Skip to content
flowss
Start freeChecking your session
PRISMA 2020 flow templates

ML in Radiology Review (PRISMA)

PRISMA flow for a deep-learning literature review.

PRISMA 2020 flowResearch & AcademicPRISMA 2020 flow syntax docs
Template previewPRISMA 2020 flow
Deep learning in radiology review3,460 records identified · 3 exclusion reasonsIdentificationScreeningEligibilityIncludedRecords identified from:n = 3,460Databases: 3,400 · Registers: 60Records removed before screeningn = 920Duplicates removed: 720Excluded by automation: 200Records screened (title / abstract)n = 2,540Records excluded by reviewern = 1,960Reports sought for retrievaln = 580Reports not retrievedn = 30Reports assessed for eligibilityn = 550Reports excluded:n = 295No ground truth: n = 140Not deep learning: n = 95Review article: n = 60Studies included in reviewn = 255Every record is accounted for: all 4 stages of the flow balance.OKAll 10 stated counts are whole numbers of records.OKIdentification balances: 3,400 + 60 − 720 − 200 = 2,540 `title_screened`.OKScreening balances: 2,540 − 1,960 = 580 `abstracts_sought`.OKRetrieval balances: 580 − 30 = 550 `fulltext_assessed` (`other_sought` and `other_unavailable` are not stated and read as zero).OKEligibility balances: 550 − 295 = 255 `studies_included`.OKThe funnel descends: all 3 stage-to-stage comparisons hold — no count is larger than the one before it.OKAll 3 exclusion reasons are named and distinct, accounting for 295 reports.

Make it your own.

title "Deep learning in radiology review"
db_records 3400
register_records 60
duplicates 720
excluded_auto 200
title_screened 2540
title_excluded 1960
abstracts_sought 580
abstracts_unavailable 30
fulltext_assessed 550
exclude_reason "No ground truth" 140
exclude_reason "Not deep learning" 95
exclude_reason "Review article" 60
studies_included 255