Table 1 Characteristics of patients with radiology reports for analysis.
From: Artificial intelligence-aided clinical annotation of a large multi-cancer genomic dataset
Total number of patients and radiology reports | Number of patients with unlabeled radiology reports and number of unlabeled radiology reports | Number of patients with labeled radiology reports and # of labeled radiology reports | ||||
|---|---|---|---|---|---|---|
Patients N (%) | Reports N (%) | Patients N (%) | Reports N (%) | Patients N (%) | Reports N (%) | |
Total cohort | 13130 (100) | 304160 (100) | 10300 (100) | 272964 (100) | 2830 (100) | 31196 (100) |
Sex | ||||||
Male | 5621 (43) | 105503 (35) | 4055 (39) | 89849 (33) | 1566 (55) | 15654 (50) |
Female | 7509 (57) | 198657 (65) | 6245 (61) | 183115 (67) | 1264 (45) | 15542 (50) |
Age at next generation genomic sequencing | ||||||
<40 | 625 (5) | 14439 (5) | 488 (5) | 12835 (5) | 137 (5) | 1604 (5) |
40–49 | 1329 (10) | 30868 (10) | 999 (10) | 26490 (10) | 330 (12) | 4378 (14) |
50–59 | 3092 (24) | 75681 (25) | 2400 (23) | 67920 (25) | 692 (24) | 7761 (25) |
60–69 | 4172 (32) | 99399 (33) | 3295 (32) | 90158 (33) | 877 (31) | 9241 (30) |
70–79 | 2944 (22) | 65229 (21) | 2335 (23) | 58700 (22) | 609 (22) | 6529 (21) |
80+ | 968 (7) | 18544 (6) | 783 (8) | 16861 (6) | 185 (7) | 1683 (5) |
Race as recorded in the electronic health record | ||||||
Asian | 424 (3) | 10724 (4) | 353 (3) | 9716 (4) | 71 (3) | 1008 (3) |
African-American | 458 (3) | 10649 (4) | 348 (3) | 9470 (3) | 110 (4) | 1179 (4) |
Native American | 11 (<1) | 193 (<1) | 10 (<1) | 184 (<1) | 1 (<1) | 9 (<1) |
Pacific Islander | 4 (<1) | 144 (<1) | 4 (<1) | 144 (<1) | 0 (0) | 0 (0) |
White | 11760 (90) | 272156 (89) | 9205 (89) | 244173 (89) | 2555 (90) | 27983 (90) |
More than one race | 39 (<1) | 729 (<1) | 33 (<1) | 652 (<1) | 6 (<1) | 77 (<1) |
Other/unknown | 434 (3) | 9565 (3) | 347 (3) | 8625 (3) | 87 (3) | 940 (3) |
Cancer type | ||||||
Breast | 2029 (15) | 63789 (21) | 1676 (16) | 58209 (21) | 352 (12) | 5527 (18) |
Colorectal | 1958 (15) | 37570 (12) | 1493 (14) | 32986 (12) | 466 (16) | 4588 (15) |
Endometrial | 482 (4) | 9801 (3) | 482 (5) | 9801 (4) | 0 (0) | 0 (0) |
Gastroesophageal | 878 (7) | 19794 (7) | 878 (9) | 19794 (7) | 0 (0) | 0 (0) |
Head and neck | 461 (4) | 8796 (3) | 460 (4) | 8795 (3) | 0 (0) | 0 (0) |
Leiomyosarcoma | 144 (1) | 6241 (2) | 144 (1) | 6241 (2) | 0 (0) | 0 (0) |
Non-small cell lung | 3378 (26) | 82609 (27) | 2763 (27) | 73758 (27) | 614 (22) | 8838 (28) |
Melanoma | 733 (6) | 20621 (7) | 731 (7) | 20591 (8) | 0 (0) | 0 (0) |
Ovarian | 646 (5) | 22248 (7) | 646 (6) | 22248 (8) | 0 (0) | 0 (0) |
Pancreatic | 685 (5) | 7854 (3) | 295 (3) | 4477 (2) | 394 (14) | 3450 (11) |
Prostate | 617 (5) | 7506 (2) | 164 (2) | 2851 (1) | 453 (16) | 4676 (15) |
Renal cell carcinoma | 499 (4) | 4737 (2) | 84 (<1) | 1721 (<1) | 415 (15) | 3016 (10) |
Urothelial carcinoma | 620 (5) | 12594 (4) | 484 (5) | 11492 (4) | 136 (5) | 1101 (4) |
Common tumor genomic variants | ||||||
TP53 mutation | 5330 (41) | 124663 (41) | 2486 (42) | 112237 (41) | 1044 (37) | 12426 (40) |
KRAS mutation | 2785 (21) | 53735 (18) | 2012 (20) | 45775 (17) | 773 (27) | 7960 (26) |
PIK3CA mutation | 1738 (13) | 43168 (14) | 1455 (14) | 39788 (15) | 283 (10) | 3380 (11) |
APC mutation | 1215 (9) | 24381 (8) | 942 (9) | 21627 (8) | 273 (10) | 2754 (9) |
BRAF mutation | 688 (5) | 15938 (5) | 587 (6) | 14918 (5) | 101 (4) | 1020 (3) |
