Artificial intelligence (AI) can be used to detect COVID-19 infection from people's voices using a cell phone app, according to a study by the Institute for Data Science at Maastricht University in the Netherlands.  The artificial intelligence model used in this study is more accurate than lateral flow/rapid antigen tests and is also cheap, fast and easy to use, meaning it can be used in low-income countries where PCR tests are expensive and/or difficult to distribute, News Medical Life Sciences writes.  COVID-19 infection usually affects the upper respiratory tract and vocal cords, resulting in changes in a person's voice. Scientists decided to see if artificial intelligence could be used to analyze the voice to detect COVID-19. They used data from Cambridge University's COVID-19 Sounds App, which contains 893 audio samples from 4,352 healthy and unhealthy participants, 308 of whom tested positive for COVID-19. The app is installed on a cell phone, participants provide basic information about their medical history and smoking status, and then they are asked to record some breathing exercises.  The researchers used Mel-spectrogram analysis, a voice analysis method that detects various voice characteristics such as loudness, strength and change over time. They found that one model called Long-Short Term Memory (LSTM) outperformed all other models. LSTM is based on neural networks that mimic the human brain and recognize basic relationships in the data. It works with sequences, which makes it suitable for modeling signals collected over time, such as voices, due to its ability to store data in memory. Its overall accuracy was 89%, its ability to correctly identify positive cases (true positive rate or "sensitivity") was 89%, and its ability to correctly identify negative cases (true negative rate or "specificity") was 83%.