Oral Presentations Abstracts: THE GOOD, THE BAD AND THE UGLY IN APP DIAGNOSIS: OUTCOMES AND IMPLICATIONS BY EXAMPLE

Sorana D. BOLBOACĂ; Adriana Elena BULBOACĂ

doi:10.24193/subbbioethica.2021.spiss.16

Authors

Sorana D. BOLBOACĂ Ph.D. Habil., Department of Medical Informatics and Biostatistics, "Iuliu Hațieganu" University of Medicine and Pharmacy, Cluj-Napoca, Romania. E-mail: sbolboaca@umfcluj.ro
Adriana Elena BULBOACĂ Department of Pathophysiology, "Iuliu Hațieganu" University of Medicine and Pharmacy, Cluj-Napoca, Romania.

DOI:

https://doi.org/10.24193/subbbioethica.2021.spiss.16

Abstract

View of Volume 66, Special Issue, September 2021

The Clinical Decision Support (CDS), a form of artificial intelligence (AI), consider physician expertise and cognitive function along with patient’s data as the input and case-specific medical decision as an output. The improvements in physician’s performances when using a CDS ranges from 13% to 68%.

The AI applications are of large interest nowadays, and a lot of effort is also put in the development of IT applications in healthcare. Medical decision support systems for non-medical staff users (MDSS-NMSF) as phone applications are nowadays available on the market. A MDSS-NMSF app is generally not accompanied by a scientific evaluation of the performances, even if they are freely available or not.

Two clinical scenarios were created, and Doctor31 retrieved the diagnosis decisions. First scenario: man, 29 years old, and three symptoms: dysphagia, weight loss (normal body mass index), and tiredness. Second scenario: women, 47 years old with L5-S1 disk herniation, abnormal anti-TPO antibodies, lower back pain (burning sensations), constipation, and tiredness.

The outcome possible effects and implications, as well as vulnerabilities induced on the used, are highlighted and discussed.