Radiology Could Play a Bigger Role in the Early Diagnosis of Alzheimer's Disease

MRI data can help identify disease progression for personalized treatment of Alzheimer's

According to the Alzheimer’s Association, amnestic mild cognitive impairment (aMCI), the subtype of MCI that primarily affects memory, may cause a person to begin forgetting important information they would previously have easily recalled, such as appointments, conversations, or recent events. Not all aMCI patients will progress to dementia, and those that do have heterogeneous clinical progression, making it difficult to identify individual patients at highest risk.

“Understanding the individual deviations from the typical brain-aging trajectory in aMCI is important for the early identification of and intervention for patients at high risk of developing Alzheimer’s disease (AD),” said Weijie Huang, PhD, a researcher at the State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University.

“Research typically categorizes the clinical progression of Alzheimer’s disease into three stages—normal, MCI and dementia,” Dr. Huang explained. “However, the reality is that disease progression is continuous and does not suddenly jump from one stage to the next.”

To better quantify the severity of aMCI, Dr. Huang conducted a study using the brain age model, the results of which were published in Radiology: Artificial Intelligence.

Using Brain Age Model to Calculate Predicted Age Difference

The brain age model is a methodological framework for quantifying deviations from healthy brain aging. It does this by converting a brain scan into a set of simplified data that can be used to estimate the age of a patient’s brain.

The difference between the patient’s predicted age based on MRI data and the chronological age is called the predicted age difference (PAD). In general, the bigger the PAD, the higher the risk for developing AD—making PAD an important marker for early diagnosis of cognitive impairment and monitoring how a patient responds to treatment. 

To train a reliable model to calculate PAD, the study combined data from the Beijing Aging Brain Rejuvenation Initiative with that of the Alzheimer’s Disease Neuroimaging Initiative. The dataset included 974 healthy controls. After the model was trained, it was applied to a dataset consisting of 206 healthy controls and 249 aMCI patients.

“We found that patients with MCI deviate from healthy brain aging and that individual deviation relates to the patient’s clinical progression,” Dr. Huang said.

Furthermore, the study demonstrated that T1-weighted MRI can be used to calculate predicted age difference, and that PAD is significantly associated with individual cognitive impairment.

When considering different AD-related risk factors, those patients carrying APOE4, the strongest risk factor gene for AD, have higher PADs than noncarriers, while patients with amyloid-positive aMCI have higher PADs than patients with amyloid-negative status, further validating PAD as a predictor of AD risk. 

Use of Neuroimaging in Cognitive Neurology

Not only do these findings have the potential to enhance radiology’s role in the early diagnosis of AD, particularly in the prediction of conversion from aMCI to AD, they could also pave the way toward earlier and more accurate detection of cognitive impairment. For example, patients may benefit from improved diagnostic precision, risk assessment, and potentially tailored treatments based on their individual cognitive profiles, genetics, and biomarker status.   

The study’s findings may further contribute to the development of predictive tools for identifying those at higher risk of disease progressions, ultimately enhancing patient outcomes and quality of life.

“Our results could open the door to more precise and personalized approaches to patient care and further the development of neuroimaging as a valuable tool in the field of cognitive neurology,” Dr. Huang said. 

Researchers are now further refining and validating its age prediction model via a large-sample, multi-site dataset. This enhanced model will then be applied to longitudinal data as a means of validating whether PAD can assess treatment response.

“The ultimate goal is to translate these findings into routine clinical practice, enabling earlier detection, personalized care and improved outcomes for patients experiencing cognitive impairment,” Dr.  Huang concluded.

For More Information

Access the Radiology: Artificial Intelligence study, “Accelerated Brain Aging in Amnestic Mild Cognitive Impairment: Relationships with Individual Cognitive Decline, Risk Factors for Alzheimer Disease, and Clinical Progression.

Read previous RSNA News articles on brain aging and Alzheimer’s Disease: