Table 5 Distributions of predictive values of algorithms and pulmonologists in each image pattern.
From: A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography
CXR patterns | Favor TB pattern (n = 30) | Abnormal pattern (n = 43) | Normal pattern (n = 32) | p value |
|---|---|---|---|---|
Model 1 | 1.0 (0.87–1.0) | 0 (0–0.89) | 0 (0–0.03) | < 0.001 |
Model 2 | 0.97 (0.81–1.0) | 1.0 (0.98–1.0) | 0.01 (0–0.13) | < 0.001 |
Model 3 | 0.97 (0.64–0.99) | 0.5 (0.5–0.9) | 0.03 (0.005–0.13) | < 0.001 |
V1 | 0.8 (0.8–0.9) | 0.2 (0.1–0.4) | 0 (0–0) | < 0.001 |
V2 | 0.8 (0.65–0.9) | 0.4 (0.3–0.5) | 0 (0–0) | < 0.001 |
V3 | 0.8 (0.6–0.8) | 0.2 (0.2–0.4) | 0 (0–0) | < 0.001 |
V4 | 0.9 (0.7–0.9) | 0.3 (0.1–0.3) | 0 (0–0) | < 0.001 |
V5 | 0.6 (0.2–0.7) | 0.1 (0.1–0.15) | 0.05(0.05–0.05) | < 0.001 |
