CareerLocatorContactINVESTORS 
Skip to main content
SYNLAB

Patients & Individuals

Healthcare providers

Organisations
Newsroom Lab Locator Value of Diagnostics Partners of Expertise Quality Focus on patient health Career Investor Relations

Rossana's Interview & Publication

SYNLAB Italy

Rossana Castaldo

Biomedical Engineer, IRCCS SDN-SYNLAB, Naples

What inspired your research and what did it cover?
I am convinced that research and particularly the use of Artificial Intelligence (AI) in healthcare can enhance preventive care and quality of life. New technologies support us in producing more accurate diagnoses and treatment plans and lead to better patient outcomes.

My research aims to use AI for a symptom-based diagnosis to detect pneumonia. The goal is to develop an easy-to-use, diagnostic aid for healthcare workers in low-resource settings, which have generally poor environmental and operational conditions.

Looking at the potential of your findings, what difference can they make?
When resources are scarce, AI has been promoted as a potential means of strengthening healthcare systems. My study highlights that an evidence-based and interpretable AI model can automatically detect patients affected by pneumonia. This classification model is highly suitable for creating an easy-to-use diagnostic tool, such as a mobile phone application for patient screening. Diagnostic tools like these could be used with minimal training or even by patients themselves, which can facilitate task shifting (i.e., rational redistribution of tasks among health teams) and reduce the pressure on clinical staff. Indeed, there is an ongoing global need for such technology, emphasised greatly by the COVID-19 pandemic.

Publication

A machine learning model for supporting symptom-based referral and diagnosis of bronchitis and pneumonia in limited resource settings

ResearchGate – https://www.researchgate.net/publication/354661507

This study presents a machine learning (ML) application for pneumonia detection in low-resource settings (LRS). Such diagnostic tools can also be used by patients themselves with a minimum of training, which can ease task distribution and reduce pressure on clinicians. Indeed, there is a growing need for such technology worldwide, which is exacerbated by the COVID-19 pandemic.

 

 back