Evaluation of a prototype machine learning tool to semi-automate data extraction for systematic literature reviews

Table 1 Definitions of the entity types

Entity type	Description	Example entity*
SLR 1
Arm description	Treatment arm description phrase	‘dexamethasone’
Arm dosage	Amount or frequency of a treatment	‘We treated patients with drug X at 1.3 mg/m²’
PFS metric	Metric used to describe PFS	‘median progression-free survival’
PFS result	Numeric measurement associated with a PFS metric	‘10 months’
Study type	Type of study design	‘randomized controlled trial’
Title	Title of the publication	‘A study to investigate multiple myeloma’
Authors	Authors of the publication	‘M Smith’
SLR 2
Age metric	Metric used to measure the age of patient populations	‘mean (SD) age’
Age number	Numeric measurement associated with an age metric	‘60 years’
Arm description	Treatment arm description phrase	‘dexamethasone’
Arm dosage	Amount or frequency of a treatment	‘We treated patients with drug X at 1.3 mg/m²’
eGFR metric	Metric used to describe eGFR	‘mean eGFR’
eGFR number	Numeric measurement associated with an eGFR metric	‘20 mL/min/1.73 m²’
eGFR subgroup	Population subgroup	‘Among patients >60 years old’
eGFR time point	Time period over which the metric was measured	‘at 4-month follow-up’
Study type	Type of study design	‘observational study’

eGFR Estimated glomerular filtration rate, PFS Progression-free survival, SLR Systematic literature review
^*These examples serve to illustrate the definition of the entity types and were not taken from the dataset

ISSN: 2046-4053