prEN ISO 24051-2

• Abstract:

  This document specifies requirements and gives recommendations for the digitalization of slide-mounted, stained sections, processing of digital whole slide images, and artificial intelligence (AI) based image analysis to support anatomical pathology examination. This document focuses on sections from formalin-fixed paraffin-embedded (FFPE) tissues (derived e.g. from surgical resection, biopsy or autopsy). It is also applicable to paraffin-embedded tissue fixed by fixatives other than formalin. ; This document is applicable to in vitro diagnostic examinations using digital pathology and AI-based image analysis performed by medical laboratories, in particular but not limited to anatomical pathology laboratories. It is also intended to be used by health institutions, in vitro diagnostics developers and manufacturers, and regulatory authorities. ; NOTE International, national or regional regulations or requirements can also apply to specific topics covered in this document.

• Status: Not Published
• Publication Date: 13-Apr-2028
• ICS: 11.100.01 - Laboratory medicine in general
• Technical Committee: CEN/TC 140 - In vitro diagnostic systems
• Drafting Committee: CEN/TC 140 - In vitro diagnostic systems
• Current Stage: 4020 - Submission to enquiry - Enquiry
• Start Date: 22-Oct-2026
• Due Date: 16-Sep-2027
• Completion Date: 22-Oct-2026
• Ref Project: oSIST prEN ISO 24051-2:2026 - Medical laboratories - Part 2: Digital pathology and artificial intelligence (AI)-based image analysis (ISO/DIS 24051-2:2026)

prEN ISO 24051-2 is a draft standard for medical laboratories on digital pathology and artificial intelligence (AI)-based image analysis. It specifies requirements and recommendations for digitizing slide-mounted, stained tissue sections, processing whole slide images, and using AI-based image analysis to support anatomical pathology examination.

What does prEN ISO 24051-2 specify?

prEN ISO 24051-2 specifies the workflow for digital pathology from slide preparation through scanning, viewing, data handling, and AI-based image analysis. It is written for sections from formalin-fixed paraffin-embedded (FFPE) tissues, and it also applies to paraffin-embedded tissue fixed by other fixatives.

The document applies to in vitro diagnostic examinations carried out by medical laboratories, especially anatomical pathology laboratories. It is also intended for health institutions, in vitro diagnostics developers and manufacturers, and regulatory authorities.

The structure is straightforward: Clause 1 gives the scope, Clause 2 lists the normative reference, Clause 3 defines terms, Clauses 4 to 9 cover the workflow and controls, and Annex A lists software-related standards.

The introduction says the standard focuses on the digital pathology workflow and AI-based image analysis, while other topics are handled elsewhere.

Annex	What it covers
Annex A	Software-related standards for software life cycle, usability, risk management, and further topics

What are the key requirements of prEN ISO 24051-2?

prEN ISO 24051-2 treats digital pathology as an end-to-end process. Clause 5 says the quality of the final result depends on the whole workflow, from primary sample collection and tissue processing to staining, scanning, viewing, and AI-based image analysis, and that manufacturer and user requirements shall be validated and verified.

Pre-examination and slide preparation

Clause 6 requires the pre-examination process to follow ISO 20166-4:2021 for FFPE tissue and other paraffin-embedded tissue fixed by non-formalin fixatives. The staining method has to suit the intended purpose. In practice, this means the laboratory must control fixation, section quality, and staining before scanning starts.

Scanner, workstation, and image acquisition

Clause 7.1 requires Installation Qualification and Operational Qualification before the scanner and associated workstation are put into use, and the user shall independently verify system performance. Clause 7.1.2 asks the manufacturer to document the scanner’s technical and performance characteristics, including the slide feeder, light source, optics, resolution, stage movement, sensor, image processing software, calibration, file format, computer environment, tissue coverage, stitching error, turnaround time, human factors studies, and analytical and clinical performance. This matters because the laboratory needs evidence that the system can generate usable diagnostic images under its intended conditions.

Clause 7.1.2 also requires the image file format to be DICOM or convertible to DICOM without loss of information. That supports exchange, storage, and later review of both image and metadata.

Slide quality control and scanning checks

Clause 7.2 requires procedures to produce slides suitable for scanning and makes pre-scanning quality control part of the process. The slide should be checked for label integrity, slide suitability, correct mounting, coverslip placement, cleanliness, staining quality, section quality, and the absence of contamination or missing tissue. Clause 7.2.3 adds corrective measures such as cleaning, recoverslipping, relabelling, or reprocessing when needed.

Clause 7.3 requires recalibration according to the manufacturer’s instructions, correct loading of slides, parameter selection according to intended use, and post-scanning quality control. The post-scan check includes incomplete coverage, blur, focus alerts, colour fidelity, and stitching problems. Clause 7.3.7 requires metadata for each slide, including patient identification, slide and block numbers, stain information, scanner type, scanning parameters, and QC parameters.

Image storage, viewing, and display quality

Clause 8 requires image storage and archiving in a system that preserves DICOM or DICOM-convertible files, metadata, and provenance. Any modification must be documented, and redundant storage is required when only digital data are retained. In practice, the laboratory needs traceability for image changes and retention aligned with local legal requirements.

Clause 8.3 and 8.3.4 require validated viewing tools and monitors. The viewer manufacturer has to specify the display parameters that affect image quality, and the user must verify suitability for intended use. Monitor verification must be repeated regularly, at least annually. The slide viewer shall support DICOM, preserve the scanner’s colour profile, and allow accurate panning, zooming, measurements, and annotations.

AI-based image analysis

Clause 9 says AI-based image analysis is optional but must be developed, verified, and validated in the context of the whole digital pathology workflow. The AI system description shall cover intended use, design choices, training and validation data, performance results, explainability, and post-marketing surveillance. Training, test, and validation datasets must be relevant, representative, and managed with clear separation and provenance, and the manufacturer shall address bias and data gaps.

Clause 9.3 requires the user interface to support human oversight, including the ability to detect anomalies and identify generative AI output. A human oversight supervisor shall be appointed for diagnostic use. Clause 9.4 requires documentation of the AI system version, manufacturer instructions, and the examination report information needed to show that AI was used. Clause 9.5 requires ongoing quality control, re-verification after changes, internal controls, incident reporting, and participation in external quality control where available.

What terms does prEN ISO 24051-2 define?

prEN ISO 24051-2 defines key terms used across the workflow so that laboratories, suppliers, and regulators use the same language.

• Artificial intelligence (AI) - the document uses AI for computational systems that perform tasks associated with human intelligence, such as learning, reasoning, and decision-making.
• Formalin-fixed paraffin-embedded tissue (FFPE tissue) - tissue that has been fixed in formalin, processed, and embedded in paraffin for sectioning.
• Whole slide image (WSI) - the digitized image of an entire stained tissue sample on a glass slide.
• Slide viewer - software and/or a device for reviewing, handling, and analyzing whole-slide images.
• Validation - confirmation, through objective evidence, that the requirements for a specific intended use have been fulfilled.
• Verification - confirmation, through objective evidence, that specified requirements have been fulfilled.
• Explainability - the property of an AI system to present important factors behind its results in a way humans can understand.
• Turnaround time - the elapsed time between the start of an examination and when the result becomes available.

Who uses prEN ISO 24051-2?

prEN ISO 24051-2 is intended for medical laboratories, especially anatomical pathology laboratories, that use digital pathology in in vitro diagnostic examinations. It is also relevant to histopathology technicians and pathologists because it covers slide handling, scanning, review, interpretation support, and reporting.

Health institutions use it when they set up digital pathology workflows or shared consultation services. IVD developers and manufacturers use it to design scanners, viewers, AI systems, and supporting software so they can document performance and intended use. Regulatory authorities can use it when assessing the evidence package for a digital pathology or AI-based diagnostic solution.

Which standards are used with prEN ISO 24051-2?

prEN ISO 24051-2 has one normative reference: ISO 15189, Medical laboratories - Requirements for quality and competence. That reference anchors the document in medical laboratory quality and competence requirements.

The text also names related standards and companion documents that contribute to implementation:

• ISO 20166-4:2021 for pre-examination processes for FFPE tissue.
• ISO/DIS 24051-1 for general principles for applying AI in medical laboratories.
• ISO 23494-1:2026 for provenance information on biological material and data.

Annex A lists software-related standards for software life cycle, usability, risk management, and related topics, including IEC 62304, ISO/IEC/IEEE 12207, IEC 82304-1, IEC 62366-1, ISO 14971:2019, ISO/TR 24971, IEC TR 80002-1, ISO 12052, ISO/IEC 5338, ISO/IEC 27034-1, ISO/TR 21332, ISO 27799, and several related guidance documents.

What does the prEN ISO 24051-2 document contain?

The document contains an overview of digital pathology and AI-based image analysis, followed by detailed requirements for the full workflow from slide preparation to report generation. It includes defined terms for pathology, imaging, AI, validation, verification, bias, provenance, and human oversight.

Clause 7 contains the main scanner and slide-scanning requirements, including technical characteristics, quality control before and after scanning, calibration, metadata, and performance checks. Clause 8 covers image storage, display, editing, sharing, and quality control for viewers and monitors.

Clause 9 contains the AI-specific material: dataset governance, training and validation data, human-machine interface, human oversight, documentation, user instructions, examination report content, record keeping, and quality control. Annex A then points readers to software-related standards that support implementation, especially for lifecycle, usability, and risk management.

SLOVENSKI STANDARD
01-december-2026
Medicinski laboratoriji - 2. del: Digitalna patologija in analiza slik na podlagi
umetne inteligence (UI) (ISO/DIS 24051-2:2026)
Medical laboratories - Part 2: Digital pathology and artificial intelligence (AI)-based image
analysis (ISO/DIS 24051-2:2026)
Medizinische Laboratorien - Teil 2: Digitale Pathologie und auf künstlicher Intelligenz (KI)
basierende Bildanalyse (ISO/DIS 24051-2:2026)
Laboratoires médicaux - Partie 2: Pathologie numérique et analyse d'images basée sur
l'intelligence artificielle (IA) (ISO/DIS 24051-2:2026)
Ta slovenski standard je istoveten z: prEN ISO 24051-2
ICS:
11.100.01 Laboratorijska medicina na Laboratory medicine in
splošno general
35.240.80 Uporabniške rešitve IT v IT applications in health care
zdravstveni tehniki technology
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

DRAFT
International
Standard
ISO/DIS 24051-2
ISO/TC 212
Medical laboratories —
Secretariat: ANSI
Part 2:
Voting begins on:
Digital pathology and artificial 2026-10-23
intelligence (AI)-based image
Voting terminates on:
2027-01-15
analysis
ICS: 35.240.80; 11.100.01
THIS DOCUMENT IS A DRAFT CIRCULATED
FOR COMMENTS AND APPROVAL. IT
IS THEREFORE SUBJECT TO CHANGE
AND MAY NOT BE REFERRED TO AS AN
INTERNATIONAL STANDARD UNTIL
PUBLISHED AS SUCH.
This document has not been edited by the ISO Central Secretariat.
IN ADDITION TO THEIR EVALUATION AS
BEING ACCEPTABLE FOR INDUSTRIAL,
TECHNOLOGICAL, COMMERCIAL AND
USER PURPOSES, DRAFT INTERNATIONAL
STANDARDS MAY ON OCCASION HAVE TO
ISO/CEN PARALLEL PROCESSING
BE CONSIDERED IN THE LIGHT OF THEIR
POTENTIAL TO BECOME STANDARDS TO
WHICH REFERENCE MAY BE MADE IN
NATIONAL REGULATIONS.
RECIPIENTS OF THIS DRAFT ARE INVITED
TO SUBMIT, WITH THEIR COMMENTS,
NOTIFICATION OF ANY RELEVANT PATENT
RIGHTS OF WHICH THEY ARE AWARE AND TO
PROVIDE SUPPORTING DOCUMENTATION.
Reference number
ISO/DIS 24051-2:2026(en)
ISO/DIS 24051-2:2026(en)
© ISO 2026
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
ii
ISO/DIS 24051-2:2026(en)
Contents Page
Foreword .v
Introduction .vi
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Overview on digital pathology and AI-based image analysis .10
5 General requirements . 10
6 Pre-examination process (according to ISO 20166-4:2021) .11
7 Digital image acquisition (slide scanning).11
7.1 General requirements for the scanner and workstation .11
7.1.1 General .11
7.1.2 Technical and performance characteristics. 12
7.1.3 Manufacturers’ instructions for users .14
7.2 Slide preparation requirements for digital image acquisition .14
7.2.1 General .14
7.2.2 Pre-scanning Quality Control - Evaluation of pre-examination artefacts .14
7.2.3 Measures to eliminate pre-examination artefacts . 15
7.3 Slide scanning requirements . 15
7.3.1 Calibration . . . 15
7.3.2 Loading of slides . 15
7.3.3 Selection of scanning mode/profile/parameters . 15
7.3.4 Start of scanning process.16
7.3.5 Scanning quality control & corrective adjustments during scanning .16
7.3.6 Post-scanning quality control of digital images .16
7.3.7 Metadata refer to (data) provenance.16
8 Image/data management and viewing .16
8.1 General .16
8.2 Image/data storing and archiving.16
8.3 Image/data viewing .17
8.3.1 Viewing process .17
8.3.2 Monitor parameters .17
8.3.3 Monitor validation and verification .18
8.3.4 Slide viewer .18
8.4 Image/data editing. 20
8.5 Image/data sharing . 20
8.6 Quality control . 20
8.6.1 General . 20
8.6.2 Pre-implementation.21
8.6.3 Implementation .21
8.6.4 Post-implementation .21
9 AI-based image analysis .22
9.1 General . 22
9.1.1 General . 22
9.1.2 AI system description . 22
9.2 Training, test and validation data . 22
9.3 Al-based tool control .24
9.3.1 Human-machine interface.24
9.3.2 Human Oversight .24
9.4 Documentation . 25
9.4.1 General . 25
9.4.2 Manufacturers’ instructions for users . 25

iii
ISO/DIS 24051-2:2026(en)
9.4.3 Al-based tool application(s) . 26
9.4.4 Examination report for clinicians . 26
9.4.5 Record keeping by the user.27
9.5 Quality control (by the user) .27
Annex A (informative) Software-related standards.29
Bibliography .30

iv
ISO/DIS 24051-2:2026(en)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out through
ISO technical committees. Each member body interested in a subject for which a technical committee
has been established has the right to be represented on that committee. International organizations,
governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely
with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of ISO document should be noted. This document was drafted in accordance with the editorial rules of the
ISO/IEC Directives, Part 2 (see www.iso.org/directives).
ISO draws attention to the possibility that the implementation of this document may involve the use of (a)
patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent
rights in respect thereof. As of the date of publication of this document, ISO [had/had not] received notice of
(a) patent(s) which may be required to implement this document. However, implementers are cautioned that
this may not represent the latest information, which may be obtained from the patent database available at
www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO's adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 212, Clinical laboratory testing and in vitro
diagnostic test systems.
A list of all parts in the ISO 24051 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.

v
ISO/DIS 24051-2:2026(en)
Introduction
Technological advances and the increased focus on precision medicine demanding precision diagnostics
created a need for application of digital pathology and artificial intelligence (AI)-based image analysis.
Digital pathology using high-resolution digitized (whole slide) images of slide-mounted, stained tissue
sections from e.g. surgical resection, biopsy or autopsy specimens, gained increasing relevance in the
context of medical diagnostics and precision medicine. The stained, slide-mounted tissue sections can result
from various in situ detection techniques, such as classical histological staining, (e.g. Hematoxylin & Eosin
staining), histochemical techniques, immunohistochemical staining, and hybridization-based techniques).
The increasing application of digital pathology was enabled by the availability of slide scanners generating
digital images with high resolution and information content comparable to classical light microscopes. The
generation and processing of these digital (whole slide) images required the modification of diagnostic
workflows and created the need for solutions (e.g. slide viewers/image viewers) to properly manage
large image files and to present them via monitors (e.g. as part of a workstation) to the pathologists for
[1] [2] [3] [4]
histopathologic diagnosis making  .
The use of AI-based image analysis solutions allows computational analysis of image features and their
quantitation. These features could be for example the detection and quantification of mitotic cells, of cells
stained with a certain antibody, or of complex features such as tumour and stroma (e.g. for quantification of
[5] [6]
tumour cell content) or of specific tumour features, such as Gleason grade, or tumour-adipose feature
[7] [8]
. The result of the AI-based image analysis is typically presented to the pathologist as decision support
for diagnosis making.
Digital pathology and AI-based image analysis can be applied in disease detection, diagnosis and prediction
of prognosis. Furthermore, it supports the identification of the optimal treatment regimen based on patient
[1] [2] [9] [10] [11] [12]
profiles as well as the discovery of novel biomarkers and drug targets   . The application of
AI-based applications including machine learning (ML) and deep learning methods in context of pathology
are often referred to as computational pathology.
The global digital pathology market was valued at USD 736 million in 2021 and is expected to grow at a
Compound Annual Growth Rate (CAGR) of 13.2% for the period of 2021-2026. This is driven by rising
incidence of cancer, global shortage of pathologists fostering the adoption of digital pathology, telepathology,
and growing applications of digital pathology in drug development, complementary diagnostics and
companion diagnostics. The immunohistochemistry sector (using stained tissue sections) is estimated to
register the highest growth rate in the market sector for diagnostic application; the highest growth rate by
end user is expected in the pharmaceutical and biopharmaceutical companies segment due to rising R&D
[13] [14] [15]
investments  .
Typically, digital pathology and AI-based image analysis for anatomical pathology examinations requires
the following:
— Workflow adaptations for digital pathology and AI-based image analysis comprising the pre-examination
workflow to generate slide-mounted sections from paraffin-embedded tissue (as defined in ISO 20166-4
[16]
), the in situ detection applied to stain these slides (not covered in this standard), the preparation (e.g.
cleaning) of slides for digitization, the acquisition of digital images (i.e. scanning), the storing, viewing,
editing and sharing of images, the selection and application of AI-based image analysis solutions, and
the evaluation and interpretation of the analysis result by the pathologist for anatomical pathology
examinations.
— Medical devices for digital pathology that have been placed on the market typically comprise an image
acquisition device (i.e. slide scanner) and a workstation (comprising hard- and software components)
for slide viewing and pathological examination in combination referred to as whole slide imaging (WSI)
system. This requires data storage and management solutions and may be used with and without
application of AI-based image analysis in the context of computational pathology solutions. The US Food
and Drug Administration (FDA) has published a guidance document and approved the first registrations
[11] [17]
of WSI equipment as medical devices in 2017 .
These aspects will be covered within this document.

vi
ISO/DIS 24051-2:2026(en)
The following will not be covered within this document:
— the in situ detection techniques;
— information on technical security and privacy requirements;
— general software-related issues (e.g. software lifecycle management, cybersecurity, interoperability
with information management systems);
— diagnosis making by pathologists.
Some of these issues are addressed in other standards (for overview see Annex A).
Digital pathology together with AI-based analysis of digital images poses new requirements on the
standardization of the entire workflow to obtain high-quality (whole slide) images and reproducible,
qualitative and quantitative analysis results. Factors which can impact the quality of the image and the
quality of the computational analysis result include, for example:
— pre-examination (pre-analytical) and examination (analytical) factors associated with the collection
and processing of tissue specimens and production of stained slides (e.g. tissue fixation and processing,
embedding, section quality [e.g. uniformity of section thickness], intactness, size, material and thickness
of the slide, staining intensity and homogeneity, placing of the coverslip, air bubbles, dust, or markings
[18]
on the slide ;
— factors such as errors occurring during scanning e.g. bar code detection failure, focus failure, incomplete
[18]
slide scanning (no detection of tissue or missing of tissue parts), improper stitching, line artefacts ,
or variable colour calibration and different settings of brightness, intensity disparity, and boundary
[19]
intensity during scanning ;
[19]
— factors associated with visualization of digitized images on monitors (e.g. colour calibration ;
— factors associated with the AI-based analysis (e.g. wrong detection of features or “Clever Hans effect”,
[20]
performance of an AI algorithm not demonstrated for features present on a whole slide image) ; and
— performance drift, which could arise due to minor differences in slide preparation protocols between
sites or due to changes in preparation materials as they may change incrementally over time.
Therefore, a standardization of the entire digital pathology and AI-based image analysis workflow, and
requirements and recommendations for the respective devices including the performance verification and
validation are needed to secure the proper quality of the image and the AI-based analysis result to support
anatomical pathology examinations.
This standard provides guidance for the use of digital pathology and AI-based image analysis and
[21]
complements ISO/DIS 24051-1 .
In this document, the following verbal forms are used:
— "shall" indicates a requirement;
— "should" indicates a recommendation;
— "may" indicates a permission;
— "can" indicates a possibility or a capability.

vii
DRAFT International Standard ISO/DIS 24051-2:2026(en)
Medical laboratories —
Part 2:
Digital pathology and artificial intelligence (AI)-based image
analysis
1 Scope
This document specifies requirements and gives recommendations for the digitalization of slide-mounted,
stained sections, processing of digital whole slide images, and artificial intelligence (AI) based image
analysis. This document focuses on sections from formalin-fixed paraffin-embedded (FFPE) tissues (derived
e.g. from surgical resection, biopsy or autopsy). It is also applicable to paraffin-embedded tissue fixed by
fixatives other than formalin.
This document is applicable to in vitro diagnostic examinations using digital pathology and AI-based
image analysis performed by medical laboratories, in particular but not limited to anatomical pathology
laboratories. It is also intended to be used by health institutions, in vitro diagnostics developers and
manufacturers, and regulatory authorities.
NOTE International, national or regional regulations or requirements can also apply to specific topics covered in
this document.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements of this document. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
ISO 15189, Medical laboratories — Requirements for quality and competence
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 15189 as well as the following
apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— IEC Electropedia: available at http:// www .electropedia .org/
— ISO Online browsing platform: available at http:// www .iso .org/ obp
3.1
anatomic pathology
branch of laboratory medicine that examines tissue samples
[22]
[SOURCE: ISO/TS 23824:2024 , 3.1]

ISO/DIS 24051-2:2026(en)
3.2
artificial intelligence
AI, noun
research and development of mechanisms and applications of AI systems
Note 1 to entry: A definition that better explains how AI differs from other laboratory software is that AI is “the
capability of computational systems to perform tasks typically associated with human intelligence, such as learning,
reasoning, problem-solving, perception, and decision-making.”
Note 2 to entry: See Clause 4 for a description of the characteristics of AI that differ from other systems.
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.1.3, modified - Notes added]
3.3
AI system, noun
artificial intelligence system
engineered system that generates outputs such as content, forecasts, recommendations or decisions for a
given set of human-defined objectives
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.1.4]
3.4
automatic
automation, noun
automated, noun
pertaining to a process or system that, under specified conditions, functions without human intervention
[24]
[SOURCE: ISO/IEC 2382:2015 , 2121282, modified — In the definition, “a process or equipment” has been
replaced by “a process or system” and preferred terms of “automated” and “automation” are added; Notes to
entry 1 and 2 have been removed.]
3.5
bias
systematic difference in treatment of certain objects, people or groups in comparison to others
Note 1 to entry: Treatment is any kind of action, including perception, observation, representation, prediction or
decision.
[25]
[SOURCE: ISO/IEC TR 24027:2021 , 3.2.2]
3.6
control
purposeful action on or in a process to meet specified objectives
[26]
[SOURCE: IEC 61800-7-1:2015 , 3.2.6]
3.7
controllability
controllable, noun
property of an AI system (3.3) that allows a human or another external agent to intervene in the system’s
functioning
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.5.6]
3.8
dataset(s)
collection of data with a shared format
EXAMPLE 1 Whole slide images of prostate tissue, each one labelled as being benign or malignant.
EXAMPLE 2 A collection of blood examination results from 10,000 patients with recorded clinical outcomes in a
structured data format.
EXAMPLE 3 Numerical results for 1000 patients from a laboratory instrument.

ISO/DIS 24051-2:2026(en)
Note 1 to entry: Datasets can be used for training (3.49), validating or verifying the performance of an AI model (3.31).
3.9
deep learning
approach to creating rich hierarchical representations through the training (3.49) of neural networks with
more hidden layers
Note 1 to entry: Deep learning uses multi-layered networks of simple computing units (or “neurons”). In these neural
networks each unit combines a set of input values to produce an output value, which in turn is passed on to other
neurons downstream.
[27]
[SOURCE: ISO/IEC TR 29119-11:2020 , 3.1.26, modified — The words “one or” have been removed from
“with one or more hidden layers”.]
3.10
deviation
departure from an approved instruction, procedure and/or method or established standard
[28]
[SOURCE: ISO 15378:2017 , 3.7.5, modified — “Approved (3.7.1) standard operating procedure (SOP)
(3.7.10)” replaced by “instruction, procedure and/or method”.]
3.11
diagnosis
identification of a health or disease state from its signs and symptoms, where the diagnostic process can
involve examination (3.12) and tests for classification of an individual’s condition into separate and distinct
categories or subclasses that allow medical decisions about treatment and prognosis to be made
[16]
[SOURCE: ISO 20166-4 :2021, 3.12]
3.12
examination
set of operations having the objective of determining the numerical value, text value or characteristics of a
property
Note 1 to entry: An examination (3.12) may be the total of a number of activities, observations or measurements
required to determine a value or characteristic.
Note 2 to entry: Laboratory examinations that determine a numerical value of a property are called "quantitative
examinations"; those that determine the characteristics of a property are called "qualitative examinations".
Note 3 to entry: Laboratory examinations are also called "assays" or "tests".
[SOURCE: ISO 15189:2022, 3.8]
3.13
explainability
AI explainability, noun
property of an AI system (3.3) to express important factors influencing the AI system (3.3) results in a way
that humans can understand
Note 1 to entry: It is intended to answer the question “Why?” without attempting to argue that the course of action
that was taken was necessarily optimal.
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.5.7]
3.14
formalin
saturated aqueous formaldehyde solution which at 100 % contains 37 % formaldehyde by mass
(corresponding to 40 % by volume)
[16]
[SOURCE: ISO 20166-4 :2021, 3.18]

ISO/DIS 24051-2:2026(en)
3.15
formalin fixation
treatment of a sample (3.41) with standard buffered formalin solution (3.45) for stabilization
[16]
[SOURCE: ISO 20166-4 :2021, 3.19]
3.16
formalin-fixed, paraffin-embedded tissue
FFPE tissue, noun
tissue specimens/samples having undergone fixation in formalin (3.14), tissue processing and paraffin
embedding (3.34) in a tissue cassette
Note 1 to entry: Tissue processing is the process during the generation of FFPE tissue (3.16) consisting of the four
consecutive steps tissue fixation, dehydration, clearing, and impregnation with paraffin (3.22).
[16]
[SOURCE: ISO 20166-4 :2021, 3.17]
3.17
grossing
gross examination, noun
inspection of pathology specimens with the bare eye to obtain diagnostic information, while being processed
for further microscopic examination (3.12)
[16]
[SOURCE: ISO 20166-4 :2021, 3.21]
3.18
histological staining
in situ detection (3.24) technique undertaken to prepare tissue sections by using histological stains (e.g.
Haematoxylin-Eosin (HE), Chromotrope Aniline Blue (CAB)) to highlight features of the tissue section (3.48)
and enhance tissue section (3.48) contrast
[16]
[SOURCE: ISO 20166-4 :2021, 3.23]
3.19
histopathology
anatomic pathology, noun
branch of medicine that evaluates tissues and cellular samples under the microscope
Note 1 to entry: This branch of medicine is also referred to as histopathology (3.19) or cellular pathology.
[22]
[SOURCE: ISO/TS 23824:2024 , 3.1, modified — “Under the microscope” has been added to the definition
and “histopathology” has been added as preferred term.]
3.20
image
〈digital〉 graphical content intended to be presented visually
Note 1 to entry: This includes graphics that are encoded in any electronic format, including, but not limited to, formats
that are comprised of individual pixels (e.g. those produced by paint programs or by photographic means) and formats
that comprised of formulas (e.g. those produced as scalable vector drawings).
[29]
[SOURCE: ISO/IEC 20071-11:2019 , 3.2.1]
3.21
image analysis
examination (3.12) of a digital image (3.20) for the purpose of making a diagnosis (3.11)
Note 1 to entry: Image analysis can be performed by a pathologist or other qualified person or by using an AI system
(3.3).
ISO/DIS 24051-2:2026(en)
3.22
impregnation
impregnation with paraffin, noun
process step in tissue processing for replacement of clearing agent and infiltration of tissue with molten
paraffin (3.33)
[16]
[SOURCE: ISO 20166-4 :2021, 3.26]
3.23
input data
data provided to or directly acquired by an AI system (3.3) on the basis of which the system produces an
output or data for which an AI system (3.3) calculates a predicted output or inference
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.2.9]
3.24
in situ detection
technique that allows for precise localization of a specific biomolecule within a slide-mounted section
EXAMPLE In situ detection techniques used in context of glass-mounted slides are immunohistochemistry for
protein detection, in situ hybridization for nucleic acid detection, histological staining (3.18), histochemical techniques,
spatial transcriptomics, MALDI imaging mass spectrometry, imaging mass cytometry, in situ Raman spectroscopy.
[16]
[SOURCE: ISO 20166-4 :2021, 3.27 - Examples modified.]
3.25
in situ detection system
detection system, noun
set of reagents used to visualize the binding of molecules to their target in situ within a tissue section (3.48)
Note 1 to entry: Examples of molecules are antibody, affinity binder, and nucleic acid.
Note 2 to entry: Most common for in situ detection are enzyme- and fluorophore-based detection systems.
[16]
[SOURCE: ISO 20166-4 :2021, 3.28]
3.26
installation qualification
establishment by objective evidence that all key aspects of the process equipment and ancillary system
installation adhere to the manufacturer's (3.30) approved specification and that the recommendations of the
supplier of the equipment are suitably considered
[30]
[SOURCE: ISO/ASTM TS 52930:2021 , 3.1]
3.27
interpretability
causability, which defines the quality of explainability (3.13) enabling the user to interpret the result
3.28
machine learning
ML, noun
process using computational techniques to enable systems to learn from data or experience
[27]
[SOURCE: ISO/IEC TR 29119-11:2020 , 3.1.43]
3.29
machine learning algorithm
algorithm to determine parameters of a machine learning (3.28) model (3.31) from data according to given
criteria
EXAMPLE Consider solving a univariate linear function y = θ0 + θ1x where y is an output or result, x is an input,
θ0 is an intercept (the value of y where x=0) and θ1 is a weight. In machine learning (3.28), the process of determining
the intercept and weights for a linear function is known as linear regression.

ISO/DIS 24051-2:2026(en)
3.30
manufacturer
natural or legal person with responsibility for design and/or manufacture of a medical device with the
intention of making the medical device available for use, under his name; whether or not such a medical
device is designed and/or manufactured by that person himself or on his behalf by another person(s)
Note 1 to entry: This “natural or legal person” has ultimate legal responsibility for ensuring compliance with all
applicable regulatory requirements for the medical devices in the countries or jurisdictions where it is intended to
be made available or sold, unless this responsibility is specifically imposed on another person by the Regulatory
Authority (RA) within that jurisdiction.
Note 2 to entry: The manufacturer’s responsibilities are described in other GHTF guidance documents. These
responsibilities include meeting both pre-market requirements and post-market requirements, such as adverse event
reporting and notification of corrective actions.
Note 3 to entry: “Design and/or manufacture”, as referred to in the above definition, may include specification
development, production, fabrication, assembly, processing, packaging, repackaging, labelling, relabelling,
sterilization, installation, or remanufacturing of a medical device; or putting a collection of devices, and possibly other
products, together for a medical purpose.
Note 4 to entry: Any person who assembles or adapts a medical device that has already been supplied by another
person for an individual patient, in accordance with the instructions for use, is not the manufacturer, provided the
assembly or adaptation does not change the intended use of the medical device.
Note 5 to entry: Any person who changes the intended use of, or modifies, a medical device without acting on behalf
of the original manufacturer and who makes it available for use under his own name, should be considered the
manufacturer of the modified medical device.
Note 6 to entry: An authorized representative, distributor or importer who only adds its own address and contact
details to the medical device or the packaging, without covering or changing the existing labelling, is not considered a
manufacturer.
Note 7 to entry: To the extent that an accessory is subject to the regulatory requirements of a medical device, the
person responsible for the design and/or manufacture of that accessory is considered to be a manufacturer.
[31]
[SOURCE: ISO 13485:2016 , 3.10]
3.31
model
physical, mathematical or otherwise logical representation of a system, entity, phenomenon, process or data
[32]
[SOURCE: ISO/IEC 18023-1:2006 , 3.1.11 - modified]
3.32
operational qualification
establishment by objective evidence process control limits and action levels which result in product that
meets all predetermined requirements
[30]
[SOURCE: ISO/ASTM TS 52930:2021 , 3.2]
3.33
paraffin
paraffin wax, noun
product obtained from distillates, consisting essentially of a mixture of saturated hydrocarbons, solid at
room temperature used for paraffin embedding (3.34) of formalin-fixed tissue specimens /samples
[16]
[SOURCE: ISO 20166-4 :2021, 3.32]

ISO/DIS 24051-2:2026(en)
3.34
paraffin embedding
embedding in paraffin, noun
process following tissue processing in which a tissue sample (3.41), is placed in melted paraffin (3.33) to
achieve a hard surrounding matrix so that thin microscopic sections can be cut
[16]
[SOURCE: ISO 20166-4 :2021, 3.33]
3.35
performance of an AI system
ability of an AI system (3.3) to achieve its intended use as claimed by the manufacturer (3.30)
Note 1 to entry: The definition has been adapted from the EU AI Act, Article 3 (from 2 February 2025).
3.36
personal data, noun
personally identifiable information
information that can be used in a given context to identify, contact, or locate a single person, or to identify an
individual in context
[SOURCE: ISO 19414:2020, 3.1, modified — “Personal data” has been added as a preferred term.]
3.37
pre-examination process
pre-examination phase, noun
process that starts, in chronological order, from the user’s request and includes the examination (3.12)
request, preparation and identification of the patient, collection of the specimens, transportation to and
within the medical laboratory, evaluation, tissue processing, paraffin embedding (3.34), FFPE block storage
(3.46) and retrieval, sectioning (including mounting onto glass slides and drying), storage (3.46) of unstained
slide-mounted tissue sections (if applicable), and pre-treatment steps for in situ detection (3.24) techniques,
and ending when the examination (3.12) begins
Note 1 to entry: The pre-examination phase (3.37) for in situ detection (3.24) includes preparative processes, e.g.
antigen/epitope retrieval and pre-hybridization procedures, which influence the outcome of the intended examination
(3.12).
Note 2 to entry: Pre-examination processes are also called “pre-analytical workflow” or “pre-analytical phase”.
[16]
[SOURCE: ISO 20166-4 :2021, 3.35, modified - ”Clinician’s” has been replaced by “user’s” as preferred
term.]
3.38
primary sample
specimen, noun
discrete portion of a body fluid, breath, hair or tissue taken for examination (3.12), study or analysis of one
or more quantities or properties assumed to apply for the whole
Note 1 to entry: The International Medical Device Regulators Forum (IMDRF) uses the term specimen in its
harmonized guidance documents to mean a sample (3.41) of biological origin intended for examination (3.12) by a
medical laboratory.
[SOURCE: ISO 15189:2022, 3.25]
3.39
reliability
property of consistent intended behaviour and results
[33]
[SOURCE: ISO/IEC 27000:2018 , 2.55]

ISO/DIS 24051-2:2026(en)
3.40
risk
combination of the probability of occurrence of harm and the severity of that harm
[34]
[SOURCE: ISO 14971:2019 , 2.16]
3.41
sample
one or more parts taken from a primary sample (3.38)/specimen (3.38)
[SOURCE: ISO 15189:2022, 3.28, modified — Term “specimen” has been added.]
3.42
scanner
slide scanner, noun
device designed and used (in whole slide imaging ) to capture high-resolution digital images of entire
microscopic glass slides containing stained sections from biological materials
3.43
slide viewer
high-performance digital slide viewing software and/or device designed for digital pathology and whole-
slide image review, handling (e.g. zooming, rotating, annotating), and analysis
Note 1 to entry: Slide viewers are also called image viewers or WSI (3.56) viewers.
3.44
stability
ability of a sample (3.41) material, when stored under specified conditions, to maintain a stated property
value within specified limits for a specified period of time
Note 1 to entry: The analytes for the purpose of this document are biomolecules located in situ/in the tissue.
[16]
[SOURCE: ISO 20166-4 :2021, 3.44]
3.45
standard buffered formalin solution
neutral buffered formalin, noun
NBF, noun
10 % formalin (3.14) solution in water with a mass fraction of 3,7 % (corresponding to a volume fraction of 4
%) formaldehyde buffered to pH 6,8 to pH 7,2
Note 1 to entry: Standard buffered formalin (3.14) solutions often contain small amounts of methanol to inhibit
oxidation and polymerization of formaldehyde.
[16]
[SOURCE: ISO 20166-4 :2021, 3.45]
3.46
storage
maintenance of biological material or data under appropriate conditions to preserve its properties for
intended use
[16]
[SOURCE: ISO 20166-4 :2021, 3.46, modified — “Or data” and “to preserve its properties” have been
added.]
3.47
test data
data used to assess the performance of a final AI (3.2) model (3.31)
Note 1 to entry: Test data is distinct from training data (3.50).
[23]
[SOURCE: ISO/IEC 22989:2022 , 3.2.14, modified — The synonym “evaluatio
...