Insilico Medicine uses AI to predict biological cell age

Charlotte Edwards 17 January 2018 (Last Updated January 17th, 2018 17:06)

Insilico Medicine has created a new artificial intelligence (AI) algorithm that can accurately determine the biological age of a person’s cells. It could also hold potential to reveal whether lifestyle changes or certain medications could increase their chances of living longer.

Insilico Medicine uses AI to predict biological cell age
A 95-year-old great grandmother leading an active lifestyle highlights chronological age is just a number. Credit: Amy Cicconi

Insilico Medicine has created a new artificial intelligence (AI) algorithm that can accurately determine the biological age of a person’s cells. It could also hold potential to reveal whether lifestyle changes or certain medications could increase their chances of living longer.

The Baltimore, US-based AI firm, which specialises in next-generation medical technology techniques, claims its latest research is key to assessing the future of anti-ageing therapies.

The formula, known as Aging.AI, works by applying a simple algorithm to patient blood sample results to determine their biological age. To date, Insilico has provided accurate results for 130,000 people using the algorithm.

Each blood sample is inspected by a medical doctor for any abnormalities that could add years to a person’s biological age, such as cholesterol level issues or signs of inflammation. Healthier blood samples suggest a younger biological age, which Insilico considers to be indicative of a better chance to live a longer, healthier life.

Insilico recently published the results of the study in the Journal of Gerontology: Biological Sciences with researchers giving a detailed presentation of their deep learning-based haematological ageing clock model. The model was tested on a range of nationalities to achieve the most accurate results.

“Given ethnic differences in health, diet, lifestyle, behaviour, environmental exposures and even average rate of biological aging, it stands to reason that ageing clocks trained on datasets obtained from specific ethnic populations are more likely to account for these potential confounding factors, resulting in an enhanced capacity to predict chronological age and quantify biological age,” the researchers said.

As increased age makes it more likely for people to be affected by the most widespread health problems, such as cancer, dementia and cardiovascular diseases, the results of Insilico’s algorithm method could have an important impact on the lives of those likely to suffer from these illnesses.

The researchers have proposed that the formula could even assess the effectiveness of medical treatments on a user’s health by measuring their biological age before and after they start the medication. If so, success rates of medications in the future could be predicted based on how likely they are to reduce a patient’s biological age and by how much.

Insilico will discuss its work further at the 5th annual Advanced Pharma Analytics Europe Summit at the end of this month, with a particular focus on how the company can use AI for drug discovery.