a project that could make a real difference."

In his second year of PhD studies, Abtin stepped into the role of lead data scientist at Health New Zealand, Te Whatu Ora. He says balancing doctoral studies with a leadership role in the health sector was intense, but it gave him a unique perspective on how research can directly influence real-world outcomes.

Now, he's an associate director and lead data scientist at KPMG, where he works on state-of-the-art AI projects across a range of industries.

"The future of AI is moving beyond predictive analytics into autonomous, intelligent, and context-aware systems," he says.

"Imagine a network of AI agents managing hospital operations, supply chains, and customer service simultaneously, while learning and adapting in real time."

Of course, with this power comes responsibility, says Abtin, and the future of AI will require clear governance, ethical frameworks and transparency.

"My PhD taught me how to approach complex problems with a rigorous, evidence-based mindset and how to design solutions that are innovative and practical. This experience gives me a unique edge in my role, and having that research background helps me bring structure and scientific thinking to projects."

Looking back, Abtin says the combination of academic development, professional opportunities, and the people he worked with made his doctoral experience one of the most rewarding periods of his life.

"And yes, I'd do it all over again … maybe with a little less coffee!"

Media contact:

Sophie Boladeras, media adviser M: 022 4600 388 E: [email protected]