HistoAge Model Developed to Predict Histopathologic Brain Age

Researchers from the Icahn School of Medicine at Mount Sinai have developed an AI algorithm called HistoAge that predicts a person's age at death by analyzing the cellular composition of human brain tissue. The algorithm demonstrates an average accuracy of 5.45 years. Additionally, HistoAge can identify regions of the brain vulnerable to age-related changes, indicating potential cognitive diseases.

To develop HistoAge the team examined nearly 700 digitized images of human hippocampal sections from aged brain donors. The hippocampus is closely associated with both brain aging and age-dependent neurodegenerative diseases. Using these images, they trained a machine learning model to estimate a person's age at death based solely on the digital brain section. This task would be nearly impossible for a human observer to perform with accuracy. The difference between the predicted age by the model and the actual age is used to determine age acceleration in the brain.

Compared to current methods of measuring age acceleration, such as DNA methylation, HistoAge-based age acceleration has shown stronger associations with cognitive impairment, cerebrovascular disease, and the levels of abnormal degenerative protein aggregation seen in Alzheimer's disease.

Compared to current methods of measuring age acceleration, such as DNA methylation, HistoAge-based age acceleration has shown stronger associations with cognitive impairment, cerebrovascular disease, and the levels of abnormal degenerative protein aggregation seen in Alzheimer's disease.

According to the paper published in Acta Neuropathologica, HistoAge opens new horizons for assessing aging and neurodegeneration in human samples. The researchers believe it can be easily deployed on a large scale in clinical and translational research laboratories, providing more rigorous, unbiased, and robust metrics for cellular changes underlying degenerative diseases.

The team also plans to establish a multicenter collaboration to build a comprehensive AI-ready dataset. This dataset will be used to develop even more powerful AI models with the potential to transform and enhance our understanding of brain diseases.

Mount Sinai researchers see AI as a paradigm shift in brain research, revolutionizing the way we assess human diseases. They anticipate that AI will not replace compassionate care but will significantly improve diagnosis and treatment, propelling us toward the next generation of cures. The HistoAge model, along with similar algorithms, can uncover crucial causal aspects of debilitating brain diseases, such as Alzheimer's, by shedding light on the mechanisms of aging and neurodegeneration. Additionally, it can help identify genes that protect against brain aging or accelerate it and uncover environmental risk factors that impact brain aging.