AI on phone to detect children's lung issues
The study focused on developing an AI-based technique to analyze children's lung sounds using smartphones, aiming to simplify home respiratory assessments and support telemedicine services, especially in areas with limited access to healthcare. A research team from Seoul National University collected lung sound recordings from 611 children aged between zero and six years, using medical headphones and smartphones. They developed a deep learning model that adapts to device differences through audio processing techniques and variance reduction methods. Results showed that the new model achieved high accuracy, with an F1 score of 0.802 and an area under the ROC curve of 0.911 when tested on new recordings, highlighting its ability to identify abnormal lung sounds across different devices. This advancement is promising, as it could allow parents to record their children's breathing sounds and share them with doctors, thereby enhancing remote monitoring. However, further studies will be necessary to verify its effectiveness and safety before full clinical adoption.
Notice: This Is an AI-Generated Summary
Comments (0)