Skip to main content

Table 1 Keywords and rules for concepts of interest

From: RegEMR: a natural language processing system to automatically identify premature ovarian decline from Chinese electronic medical records

Field

Target concept

Keyword

Rule

Menstrual history

初潮年龄 (Menarche age)

初潮 (Menarche)

C1

月经周期 (Menstrual cycle)

月经周期 | / (Menstrual cycle)

C1

月经量 (Menstruation amount)

量 (Amount)

C3

Hormone test

日期 (Date)

年 | 月 | 日 (Year, month, day)

C2

FSH (Follicle stimulating hormone)

FSH

C1

LH (Luteinizing hormone)

LH

C1

E2 (Estrogen)

E2 | E

C1

P (Progesterone)

P | PRGE

C1

PRL (Prolactin)

PRL

C1

T (Testosterone)

T | TEST

C1

AMH (Anti-Müllerian hormone)

AMH | 抗苗勒管(氏)激素(Anti-Müllerian hormone)

C1

Ultrasonographic measures

子宫内膜厚度 (Endometrial thickness)

内膜(厚) | En (Endometrium)

C1

子宫位置 (Uterine position)

子宫 | UT (uterus)

C3

左卵巢卵泡个数 (left antral follicle counting, LAFC)

左(侧)卵巢 | 左(侧)附件 | Lov (Left ovary, left adnexa)

C1

右卵巢卵泡个数 (right antral follicle counting, RAFC)

右(侧)卵巢 | 右(侧)附件 | Rov (Right ovary, right adnexa)

C1

  1. We developed 3 rule categories. C1: keyword + quantity + unit, e.g., FSH 5.32 mIU/ml; C2: keyword + numeral, e.g., 量中; C3: year + month + day, e.g., 2018年10月3日. Different keywords of the same target concept were split by "|"