Limitations
After the development of an AI-ECG model, rigorous external validation
in diverse populations and clinical trials to demonstrate superior
clinical outcomes to the standard of care is needed. To date, most
models have not undergone rigorous evaluation and thus their results
remain confined to their own datasets. Furthermore, due to the several
knowledge gaps surrounding AF, it becomes challenging to prove immediate
benefits in clinical outcomes.
A limitation to widespread applicability of these models is the need to
be tailored to the target application. Although, cross domain utility of
a pre-trained model to perform the same function has shown promising
results. For instance, in a study a DNN model was trained to detect
paroxysmal AF from 24-hour Holter ECG and then fine tuned to a
completely different input data of smartwatch PPG signal with superior
results (AUC of 0.97).79 However, as noted by authors,
this study was limited by a lack of gold standard labels to study true
performance.79
Several other limitations surrounding AI such as data security,
perpetuating bias, physician acceptance and regulatory considerations
are also important to be addressed and have been discussed elsewhere in
more detail.8,89