Skip to main content

Table 1 Diagnostic and treatment characteristics in operative and pathology report cohorts. SD; standard deviation, LN; lymph node

From: Automated medical chart review for breast cancer outcomes research: a novel natural language processing extraction system

Pathology Reports

Training Cohort

(n = 50)

Validation Cohort

(n = 50)

Laterality

 Unilateral

45 (90%)

49 (98%)

 Bilateral

5 (10%)

1 (2%)

Cancer Type

 Invasive

42 (76.4%)

41 (80.4%)

 Non-Invasive

13 (23.6%)

10 (19.6%)

Margins

 Positive

10 (18.2%)

9 (17.6%)

 Negative

45 (81.8%)

42 (82.4%)

Lymph Nodes

 Avg. LNs Examined (SD)

4.6 (4.3)

4.3 (3.4)

 Micro/Macro Metastasis

16 (29.1%)

13 (25.5%)

 Extranodal Extension

8 (14.5%)

3 (5.9%)

Pathologic Diagnosis

 Avg. Number of Foci (SD)

1.9 (2.0)

2.0 (2.2)

 Avg. Nottingham Score (SD)

6.3 (1.6)

6.7 (1.6)

 Avg. Tumour Size in mm. (SD)

28.2 (28.8)

27.9 (20.0)

 Lymphovascular Invasion

11 (20%)

10 (19.6%)

Operative Reports

Training Cohort

(n = 50)

Validation Cohort

(n = 50)

Laterality

 Unilateral

48 (96%)

49 (98%)

 Bilateral

2 (4%)

1 (2%)

Procedure Type

 Lumpectomy

19 (36.5%)

16 (31.4%)

 Nipple-Sparing Mastectomy

15 (28.9%)

21 (41.2%)

 Skin-Sparing Mastectomy

16 (30.7%)

14 (27.5%)

 Total Mastectomy

2 (3.8%)

0

Neoadjuvant Treatment

 Chemotherapy

4 (7.7%)

14 (27.5%)

 None

48 (92.3%)

37 (72.5%)

Immediate Reconstruction

 Mentioned

50 (96.2%)

45 (88.2%)

 Not Mentioned

2 (3.8%)

6 (11.8%)

Axillary Surgery

 Sentinel LN Biopsy

40 (76.9%)

38 (74.5%)

 Axillary LN Dissection

4 (7.7%)

5 (9.8%)

 None

8 (15.4%)

8 (15.6%)