Models | Cohorts | AUC (95% CI) | Sensitivity (%) (95% CI) | Specificity (%) (95% CI) | Accuracy (%) (95% CI) | PPV (%) (95% CI) | NPV (%) (95% CI) |
---|---|---|---|---|---|---|---|
GTV | Training set | 0.730 (0.654–0.798) | 68.57 (24/35) (52.17–84.97) | 67.77 (82/121) (59.44–76.10) | 67.95 (106/156) (60.63–75.27) | 38.10 (24/63) (26.10-50.09) | 88.17 (82/93) (81.61–94.74) |
 | Validation set 1 | 0.652 (0.546–0.748) | 73.91 (17/23) (52.17–84.97) | 57.14 (40/70) (45.55–68.74) | 61.29 (57/93) (51.39–71.19) | 36.17 (17/47) (22.43–49.91) | 86.96 (40/46) (77.22–96.69) |
 | Validation set 2 | 0.693 (0.609–0.768) | 72.00 (18/25) (54.40–89.60) | 62.28 (71/114) (53.38–71.18) | 64.03 (89/139) (56.05–72.01) | 29.51 (18/61) (18.06–40.95) | 91.03 (71/78) (84.68–97.37) |
 | Validation set 3 | 0.669 (0.576–0.753) | 75.00 (18/24) (57.68–92.32) | 59.14 (55/93) (49.15–69.13) | 62.39 (73/117) (53.62–71.17) | 32.14 (18/56) (19.91–44.37) | 90.16 (55/61) (82.69–97.64) |
GPTV3 | Training set | 0.872 (0.810–0.920) | 82.86 (29/35) (70.37–95.34) | 80.99 (98/121) (74.00-87.98) | 81.41 (127/156) (75.31–87.52) | 55.77 (29/52) (42.27–69.27) | 94.23 (98/104) (89.75–98.71) |
 | Validation set 1 | 0.779 (0.681–0.858) | 78.26 (18/23) (61.40-95.12) | 68.57 (48/70) (57.70-79.45) | 70.97 (66/93) (61.74–80.19) | 45.00 (18/40) (29.58–60.42) | 90.57 (48/53) (82.70-98.44) |
 | Validation set 2 | 0.803 (0.727-865) | 76.00 (19/25) (59.26–92.74) | 75.44 (86/114) (67.54–83.34) | 75.54 (105/139) (68.39–82.69) | 40.43 (19/47) (26.40-54.46) | 93.48 (86/92) (88.43–98.52) |
 | Validation set 3 | 0.752 (0.664–0.827) | 75.00 (18/24) (57.68–92.32) | 68.82 (64/93) (59.40-78.23) | 70.09 (82/117) (61.79–78.38) | 38.30 (18/47) (24.40–52.20) | 91.43 (64/70) (84.87–97.99) |
GPTV6 | Training set | 0.766 (0.691–0.830) | 80.00 (28/35) (66.75–93.25) | 70.25 (85/121) (62.10-78.39) | 72.44 (113/156) (65.42–79.45) | 43.75 (28/64) (31.60–55.90) | 92.39 (85/92) (86.97–97.81) |
 | Validation set 1 | 0.671 (0.566–0.765) | 69.57 (16/23) (50.76–88.37) | 65.71 (46/70) (54.59–76.83) | 66.67 (62/93) (57.09–76.25) | 40.00 (16/40) (24.82–55.18) | 86.79 (46/53) (77.68–95.91) |
 | Validation set 2 | 0.760 (0.680–0.828) | 76.00 (19/25) (59.26–92.74) | 68.42 (78/114) (59.89–76.95) | 69.78 (97/139) (62.15–77.42) | 34.55 (19/55) (21.98–47.11) | 92.86 (78/84) (87.35–98.36) |
 | Validation set 3 | 0.726 (0.636–0.805) | 79.17 (19/24) (62.92–95.41) | 61.29 (57/93) (51.39–71.19) | 64.96 (76/117) (56.31–73.60) | 34.55 (19/55) (21.98–47.11) | 91.94 (57/62) (85.16–98.71) |
GPTV9 | Training set | 0.809 (0.739–0.868) | 74.29 (26/35) (59.81–98.71) | 76.03 (92/121) (68.43–83.64) | 75.64 (118/156) (68.91–82.38) | 47.27 (26/55) (34.08–60.47) | 91.09 (92/101) (85.53–96.65) |
 | Validation set 1 | 0.713 (0.610–0.802) | 82.61 (19/23) (67.12–98.10) | 65.71 (46/70) (54.59–76.83) | 69.89 (65/93) (60.57–79.22) | 44.19 (19/43) (29.34–59.03) | 92.00 (46/50) (84.48–99.52) |
 | Validation set 2 | 0.725 (0.643–0.797) | 72.00 (18/25) (54.40–89.60) | 67.54 (77/114) (58.95–76.14) | 68.35 (95/139) (60.61–76.08) | 32.73 (18/55) (20.33–45.13) | 91.67 (77/84) (85.76–97.58) |
 | Validation set 3 | 0.680 (0.588–0.763) | 70.83 (17/24) (52.65–89.02) | 65.59 (61/93) (55.94–75.25) | 66.67 (78/117) (58.12–75.21) | 34.69 (17/49) (21.37–48.02) | 89.71 (61/68) (82.48–96.93) |
GPTV12 | Training set | 0.764 (0.690–0.828) | 77.14 (27/35) (63.23–91.05) | 68.60 (83/121) (60.32–76.87) | 70.51 (110/156) (63.36–77.67) | 41.54 (27/65) (29.56–53.52) | 91.21 (83/91) (85.39–97.03) |
 | Validation set 1 | 0.645 (0.539–0.741) | 73.91 (17/23) (55.97–91.86) | 67.14 (47/70) (56.14–78.15) | 68.82 (64/93) (59.40-78.23) | 42.50 (17/40) (27.18–57.82) | 88.68 (47/53) (80.15–97.21) |
 | Validation set 2 | 0.690 (0.606–0.766) | 80.00 (20/25) (64.32–95.68) | 57.02 (65/114) (47.93–66.11) | 61.15 (85/139) (53.05–69.25) | 28.99 (20/69) (18.28–39.69) | 92.86 (65/70) (86.82–98.89) |
 | Validation set 3 | 0.633 (0.539–0.720) | 62.50 (15/24) (43.13–81.87) | 66.67 (62/93) (57.09–76.25) | 65.81 (77/117) (57.22–74.41) | 32.61 (15/46) (19.06–46.16) | 87.32 (62/71) (79.58–95.06) |
GPTV15 | Training set | 0.727 (0.650–0.795) | 71.43 (25/35) (56.46–86.40) | 68.60 (83/121) (60.32–76.87) | 69.23 (108/156) (61.99–76.47) | 39.68 (25/63) (27.60-51.76) | 89.25 (83/93) (82.95–95.54) |
 | Validation set 1 | 0.629 (0.522–0.727) | 60.87 (14/23) (40.92–80.82) | 64.29 (45/70) (53.06–75.51) | 63.44 (59/93) (53.65–73.23) | 35.90 (14/39) (20.84–50.95) | 83.33 (45/54) (73.39–93.27) |
 | Validation set 2 | 0.652 (0.567–0.731) | 56.00 (14/25) (36.54–75.46) | 67.54 (77/114) (58.95–76.14) | 65.47 (91/139) (57.56–73.37) | 27.45 (14/51) (15.20–39.70) | 87.50 (77/88) (80.59–94.41) |
 | Validation set 3 | 0.570 (0.484–0.670) | 79.17 (19/24) (62.92–95.41) | 43.01 (40/93) (32.95–53.07) | 50.43 (59/117) (41.37–59.49) | 26.39 (19/72) (16.21–36.57) | 88.89 (40/45) (79.71–98.07) |