Table 1 Overview of studies using radiological features to predict lymph node metastasis in patients with cervical cancer
Imaging modality | Author, year | Training/testing/validation set | Radiomics/clinical features used | Algorithm | AUC (95%CI) |
|---|---|---|---|---|---|
PET/CT | Zhang, 2022 [18] | 104/44/0 | 14/0 | Machine learning | 0.786 (0.636–0.895) |
MRI | Song, 2021 [15] | 90/42/0 | 7/4 | Logistic regression | 0.75 (-) |
CT | Liu, 2021 [20] | 148/74/51 | 464/0 | Artificial neural network | 0.859 (0.776–0.941) |
MRI | Xiao, 2020 [19] | 155/78/0 | 23/0 | Logistic regression | 0.883 (0.809–0.957) |
CT | Dong, 2020 [21] | 176/50/0 | 5/2 | Logistic regression, support vector machine, deep neural network | 0.99 (-) |
Ultrasound | Jin, 2020 [14] | 100/72/0 | 6/0 | Logistic regression | 0.77 (0.65–0.88) |
CT | Chen, 2020 [22] | 104/46/0 | 2/1 | Ridge logistics regression | 0.75 (0.53–0.93) |
MRI | Kan, 2019 [17] | 100/43/0 | 10/0 | Support vector machine | 0.754 (0.584–0.924) |
MRI | Wu, 2019 [16] | 126/63/0 | 14/1 | Support vector machine | 0.847 (-) |