Coronavirus and big data, a possible solution

Big-Data-Coronavirus

An interesting proposal by Dr. Roberto Grinta, Director of the Department of Services AV2 ASUR Marche, which we gladly receive and publish.

The International Committee on Taxonomy of Viruses has named the new coronavirus “severe acute respiratory syndrome coronavirus 2” (SARS-CoV-2). The outbreak of COVID-19 (where “CO” stands for coronavirus, “VI” for virus, “D” for disease and “19” indicates the year in which it occurred), has been declared by the World Health Organization as a public health emergency of international importance. To date, the virus has reached all countries in the world, so much so that the World Health Organization has changed the classification of the Coronavirus from epidemic to pandemic.

A pandemic (from the Greek pan-demos, “all the people”) is precisely an epidemic disease that expands rapidly, spreading to more geographical areas of the world, and which involves many people. This situation presupposes the lack of human immunization towards a highly virulent pathogen.

According to the WHO, the conditions for classifying the pandemic are essentially three:

  1. the appearance of a new pathogen agent,
  2. the ability of that agent to affect humans,
  3. the ability of that agent to spread rapidly by contagion.

Against the Coronavirus, the States have put in place measures to contain the spread of the epidemic, as a lower incidence of new cases could allow to slow down the epidemic curve and, therefore, reprogram access in ICU for subjects with compromised lung diagnosis, with EGA in O2 for patients with SpO2<90% in ambient air, chest Rx showing bilateral pulmonary thickening such as interstitial pneumonia.

The average stay in ICU of the COVID-19 patient, is about 21 days, so it becomes important to define the path upstream of the process, separating the suspicious cases waiting for a swab from the ascertained patients who may have to go into resuscitation if their functions are compromised.

The response of our Health System has been to work on a deep and continuous transformation of the hospitals of our National Health Service, both Hub and Spoke, trying to improve the efficiency of the path of the patient affected by COVID, separating the paths with those NO-COVID.

The population most affected by the Coronavirus, at least from the first studies (which, however, are still too far from having a useful database to express effective guidelines against the virus), seems to be represented by over-65s with proven pathologies and comorbidities, 80% of which with an average age of eighty years.

From this point of view, however, the health system is equipped with digital information that will be the core of the new organizational model, especially the reorganization of health planning.

Patients with comorbidities with more fragile multiple pathologies are known and therefore “ideal” target for the transmission of Coronavirus and subsequent complications.

The registry of these patients is able to provide all the clinical indications of the patient, also through the patient summary of the General Practitioner who connects to the Electronic Health Record.

The prevention and therefore the continuous remote monitoring of “fragile” subjects can reduce the incidence and therefore the access to the hospital. Moreover, through artificial intelligence it is possible to digitally apply algorithms able to simulate human cognitive skills in analysing clinical data and to come to conclusions autonomously, without further human input, thus supporting the clinician in identifying signs of probable pathologies.

Secondly, it is possible to think about the possibility of data extraction and simulate, through new algorithms, predictive signals of pathology related to the spread, linking the simulation also to the power with which the virus binds to the contact surfaces. In this way, standards could be redefined according to the pathology of the subject through both pathology and condition exemptions.

Of course, the information network cannot escape the constraints dictated by privacy requirements and data management to decision-makers in health planning.

The subjects assessed most at risk could be monitored remotely, through consoles set up at regional and/or company level, such as to allow the possibility of immediate intervention if standard parameters defined for the pathology should be altered.

Managing the health data platform through passwords provided to the patients, who can access the central database to enter clinical data, such as the outcome of some predictive tests is possible.

The opportunity to manage remotely therefore provides the possibility to intervene promptly in a clinical way and directly through the activation of a network system between professionals, both in hospital and in the local area.

Moreover, the available data could be linked to algorithms that are able to establish the actual count of beds for suspected, positive and resuscitated COVID patients. In fact, it is useful to have a predictive path of beds, based on population comorbidity data and therefore approach the epidemiological problem on a predictive basis.

The hypothesized procedure – certainly feasible in a short period of time – even by funding remote telemedicine projects, would be able to introduce a new method to address the problem of the spread of Coronavirus through prevention, continuous monitoring at home, in the office, without drastically changing the habits of each citizen.

Roberto Grinta

Director of the Department of Services AV2 ASUR Marche

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s