Expanded collaboration explores how AI could help identify the right treatment for the right Alzheimer’s patient.
What if Alzheimer’s treatment became personal? Say two people, both in their early 70s, sitting in the same memory clinic. Both have been diagnosed with Alzheimer’s disease and are beginning the same treatment. A year later, their journeys look very different – one’s symptoms have stabilized, and the other has continued to decline. Why?
The why is one of the biggest challenges in Alzheimer’s medicine today. While new therapies are bringing fresh hope to patients and families, doctors still have limited ways of predicting who is most likely to benefit from a particular treatment.
A new phase of collaboration between South Korean biotech NeuroXT and Beth Israel Deaconess Medical Center (BIDMC) in Boston hopes to help change that. The organizations have announced an expanded research agreement focused on using AI and medical imaging to better understand how individual patients respond to Alzheimer’s therapies [1].
The partnership builds on an earlier collaboration launched in 2024, which focused on validating NeuroXT’s MRI-based AI technology.
Think of an MRI scan as a photograph. To a trained clinician, it already contains valuable information. NeuroXT’s AI is designed to act like an additional set of highly specialized eyes, looking for subtle patterns that may be difficult for humans to detect on their own.
The goal is to estimate Alzheimer’s-related biomarkers typically measured through PET scans, which can provide important insights into what’s happening inside the brain. PET imaging can be highly informative, but it is also expensive and less accessible than MRI. If AI can reliably extract similar insights from scans that are already widely available, it could make advanced Alzheimer’s assessment more practical and scalable.
The next phase of the collaboration takes that concept a step further. Researchers will analyze data from patients who have received Alzheimer’s therapies at BIDMC for more than a year, exploring whether NeuroXT’s imaging biomarkers can help reveal patterns in treatment response.
Rather than simply asking, “Does this patient have Alzheimer’s?”, the research is moving toward a more nuanced question: “Which treatment is most likely to help this particular patient?”
Timing is notable. For years, Alzheimer’s research was largely focused on developing therapies capable of slowing the disease. While that effort continues, the conversation is beginning to evolve.
As treatments enter clinical practice, attention is shifting toward optimization. How can physicians determine who should receive treatment? How early should intervention begin? Which patients are most likely to benefit? Those questions mirror changes already underway in fields such as cancer care, where personalized treatment strategies have become increasingly common. Alzheimer’s medicine may now be moving in a similar direction.
“Our continued collaboration with BIDMC marks an important step toward translating AI innovation into clinical impact,” said Joon-Kyung Seong, CEO of NeuroXT.
The collaboration continues under the leadership of Dr Daniel Press, Chief of the Cognitive Neurology Unit at BIDMC, and Dr Chun Lim, Medical Director of the Cognitive Neurology Unit, bringing together clinical expertise and advanced imaging research.
Of course, the significance of this work extends beyond Alzheimer’s disease itself. One of the central goals of longevity medicine is not simply helping people live longer, but helping them maintain cognitive function, independence and quality of life as they age. That makes Alzheimer’s one of the defining challenges of the longevity era.
The longevity field has spent years searching for better drugs, better diagnostics and earlier interventions. However, another question is emerging: how do we make those tools work better for each individual? That is where AI may ultimately have its greatest impact – not as a replacement for physicians, but as a tool that helps uncover patterns hidden within vast amounts of clinical data, patterns that could inform more personalized decisions and potentially improve outcomes.
