Biological ageing is a complex process, influenced by genetic and environmental
factors. Factors influencing the ageing of the mouth are in part genetic, but are mainly related to lifestyles habits, and hence often depend on socio-economic status. Numerous diseases of the oral cavity share common risk factors with other non-communicable diseases like diabetes or cancer. There is a bidirectional relationship between oral health and biological ageing. On one hand, age-related diseases, like dementia, cancer or geriatric depression, are frequent in elderly persons, and might have a direct impact on oral health. For example, oral health status may be jeopardised by frailty, disability, care dependency or limited access to professional oral health care. Another example is the side effect of several medications in causing xerostomia and hyposalivation. On the other hand, oral health problems such as periodontitis, unstable removable dentures, insufficient chewing or dysphagia are related to low Oral Health Reported Quality of Life scores, and nutritional state in geriatric individuals.

Deep convolutional neural networks (CNN) have been tackling, with impressive results, most of the pattern recognition challenges of the past few decades. With the advent of efficient digitization techniques for microscopy slides (Whole slide images, WSI), the use of deep learning models on histopathology images has been widely explored as it may save time and reduce errors in the diagnosis, prognosis or response to therapy predictions.

The objective is to study the relationship between nutrition and intrinsic capacities and understand the impact of Reactive Oxygen Species (ROS) metabolism in this context. For this, different approaches will be carried out within the framework of the INSPIRE platform to promote healthy aging. First, the epidemiological component will be made from data from human cohorts, aged 70 and over, randomized with multi-domain interventions in order to analyze the link between the nutrition and intrinsic abilities. Existing cohorts (MAPT, NOLAN, COGFRAIL) will be analyzed as well as the ongoing cohort of the INSPIRE project. We developed a mathematical model centred on the mitochondrial dysfunctions and more precisely on the dysfunction of mitochondrial respiratory chain (MRC) as it was shown to be a major point in ROS production (80% of total ROS production).