The Limited Times

Now you can see non-English news...

Psoriasis: the first artificial intelligence program for a key point of the disease is created in the country

2023-04-10T11:41:02.288Z


Doctors will now be able to determine the degree of severity only with an app on a cell phone. It was endorsed by the Argentine Society of Dermatology.


Psoriasis lesions, its famous "plaques", are easily recognizable to a trained dermatologist.

However, the disease, which

half a million Argentines

suffer according to data from the Global Atlas of Psoriasis and which skyrocketed in a pandemic, has multiple degrees of severity that are measured with an international standard, the so-called

PASI score

.

Determining this degree is 

essential for the patient 

and finding this information is a process that requires a lot of consulting time, a well-trained medical eye and several mathematical calculations.

Now, for the first time in the world,

an artificial intelligence program has managed to do the PASI by the doctor

.

And they developed it in Argentina.

Artificial intelligence

promises

to be one of the topics of the year, after a group of experts warned about its potential risks.

But the concrete thing is that today it is making

significant advances in the field of health

, where it promises sooner rather than later to cause a revolution by accelerating and improving diagnoses.

Fast PASI is an example of this.

With just a cell phone, determine the

Psoriasis Area Severity Index

(PASI), the indicator most used by dermatologists in the world.

The app was developed by a group of local entrepreneurs with the advice of doctors specializing in psoriasis and

obtained the endorsement of the Argentine Society of Dermatology (SAD)

.

“Identifying psoriasis is easy.

But performing the PASI, which is a fundamental step when choosing a treatment to measure the severity of the patient, generates a significant

loss of time

and a

need for

professional training.

Until now you had to analyze the injuries piece by piece throughout the body and make a series of calculations that could take more than half an hour.

With the application, you take five photos with your cell phone and

the algorithm does the PASI by itself

, ”explains Alberto Lavieri, coordinator of the Psoriasis group at SAD. 

Doctors, biologists and systems experts worked on the development.

Alberto Adem, one of the partners behind Fast PASI, says that it arose precisely because they saw

a need for dermatologists

.

That there are already professionals who are using it and that the goal is for the platform to be freely

available

to everyone, for which they are now looking for some type of sponsorship from the pharmaceutical industry.

The correlation of results between the PASI made by the app and that made by the human being is very high.

“We compared the severity provided by data from pivotal studies with the analysis of the app's photographs and it came within

perfect ranges

,” Lavieri says.

The dermatologist has to take a photo of the patient following the indications of the app.

Photo Fast PASI

How does the artificial intelligence algorithm work in this case?

At the base, in the same way as the controversial ChatGPT:

learning

.

The training lasted more than six months.

They “made him see” more than 2,500 photos of patients with psoriasis in which they first taught him to define the area of ​​the body.

Then to see if he could determine the redness, thickness, and scaling.

“Iterations and corrections were made until

a level of learning was reached

.

And with a part of images that were not used before, they were uploaded so that the algorithm would identify them.

One part was used for him to learn and another, for him to

take the exam

”, graphic Adem.

Why PASI is so important

The PASI was defined in 1978 by two Swedish researchers and is today the most widely used system to assess the

severity of psoriasis

.

It contemplates four variants to analyze the injuries in the different parts of the body.

Combine the severity with the affected area to get a number between 0 and 72. Based on this index, psoriasis is classified as

mild, moderate, or severe

.

One of the tests of how the application returns the results.

Photo Fast PASI

This classification is

key

for two reasons.

First, because depending on the severity of the psoriasis, the treatment will be indicated.

Second, because for the most severe degrees there are already systemic drugs with good results, but they are included in the group of so-called

high-cost drugs

, whose access is not easy because many times prepaid and social works hinder it.

And that is where the PASI comes in: it is the

fundamental evaluation that all health providers request

to comply with medication coverage.

“All funders ask you for a detailed PASI.

Here you have a detailed PASI and not made by a dermatologist but

approved by the Argentine Society of Dermatology

”, Lavieri remarks about the usefulness of this tool.

The specialist sees several.

That

democratizes access to medicine

, that facilitates faster access to treatments that improve the quality of life, that allows the doctor to have a kind of virtual clinical history of the patient in which he can follow up, and that at the same time protects your privacy because

the data cannot be shared

.

And regarding the debate over

the risks of artificial intelligence

, add your look.

“It can be a problem when it is given as perfect and there is no error, because it may be missing something that the algorithm does not listen to.

But here a diagnosis is not being made:

the diagnosis was already made by the doctor before

.

The only thing the algorithm does is speed up a lot of things and gain that time to dedicate it precisely to the doctor-patient relationship”.

ACE

look also

Psoriasis, the disease that exploded in the pandemic

Why emotions are also expressed on the skin

Source: clarin

All life articles on 2023-04-10

You may like

Trends 24h

Latest

© Communities 2019 - Privacy

The information on this site is from external sources that are not under our control.
The inclusion of any links does not necessarily imply a recommendation or endorse the views expressed within them.