Artificial Intelligence At The Service Of Health

Artificial Intelligence At The Service Of Health : Medical research is going through a digital revolution linked to the abundance of data and our ability to collect, store and above all process it. These possibilities are linked to the capacities of computers, which today are almost unlimited: they can analyze billions of pieces of information at considerable speed and compare a multitude of parameters, whereas the human brain hardly manages more than 6 or 7 variables at the same time. Combined with the expertise of doctors, the development of artificial intelligence (AI) is therefore very promising in health.

But what is AI? This field relies on two important elements: algorithms and big data.

Algorithms

Algorithms are a series of operations and/or formal rules carried out by a computer program to solve a given type of problem. More concretely, these computer programs imitate modes of reasoning that call on intelligence.

There are AI programs where the tasks to be performed have been predefined by humans. This is, for example, a program that knows how to play chess because it has been provided with the rules of the game and the computing power necessary to anticipate the consequences of its choices.

There are also AI programs with learning , where the system learns by itself, as long as it is provided with the right exercises and especially the right answers. For example, if the program is provided with a multitude of photos of objects and their associated identifications, it will subsequently be able to distinguish a car from an airplane on any new photo. Some programs, if given enough data, are even able to discover matches on their own, without any prior guidance. We then speak of deep learning or deep learning.

Massive Data, Or Big Data

Another essential element for AI: massive data, or big data. In the field of health, digital data are multiplying: health insurance files, files of causes of death or of the national health data system, medical files of patients (results of examinations and imaging, reports surgical procedures, therapeutic protocols, etc.). To which are added data from research programs.

There are therefore many issues around all this information to allow its optimal use by AI. This includes ensuring their accuracy, their representativeness of real life, but also the quality of their organization and sharing (which requires a fair balance between data confidentiality and access to this data) .

In oncology

Structure patient data

The ConSoRe project (for Continuum Care – Research), set up by Unicancer, the network of centers for the fight against cancer, aims to organize the massive data collected in the field of oncology. It is a system that aims to collect, analyze and structure this data. Associated with a search engine, it allows doctors to identify patients meeting precise search criteria, to visualize the evolution of their disease and their treatments, to locate files of rare cancers, or even to know if similar cases have already been treated elsewhere. This data can then be processed by different algorithms.

Facilitate the diagnosis

It is probably in the field of diagnostics that AI will first disrupt oncology. Several research teams around the world have already proven that programs are able to detect melanoma, a severe form of skin cancer, or breast cancer more acutely than a doctor, even when it is an atypical tumor.