Skip to main content

Table 2 Characteristics of the reviews and screening predictions for each tool

From: Performance and usability of machine learning for screening in systematic reviews: a comparative evaluation of three tools

Characteristic

Antipsychotics, N records (%)

Bronchiolitis, N records (%)

Visual Acuity, N records (%)

Screening workloada

12,156

5861

11,229

Included by title/abstractb

1178 (10)

518 (9)

224 (2)

Included in the reviewb

127 (1)

137 (2)

1 (< 1)

Includes/excludes in training set

Abstrackr, 15/185

Abstrackr, 12/188

Abstrackrc, 4/296

DistillerSR, 14/186

DistillerSR, 14/186

DistillerSR, 2/198

RobotAnalyst, 20/180

RobotAnalyst, 15/185

RobotAnalyst, 3/197

Screened by toold

11,956 (98)

5661 (97)

11,029 (98)

Predicted relevant by Abstrackr

2117 (18)

656 (12)

3639 (33)

Predicted relevant by DistillerSR

7 (< 1)

83 (1)

0 (0)

Predicted relevant by RobotAnalyst

3488 (29)

1082 (19)

3221 (29)

  1. aTotal number of records retrieved via the electronic searches. Each record was screened by two reviewers
  2. bIncluded following the initial screening by two independent reviewers (retrospective)
  3. cAll training sets were 200 records, with the exception of the Visual Acuity review which required a 300-record training set in Abstrackr before predictions were produced
  4. dAfter a 200-record training set