An AI algorithm used with x-ray helps detect tuberculosis in children, Spanish researchers have reported.
A team led by doctoral candidate Daniel Capellán Martín, of the Universidad Politécnica de Madrid (UPM) and colleagues from the Biomedical Research Networking Centre in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN) developed the algorithm. A study outlining the algorithm's use was published on 27 October in Nature Communications.
Martín's group reported the following:
- Pretraining AI models on adult chest x-rays improved performance when the model was refined with pediatric data.
- Lateral x-rays provided complementary information -- which was "especially valuable in infants and young children, where the frontal view alone may be insufficient," the investigators noted.
- The study showed that age-specific models were preferred "over models trained across all ages, due to differences in disease presentation and developmental stages among age groups," they wrote.
The algorithm is "not intended to replace the radiologist or physician, but rather to serve as a decision-support tool: it can help prioritize cases, guide screening decisions, and facilitate early detection in resource-limited settings," said co-author Begoña Santiago García, MD, of the Hospital General Universitario Gregorio Marañón in Madrid, in a statement released by UPM.
