Predictivity in medicine with Artificial Intelligence

Share this post on:
Share on email
Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp

Prevention in the medical field is a universally recognised concept, but it had and ad still has different nuances according to the technologies available, the different application frameworks and other parameters, such as the epidemiological framework and the different guidelines followed in the health sector.

What is predictive medicine?

Even the definition of predictive medicine has different forms in the literature. In this context, we refer to the ability to find out in probabilistic terms which are the factors that can favour the onset of a certain disease in a specific person.

The goal of predictive medicine is therefore to find the fragility or the appearance that predisposes a healthy person to develop a certain disease.

This type of approach has a solid foundation in traditional medicine. By differentiating environmental risk factors from individual risk factors, it is possible, thanks to clear scientific evidence, to propose plans for prevention.

To give an example we can cite GARD, an initiative of 41 Italian and international organisations under the auspices of the WHO, which aims to achieve epidemiological surveillance and diagnosis for the treatment of chronic respiratory diseases.

Predicting (diagnosing) diseases with Artificial Intelligence

Medicine based on careful observation of the patient is a guiding principle dating back to the times of Hippocrates. Over the years, this criterion has changed to reach today’s definition of Evidence-Based Medicine (EMB).

Today, artificial intelligence is increasingly the best methodological approach to exploit big data to develop predictive and diagnostic models.

A machine is able to process millions of images in a short time, that is, it is possible to train an algorithm to recognise the differences and similarities, like (or better than) a radiologist would do.

Predictive medicine with the use of Artificial Intelligence goes further because it is able to recognise patterns within big data. By using the available data, AI is able to create new knowledge.

Rest assured, dystopian scenarios, in which the doctor is a robot with a white coat, are not being painted. In this regard, the reflections of two doctors are interesting: dr. Aldo di Benedetto and dr. Giampaolo Collecchia. What we want to highlight is the support that AI can give to those who work in the medical field.

Precision medicine

The information obtained can have immediate application and improve the national health systems. Early diagnoses can avoid hospitalisation of patients and according to Elena Bonfiglioli, who leads the Health Industry business for Microsoft in Europe, the Middle East and Africa, by only integrating and analysing big data it is possible to save 30% on public spending.

The scopes of application of AI in the medical field are making it possible to trace new paths of research. Among these, one of the most promising is precision medicine: the ability to perform diagnosis and treatment tailored to the individual patient.

By integrating information from the patient through databases and clinical analyses, it is possible to identify what are called alterations of pathways that determine the onset of the disease.

An article presented on the occasion of the “12th International Conference on Health Informatics”, entitled Predictive AI Models for the Personalized Medicine, underlines the potential offered by the use of predictive models with AI to help doctors choose the most appropriate therapeutic approach.

The 4Ps: predictive, preventive, personalised and participatory express the most important concepts for the development of predictive medicine.

eHealth data

AI is entrusted with the task of finding “invisible” evidence for humans (who are unable to process huge number of data), since the latter can represent a limit.

Medicine is based on the multiparametric classification of diseases, otherwise if the data are not acquired and organised according to shared standards, not only from a quantitative but also qualitative point of view, there is a risk of having unusable data sets.

Today it is possible to acquire data thanks to devices such as smartphones, smartwatches and other types of wearable medical devices. Apps have been developed to monitor the health status of weaker patients and to alert medical staff in case of critical events.

Digital transformation in healthcare

AI is a great opportunity, but in order to make it all work, an important digital transformation plan must be launched in the field of public health. eHealth is an extraordinary tool because it enables to improve citizens’ health and to save significant financial resources.

If you are looking for an authoritative partner for e-health solutions and projects, Ippocrate AS could be the solution for you.