L’IA peut-elle prédire l’âge métabolomique d’une personne et adapter les plans de santé en conséquence ?

Artificial intelligence continues to push the boundaries of science and health exploration. From mapping proteins to neurological networks, these systems continue to make game-changing impacts on the sector. This month, a group of advantageous researchers from the Institute of Psychiatry, Psychology & Neuroscience took AI integration further. They released a étude¹ in which AI algorithms were used to predict the metabolic age of patients. Here’s what you need to know.

Horloge Chronologique vs Horloge Biologique

To grasp the importance of this development, it’s vital to understand the differences between your chronological and biological age. These two methods of determining the age of your body are used in different ways by healthcare professionals. As such, it’s crucial to be able to determine both accurately when providing health care strategies.

Horloges Chronologiques

Your chronological age is your age based on the days of your life. Your chronological age can help healthcare professionals understand what to reasonably expect from your body in terms of health status, signs of aging, lifestyles, and more. It can also help professionals develop drugs tailored for certain age groups,  enabling more effective care. Notably, your chronological age cannot be changed.

Horloges Biologiques

Your biological age represents the stage of your molecular and cellular damage. Unlike chronological aging, biological aging can be altered by changing lifestyles, environments, diets, and more. Your biological aging will be detectable in the level of performance and capabilities based on your chronological age expectations. The main goal of biological clocks is to track and demonstrate how your metabolism changes over time.

Étude des Horloges d’Âge Métabolomique Propulsées par l’IA

AI-powered biological aging clocks could revolutionize the sector. These algorithms can connect huge swatches of data to create in-depth maps that integrate advanced metabolite data. Notably, this study was the first to attempt to use AI to map and rate metabolites.

Métabolites

Metabolites are markers found in the blood that appear after metabolism. They serve multiple roles and can indicate certain aspects of health. These molecules are a relatively unexplored method of determining a patient’s health status and expectations.

In this approach, engineers introduce statistical or machine-learning algorithms. These systems are specifically designed to locate relationships between chronological age and molecular data. The results are a more in-depth and individualized understanding of a person’s health age.

Âge Métabolomique – MileAge

As part of the study, the engineers created a Metabolic Age ranking system. Each patient was given a MileAge based on the difference between metabolite-predicted and chronological age. The MileAge approach leverages a number of metabolic markers. Specifically, the engineers discovered that a select panel of metabolites explained more than half of the variance in chronological age.

Additionally, researchers found that those with higher MileAge were far more susceptible to risk and health factors. Their bodies were also weaker and couldn’t heal as fast. Notably, their cells showed shorter telomeres, which is another sign of decaying health and aging. The difference between metabolite-predicted age and chronological age is called the MileAge Delta.

Source – Science.org

Test de l’Étude d’Âge Métabolomique

To discover the optimal MileAge algorithms, engineers had to test many different machine learning algorithms. Notably, the team leveraged nuclear magnetic resonance spectroscopy to monitor cellular changes. They began by developing 17 machine learning algorithms based on data gathered from blood from over 225,000 UK Biobank participants.

Notably, the age of participants ranged from 40 – 69 years. The algorithms were programmed utilizing 168 plasma metabolites derived from UK Biobank data. It included middle-aged and older adults.

The next step was to compare the AI algorithms to see which ones were the most accurate. The team used nuclear magnetic resonance (NMR) spectroscopy to acquire data from the blood banks to benchmark each model. The comparison yielded some interesting results.

Résultats de l’Étude d’Âge Métabolomique

The study demonstrated that AI algorithms are not all equal in terms of their ability to accurately predict MileAge. The engineers noticed that Cubist rule-based regression and linear algorithms worked the most effectively at determining aging signals.

The study revealed that 116 metabolites serve roles in this function, with GlycA, omega-3, and DHA having the highest correlation with age. These revelations demonstrated that more complex nonlinear AI systems were the best options.

Avantages de l’Étude d’Âge Métabolomique

This metabolic age study brings several key benefits to the table. For one, it will allow people to track their health by empowering proactive behavior. It will also be a key tool used by professionals to determine the best care practices and strategies to implement based on individual patient needs.

Utiliser l’Âge Métabolique comme Indicateur d’Alerte Précoce

Checking your MileAge could provide you with added confidence in your health. This system makes it easier to see early signs of declining health. As such, healthcare professionals can recommend more effective measures to combat disease. It will also play a major role in improving preventive measures by making it easy to see who is at high risk.

Cas d’Utilisation de l’Âge Métabolique

There are several immediate and distant use case scenarios for MileAge-based technologies. These options range from enabling people to monitor their health better, all the way to creating advanced care and treatment methods that integrate key metabolic markers.

Évaluations de Santé

At the core of this technology will include the ability to create more useful and specific health assessments. Insurance companies and healthcare professionals will need to leverage this tech to improve and monitor their healthcare strategies currently. In the future, you may need to get a MileAge checkup prior to gaining coverage.

Stratification des Risques

Another key science that can benefit from MileAge integration is risk stratification. Research will leverage this data to find out what areas and people are at the highest risk of certain disorders and diseases. Determining the risk level of populations will help researchers locate and eliminate unnecessary risks presented to patients based on their location or habits.

Suivi Proactif de la Santé

The rise of wearables has introduced a new level of health trackability. Now, the metabolic age scale could help to improve personal health tracking even further. Imagine taking a test or even just clicking a button on your watch to find out the key areas you need to improve your health and bring you up to par with others in your area and age group.

Chercheurs de l’Âge Métabolique

The MileAge research was conducted by engineers from the Institute of Psychiatry, Psychology & Neuroscience. The study received funding from various backers, including the National Institute for Health and Care Research (NIHR) and Maudsley Biomedical Research Centre (BRC). Additionally, the UK Biobank played a vital role in providing access to data and samples.

Entreprises qui Peuvent Bénéficier de l’Étude d’Âge Métabolomique

There are several companies that could integrate this technology to improve their offerings and services. These firms work in the healthcare industry and already provide some form of service that utilizes a person’s health status as an indicator or metric when determining the products or offerings.

10x Genomics Inc

10x Genomics Inc (TXG ) est une entreprise de recherche scientifique avancée spécialisée dans la technologie de séquençage génétique. La société est entrée sur le marché en 2012 sous le nom d Avante Biosystems, Inc. avant de changer de nom pour 10x Genomics. Le projet a été fondé par Serge Saxonov, Ben Hindson et Kevin Ness afin d’améliorer la compréhension de la biologie pour les experts en santé. Aujourd’hui, l’entreprise propose divers services couvrant l’immunologie et les neurosciences.

En 2018, 10x Genomics a réalisé plusieurs acquisitions majeures qui ont amélioré sa technologie et son positionnement sur le marché. Par exemple, la société a acquis Epinomics et Spatial Transcriptomics. Ces deux manœuvres ont enrichi son offre et ont donné à l’entreprise accès à des technologies avancées.

Notably, TXG is a popular stock that continues to draw investor attention alongside the company’s accomplishments. To date, the firm received multiple awards including The Scientist Top 10 Innovations award in 2017, 2018, 2019, 2020, and 2021. Additionally, it’s considered one of the top 100 global research institutions and consistently ranks among the top 20 global pharmaceutical research firms.

Avenir de l’Étude de l’Âge Métabolique

Now that engineers have determined that the MileAge marker provides more accurate results compared to previous methods, there’s sure to be additional peer review prior to integration. This method of tracking health will quickly catch on as more agencies realize the added accuracy and savings it provides.

Âge Métabolique – Suivre votre Santé à de Nouveaux Niveaux.

The average person gains a lot from this study in regard to their future healthcare. For one, insurance companies, healthcare providers, and drug manufacturers will all be able to leverage their information to more accurately provide services to the population. As such, there is a strong demand for this data to receive priority in terms of integration. Regardless of the time frame it takes to institute, MileAge markers will soon become crucial in determining health care plans and more.

Découvrez d’autres technologies de santé intéressantes maintenant.

Référence de l’Étude :

^{1. Mutz, J., Iniesta, R., & Lewis, C. M. (2024). L’âge métabolomique (MileAge) prédit la santé et l’espérance de vie: une comparaison de plusieurs algorithmes d’apprentissage automatique. Science Advances, 10(51), eadp3743. https://doi.org/10.1126/sciadv.adp3743}