Search Articles

View query in Help articles search

Search Results (1 to 4 of 4 Results)

Download search results: CSV END BibTex RIS


Evaluating Listening Performance for COVID-19 Detection by Clinicians and Machine Learning: Comparative Study

Evaluating Listening Performance for COVID-19 Detection by Clinicians and Machine Learning: Comparative Study

We used Adam optimizer to reduce the binary cross-entropy loss function on these training samples (more details are presented in a study by Han et al [22]). Once trained, the model takes one sample, consisting of 3 respiratory sounds (ie, breathing, coughs, and voice) from the same subject as input. The sample is processed through the model, and the model outputs a two-dimensional prediction, indicating the probability of the sample being from a positive- or negative-tested participant, respectively.

Jing Han, Marco Montagna, Andreas Grammenos, Tong Xia, Erika Bondareva, Chloë Siegele-Brown, Jagmohan Chauhan, Ting Dang, Dimitris Spathis, R Andres Floto, Pietro Cicuta, Cecilia Mascolo

J Med Internet Res 2023;25:e44804