Cancer3.AI › AI in Oncology › AI Models › Paige Prostate Detect
Clinical product Closed / commercial FDA De Novo pathology

Paige Prostate Detect

The first AI software for pathology authorised by the US FDA (De Novo, September 2021): flags prostate biopsy whole-slide images likely to contain cancer so the pathologist reviews the suspicious region.

Docs { } JSON

At a glance

DeveloperPaige
Released2021-09-21
Licenceproprietary
AvailabilityClosed / commercial
KindClinical product
Regulatory statusFDA De Novo
Parametersnot disclosed by the manufacturer

What it does

A second-read aid on digitised H&E prostate needle biopsies: outputs a slide-level suspicion flag and highlights the area of concern; the diagnosis remains the pathologist's.

Notable uses in oncology

In published studies, pathologists working with the tool improved the sensitivity of cancer detection by roughly 3–16% depending on the study, with negative predictive values of about 96–100%. Larger prospective trials are still needed to confirm performance in everyday practice.

Tasks, data types and cancers

TaskDetectionClassification
Data typeHistopathology (WSI)
Clinical settingDiagnosis
CancerProstate
InputDigitised H&E whole-slide images of prostate needle biopsies, on validated scanners.
OutputFlagging of areas suspicious for cancer for the reviewing pathologist; not a standalone diagnosis.

Architecture

FamilyNot disclosed. Checked on 2026-09-06: paige.ai gives no network type, backbone, parameter count or training-data description. The public technical reference for this product is the FDA De Novo summary DEN200080, already linked on this card.
Parametersnot disclosed by the manufacturer

Training data

Not disclosed by the manufacturer. Checked on 2026-09-06: the Paige site (paige.ai) describes the Paige Prostate Suite functionally, as AI applications aiding detection and diagnosis of prostate cancer on H&E-stained whole-slide images of prostate needle biopsies, but names no training dataset, no number of whole-slide images or patients, no source institution and no architecture.

Evaluation

Benchmark / datasetMetricValueExternal validationSource
AHRQ/NCBI evidence review of the Paige Prostate Suite (NBK608438) — pathologists reading prostate needle-biopsy whole-slide images without and then with the tool sensitivity and specificity of the pathologist, unassisted vs assisted across the reviewed studies the reviewers report a sensitivity gain of 2.9% to 16% from unassisted to assisted reading, with assisted sensitivity between 90% and 96.9%; specificity stayed broadly unchanged — assisted specificity between 96.6% and 99.8%, the largest fall being 2.0% and one study reporting a rise of 0.7%. These are ranges across several studies, not the result of a single trial. yes Source

How to run

Not yet documented on this card.

Regulatory status and intended use

Regulatory statusFDA De Novo
Intended useAid to pathologists reviewing digitised prostate needle-biopsy H&E slides; not for standalone diagnosis.

FDA De Novo DEN200080 (granted 2021-09-21) for Paige Prostate as software to assist pathologists in detecting areas suspicious for cancer on prostate needle biopsy slides.

Source →

Regulatory status is quoted from the source linked above and can change. Research-use-only models must not be used for clinical decisions.

Limitations and bias

  • Cleared for prostate needle biopsies only, on validated scanners.
  • Closed weights; performance claims come from the regulatory submission.

Sources

  1. FDA De Novo decision summary DEN200080 (Paige Prostate)
  2. NCBI Bookshelf — evidence review of the Paige Prostate Suite (source of the sensitivity and specificity ranges quoted on this card)
  3. Pivotal validation study, cited after the review above: Raciti P, Sue J, Retamero JA et al. Clinical validation of artificial intelligence-augmented pathology diagnosis demonstrates significant gains in diagnostic accuracy in prostate cancer detection. Arch Pathol Lab Med 2023;147(10):1178-1185
  4. Independent (non-manufacturer) real-world validation, cited after the review above: da Silva LM et al. Independent real-world application of a clinical-grade automated prostate cancer detection system. J Pathol 2021;254(2):147-158

This page is educational — it is not medical advice and does not replace consultation with an oncologist. Diagnostic and treatment decisions are made solely by specialist physicians.