Table 4.
Application of artificial intelligence in automatic coronary calcium scoring.
| Study | Year | Algorithm | ICC | κ | Accuracy |
|---|---|---|---|---|---|
| Fischer et al. (45) | 2020 | RNN (LSTM) | NA | 0.85 | 0.903 |
| Wolterink et al. (46) | 2016 | CNN | 0.944 | 0.83 | 83% |
| Lee et al. (47) | 2021 | CNN | 0.99 | 0.94 | NA |
| de Vos et al. (48) | 2019 | CNN | 0.98 | 0.95 | 0.99 |
| van Assen et al. (49) | 2021 | CNN | 0.921 | 0.74 | 0.7 |
CNN, convolutional neural network; CCTA, coronary computed tomography angiography; ICC, Intra-class correlation coefficient; κ, Cohen's linearly weighted kappa; NA, not applicable; CAC, coronary artery calcium; LSTM, long short-term memory; RNN, recurrent neural network.