Today medicine is manly focused on the disease but many are trying to shift the attention on the patient. Not surprisingly current medical practices promote to apply similar treatments to any type of patient, without considering the peculiar health characteristics that make unique each person.
This mainly happens because drugs continue to be tested on the mass, so that if the treatment works for the average person then it works for everyone. In recent years, however, we have been understanding that this approach can be improved: the same treatment plan can have different impacts on the patients, especially regarding the evaluation of its side effects.
That’s why best physicians are starting to require to have systems able to determine the effectiveness of a treatment a priori, testing in real-time drug’s effects on the treated patient. This approach is consistent with a new vision of medicine that is leading more and more professionals in the sector to address health problems through the eyes of the patient.
The transition from a disease-centric system to a patient-centric one, which see patients less passive and more and more active and the protagonist of their own healing process, is opening a new way of doing medicine.
Modern medicine will no longer follow an epidemiological approach but will be entirely personalized: patients of the near future will no longer want to know what is their probability of developing a certain disease or recovering from it, but they will be interested in understanding what exactly will happen to them and not to the average person.
This requires that physicians cannot simply limit themselves to describe the dynamics of a disease, rather they should learn how to correctly predict the clinical scenarios patients may face on the basis of their health history.
The health system of the future will be prospective and no longer reactive like the actual one. We will be able to understand in advance if a certain disease will manifest itself without being forced to diagnose it only when it is too late.
This will be possible thanks to an insightful overview on health data – collected by hospitals or through alternative sources (mobile applications, smartwatches, wearable medical devices) – and the ability to perform a much greater number of medical tests in an extremely easier way.
We will have tons of information available that will allow us to broaden our views, thanks to the ability to spot – through the use of artificial intelligence technologies – correlations, trends and hidden patterns within our health data.
Technological advances will not diminish the figure of the physician, but on the contrary they will represent a formidable tool to improve the quality of his work, amplifying his ability to correctly diagnose diseases and structure more effective therapeutic plans.
In the future, the physician who will not make the effort to embrace this technological revolution – despite extremely competent – will no longer be able to keep up with doctors who will instead choose the path of artificial intelligence and big data.
Based on this vision, iCareX has developed iTwinDiscover, an end-to-end cloud platform that allows physicians to have an insightful overview on the patient’s health profile, make data-driven decisions and provide patients with personalized AI-based treatments. iTwinDiscover is equipped with a range of artificial intelligence models that can help the physician to predict the patient’s response to different care plans used for the treatment of a specific disease.
For example, our algorithm – K2AI (Knowledge Augmented Artificial Intelligence) – is able to combine the solid medical foundation and knowledge of disease mechanics with flexible and data-driven models and optimization routines based on the artificial intelligence of Machine Learning, to inform physicians about the impact of various treatments on patient’s biomarkers.
Another algorithm developed by our team – Crystal – allows physicians to select the treatment plan most suited to the patient’s clinical situation and to dynamically manage the dosages over time, evaluating the possible effects. This is made possible thanks to the development of a spurious representation of his health profile which is independent of external factors to the treatment that could produce a bias in the predictions of the model.
The flexibility of our algorithms allows us to operate with a very large pool of diseases, particularly in situations where national or international guidelines require the physician to make choices about the therapeutic actions to be taken in order to preserve the patient’s health. In these sensitive situations, an optimal combination of data and intelligent tools can help healthcare professionals and patients to make more conscious and informed decisions.