A model that can achieve accuracy in line with that of radiologists when labeling open-access datasets could be a key factor to overcoming limitations of artificial intelligence implementation, researchers explained in Nature Communications.

Experts involved with the study suggested that these findings could be beneficial in the future of ovarian tumor assessment by providing clinical decision making support.

Homogenous datasets can create unintended research bias that hinders the clinical efficacy of AI applications, experts recently explained in PLOS Digital Health.

A study in Lancet Digital Health reports that a previously validated, high performing AI model committed troublesome errors when confronted with atypical anatomy.

A deep convolutional neural network was able to predict hospital stay, ICU admission and intubation when scoring chest CT images of hospitalized COVID patients.

The new algorithm flags chest x-rays when it detects pneumothorax and notifies physicians of any suspicious findings.

Research published recently in Radiology found comparable sensitivity and specificity between artificial intelligence and clinicians for fracture detection. 

 "This research aims to raise a red flag regarding naive off-label usage of open-access data in the development of machine-learning algorithms," experts involved in the study said.