ISO/IEC 3532-1:2023

Information technology — Medical image-based modelling for 3D printing — Part 1: General requirements

This document specifies the requirements for medical image-based modelling for 3D printing for medical applications. It concerns accurate 3D data modelling in the medical field using medical image data generated from computed tomography (CT) devices. It also specifies the principal considerations for the general procedures of medical image-based modelling. It excludes soft tissue modelling from magnetic resonance image (MRI).

Technologies de l'information — Modélisation médicale à base d'images pour l'impression 3D — Partie 1: Exigences générales

General Information

• Status: Published
• Publication Date: 08-Jun-2023
• ICS: 25.030 - Additive manufacturing
• ICS: 35.240.80 - IT applications in health care technology
• Technical Committee: ISO/IEC JTC 1 - Information technology
• Drafting Committee: ISO/IEC JTC 1/WG 12 - 3D Printing and scanning
• Current Stage: 6060 - International Standard published
• Start Date: 09-Jun-2023
• Due Date: 24-Aug-2023
• Completion Date: 09-Jun-2023

Standards Content (Sample)

INTERNATIONAL ISO/IEC
STANDARD 3532-1
First edition
2023-06
Information technology — Medical
image-based modelling for 3D
printing —
Part 1:
General requirements
Technologies de l'information — Modélisation médicale à base
d'images pour l'impression 3D —
Partie 1: Exigences générales
Reference number
ISO/IEC 3532-1:2023(E)
© ISO/IEC 2023

---------------------- Page: 1 ----------------------
ISO/IEC 3532-1:2023(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO/IEC 2023
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/IEC 2023 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/IEC 3532-1:2023(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms, definitions and abbreviated terms . 1
3.1 Terms and definitions . 1
3.2 Abbreviated terms . 4
4 Overview of image processing for the medical industry . 5
4.1 Process flow . 5
4.1.1 3D printing process for medical applications . 5
4.1.2 Explanation of a typical use case (cranial implant case) . 5
5 General requirements . 6
6 Requirements of data processing .7
6.1 Medical image data flow . 7
6.2 Medical image acquisition/computed tomography scan . 8
6.3 Segmentation . 9
6.4 3D reconstruction and visualization . 11
6.5 Calibration and validation of 2D and 3D conversion .12
6.6 File format . 13
Annex A (informative) Reporting .14
Bibliography .15
iii
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 3 ----------------------
ISO/IEC 3532-1:2023(E)
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical
activity. ISO and IEC technical committees collaborate in fields of mutual interest. Other international
organizations, governmental and non-governmental, in liaison with ISO and IEC, also take part in the
work.
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 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 or
www.iec.ch/members_experts/refdocs).
Attention is drawn to the possibility that some of the elements of this document may be the subject
of patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent
rights. Details of any patent rights identified during the development of the document will be in the
Introduction and/or on the ISO list of patent declarations received (see www.iso.org/patents) or the IEC
list of patent declarations received (see https://patents.iec.ch).
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. In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology.
A list of all parts in the ISO/IEC 3532 series can be found on the ISO and IEC websites.
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 and
www.iec.ch/national-committees.
iv
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/IEC 3532-1:2023(E)
Introduction
This document was developed in response to the need for customization of 3D scanning and 3D printing
technology within the medical industry, which can be achieved by taking full advantage of information
and communication technology (ICT).
This document addresses the overview of medical image processing and requirements for image-
based modelling. 3D printing technology has caused a revolution in health care delivery. New classes
of medical devices embody the true meaning of personalized medicine. Medical device designers and
practitioners are able to practically and efficiently create devices that were very difficult or impossible
to create before. In addition to using 3D printing technology to create standard medical devices with
features like intricate lattice structures, clinicians and engineers work in conjunction to produce what
are known as patient-specific devices or patient-matched devices. These are medical devices designed
to fit a specific patient’s anatomy, typically using medical imaging from that patient. Anatomically
matched devices have very complex geometrical contours and shapes. Several challenges exist in the
design process between the input data and the final device design. Most of these steps definitely depend
on software-based management of medical images.
Overall, the world revenue from 3D printing technology in the healthcare industry is expected to grow
exponentially, yet very few guides exist for 3D printing for medical practice. Medical images from the
human body are different from solid objects due to the non-geometric nature of the human body. To
perform 3D printing for medical practice, an accurate and consistent approach for image processing and
data creation from medical images is needed. Standardization for 3D printing processes in medicine
is urgently required for education, diagnosis, neurosurgical treatment, developing simulation models,
medical equipment (including surgical guides) and surgical implantable devices in the clinical fields.
Regulatory bodies from several countries (US, Republic of Korea, etc.) have already published their
own guidelines for approval. However, those guidelines are not specifically designed for 3D printing
technology.
Applications of 3D printing in medicine are thriving, and include surgical simulation models, surgical
guides, educational models, surgical implants, etc. Those which are manufactured by 3D printing
technology require patient- and/or procedure-specific data (e.g. planned surgical technique and others)
and medical image data acquisition processing. Most of the processing of medical images for 3D printing
medical devices is software-based. In order to accurately and consistently visualize human body
anatomy, appropriate software-based modelling for 3D printing is needed. This document provides
requirements for software-based medical image processing for the purpose of producing 3D models for
3D printing. Valuable information related to optimized medical image data for additive manufacturing
can be found in ISO/ASTM TR 52916.
v
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO/IEC 3532-1:2023(E)
Information technology — Medical image-based modelling
for 3D printing —
Part 1:
General requirements
1 Scope
This document specifies the requirements for medical image-based modelling for 3D printing for
medical applications. It concerns accurate 3D data modelling in the medical field using medical image
data generated from computed tomography (CT) devices. It also specifies the principal considerations
for the general procedures of medical image-based modelling. It excludes soft tissue modelling from
magnetic resonance image (MRI).
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/IEC 2382, Information technology — Vocabulary
ISO/ASTM 52900, Additive manufacturing — General principles — Fundamentals and Vocabulary
3 Terms, definitions and abbreviated terms
For the purposes of this document, the terms and definitions given in ISO/IEC 2382, ISO/ASTM 52900
and the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1 Terms and definitions
3.1.1
image acquisition
scanning of the structure of interest using computed tomography (CT), magnetic resonance imaging or
other three-dimensional imaging technology
3.1.2
slice distance
slice spacing
distance between the centre of the slices, which is calculated by the difference in the slice locations of
two adjacent slices
3.1.3
hard tissue
tissue which is mineralized and has a firm intercellular matrix (such as bone, tooth enamel, dentin and
cementum)
1
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 6 ----------------------
ISO/IEC 3532-1:2023(E)
3.1.4
soft tissue
tissue that connects, supports or surrounds other structures and organs of the body, excluding hard
tissue (3.1.3)
3.1.5
solid organ
organ which has firm tissue consistency such as the heart, kidney, liver, lungs, pancreas, etc., excluding
hollow organs (such as the organs of the gastrointestinal tract) and tissue with liquid consistency (such
as blood)
3.1.6
pixel
picture element
smallest two-dimensional element of a display image that can be independently assigned attributes
such as color and intensity
[SOURCE: ISO/IEC 2382:2015, 2125999, modified — Notes to entry have been removed.]
3.1.7
voxel
volume element
smallest three-dimensional element in volume or volumetric (solid) modelling that can be independently
assigned attributes such as colour and intensity
[SOURCE: ISO/IEC 2382:2015, 2126000, modified — Notes to entry have been removed; "solid" has
been replaced by "volume or volumetric (solid)".]
3.1.8
vector data
vector image
vector model
digital description of 2D image or 3D model stored as a series of points and mathematical functions to
describe the geometric figure
[SOURCE: ISO 12651-1:2012, 4.139, modified — "image" has been replaced by "2D image or 3D model”.]
3.1.9
raster data
raster image
raster model
bitmap data
bitmap image
bitmap model
2D image or 3D model data formed by a set of picture elements (3.1.6) or volume elements (3.1.7)
arranged in a grid pattern
3.1.10
volume model
solid model
three-dimensional geometric model which deals with the solid characteristics of an object in order to
represent its internal structure as well as its external shapes
Note 1 to entry: See ISO/IEC 2382 for definitions of volume modelling and solid modelling.
Note 2 to entry: Volume model can be represented with raster model (3.1.9) or vector model (3.1.8).
2
 © ISO/IEC 2023 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/IEC 3532-1:2023(E)
3.1.11
surface model
boundary model
data set of a model which represents the surfaces of objects
Note 1 to entry: See ISO/IEC 2382 for definitions of surfacing and surface modelling.
3.1.12
facet model
faceted model
surface model (3.1.11) of which surfaces consist of group of polygons
Note 1 to entry: A triangle is widely used as a polygon.
3.1.13
segmentation
process of separating the objects of interest from their surroundings
Note 1 to entry: Segmentation can be applicable to 2D, 3D, raster or vector data (3.1.8).
3.1.14
3D visualization
presentation intended for human viewing of a scene on a flat display surface, using graphics techniques
to convey depth information and knowledge of the arrangement and shapes of the visualized scene in a
three-dimensional space
Note 1 to entry: The graphics techniques can include use of perspective, occlusion, stereoscopy, lighting and
environmental effects, and ability to navigate the viewpoint to alternate positions and orientations.
3.1.15
3D modelling
activity intended to create a digital representation of the form and arrangement of one or more 3D
objects in a three-dimensional space
Note 1 to entry: 3D models can contain geometric information such as mesh vertices, appearance, lighting, and
animation information. The created representation is a prerequisite to creating a 3D visualization (3.1.14) of the
modelled objects.
3.1.16
maximum intensity projection
MIP
scientific visualization method for 3D data that projects, in the visualization plane and with maximum
intensity, the voxels that fall in the way of parallel rays traced from the viewpoint to the plane of
projection
3.1.17
minimum intensity projection
MinIP
data visualization method that enables detection of low-density structures in a given volume
Note 1 to entry: The algorithm uses all the data in a volume of interest to generate a single two-dimensional
image. In other words, it consists of projecting the voxel with the lowest attenuation value on every view
throughout the volume onto a 2D image.
3
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 8 ----------------------
ISO/IEC 3532-1:2023(E)
3.1.18
Hounsfield value
Hounsfield unit
integer representing the intensity of the image at each image point [pixel (3.1.6)] which originates from
the x-ray scanning process and in turn represents the image intensity, which depends on the density of
the tissue at that location
Note 1 to entry: Hounsfield values rise monotonically with tissue density but are not linearly proportional to
density.
Note 2 to entry: The highest range of biological tissue Hounsfield values is for cortical bone, and they can go even
higher for image artefacts such as metallic implants, metallic sections of a hospital bed included in the image, etc.
3.1.19
multiplanar reformation
MPR
two-dimensional reformatted images that are reconstructed secondarily in arbitrary planes from the
stack of axial image data
Note 1 to entry: Multiplanar reformation (MPR) allows images to be created from the original axial plane in
either the coronal, sagittal or oblique plane.
3.1.20
volume rendering
set of techniques used to display a 2D projection of a 3D discretely sampled data set, typically a 3D
scalar field
3.2 Abbreviated terms
2D two-dimensional
3D three-dimensional
AM additive manufacturing
AMF additive manufacturing file format
ANN artificial neural network
CAD computer aided design
CT computed tomography
DICOM digital imaging and communications in medicine
HU Hounsfield unit
PACS picture archiving communication system
QC quality control
ROI region of interest
STL stereolithography
SVM support vector machine
4
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/IEC 3532-1:2023(E)
4 Overview of image processing for the medical industry
4.1 Process flow
4.1.1 3D printing process for medical applications
In general, the medical 3D printing processing flow can be divided into eight phases, as shown in
Figure 1.
NOTE Annex A contains a list of recommended items to be noted during the 3D printing process flow.
1) Image acquisition phase
In the image acquisition phase, medical images are acquired from medical imaging devices such as CT.
2) Segmentation phase
In the segmentation phase, the acquired medical images are segmented to fit the design purpose and
are processed to be divided (segmented) to extract a subset that would represent the part(s) of the
anatomy under consideration.
3) 3D modelling phase
In the 3D modelling phase, the segmented data representing the human tissue is converted
(reconstructed) into a 3D model optimized for 3D printing.
4) 3D printing phase
In the 3D printing phase, 3D printing is performed using the 3D model designed. For this phase 3D
model is processed for 3D printing by slicing, assigning build parameters, being oriented and placed
within the build space, and can have support structures generated.
5) Post-processing phase
In the post-processing phase, the 3D printed part is post-processed to become fit for actual medical use.
6) Quality control (QC) phase
In the QC phase, the 3D printed part is finally verified to meet all requirements (user/design/quality/
risk).
7) Clinical application and review phase
In the clinical application and review phase, the 3D printed part is reviewed as applicable to clinical
application by the healthcare practitioner.
8) Post-market phase
In the post-marketing stage, the 3D printed part is managed based on the post-sale market management
policy according to product life cycle issues such as tracking management/recall.
4.1.2 Explanation of a typical use case (cranial implant case)
Computed tomography (CT) is a common imaging modality for medical applications. For instance, for
patients with a skull defect visiting a neurosurgical clinic, CT has been known as the gold standard
for investigating bone-related problems. Figure 1 shows that the CT images are initially transferred to
the PACS server in DICOM file format. DICOM images have been used to reconstruct 3D image through
segmentation and 3D modelling by certain software. This 3D modelled image is transformed and
exported to design software as a stereolithography (STL) file. After completion and confirmation of
3D cranial implant by designing software, a metal AM machine builds this implant as designed. Post-
processing such as heat treatment, machining, cleaning and sanding is performed. Reverse engineering
5
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 10 ----------------------
ISO/IEC 3532-1:2023(E)
is performed to confirm the completeness of the implant before delivery by 3D scanning and matching
to the original digital blueprint. After QC, the implant is packed, sterilized and delivered. An operation
is performed to cover the defect with the 3D printed cranial implant. For this medical 3D printing
process, accuracy and reproducibility should be considered. The accuracy and reproducibility of the
parts (anatomical model, surgical guides, implant, etc.) from medical 3D printed parts are affected
by the sum of errors introduced in each step during data flow. These steps can be image acquisition,
segmentation and any subsequent post-processing of the segmented images. This document covers
processes 1, 2 and 3 as shown in Figure 1, ending with a 3D model of the relevant patient anatomy for
use in multiple other later processes. Activities related to items for processes 4 - 8 are addressed by
ISO/TC 261.
Figure 1 — Typical workflow of medical 3D printing (example: cranial implant case)
5 General requirements
To conform to this document, all of the following items shall be considered and relevant information
shall be documented.
— The medical image acquisition protocol by the CT scanner.
— The clinical purpose (bone/hard tissue) of image-based modelling.
— The segmentation method and associated parameters.
— The processes and parameters for 3D reconstruction.
Major parameters, settings and descriptions of methods used in the processes above shall be recorded.
6
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 11 ----------------------
ISO/IEC 3532-1:2023(E)
6 Requirements of data processing
6.1 Medical image data flow
There are usually two medical image data flows involved in data processing: example flow and direct
flow. MIP, MinIP, MPRs and volume rendering are used before transporting the medical image to the
PACS server. Typically, DICOM files are used to make 3D images. However, many PACS companies
provide plugged-in 3D visualization software to reform raw data to 3D images and transport 3D images
directly to the PACS server as captured images. These 3D visualizations on the PACS server are 2D
projections of a 3D object and are not suitable for 3D modelling. The 2D printers for films prints out 2D
images (X ray radiograph, CT, MRI, etc.) or 3D-modelled captured images. See Figure 2.
7
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 12 ----------------------
ISO/IEC 3532-1:2023(E)
Key
Example flow (dotted lines) Direct flow (solid lines)
1 scanned by device A scanned by device
2 convert to DICOM B 3D visualization
3 save to PACS server C save to PACS server
4 3D visualization D (transfer to STL)
5 transfer to STL or other file formats E directs to Figure 1
6 directs to Figure 1
Figure 2 — Data flow of medical images (Example)
6.2 Medical image acquisition/computed tomography scan
To make medical 3D models, sequential 2D images are necessary and should be acquired from sectional
images. Generally, 2D slice images are acquired from a CT scan of the patient's body at regular intervals
depending on the scanning needs. Each CT image set has its own strengths and weaknesses with
respect to the different objectives of the observation. Bones are typically clearly identified. Variabilities
8
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 13 ----------------------
ISO/IEC 3532-1:2023(E)
of output depend on factors such as spatial resolution/voxel size of the images, which in turn depends
on the x-ray dosage, the quality of the scanned images, operator capability, and low and high resolution
on 2D to 3D conversion algorithms. Features smaller than 0,3 mm cannot be printed successfully with
some printing processes if smaller features are needed. This will require special considerations for
process selection and post-processing operations. Care should be taken in choosing the appropriate
3D printing technology and the part manufacturers should be requested to consider the required
resolutions.
For medical image acquisition, the typical slice distance of less than 1 mm is sufficient and the following
points shall be addressed.
— Required accuracy and clinical purposes shall be compatible. The CT scanning protocol shall be
specified beforehand to achieve the required accuracy of the final models.
— The highest accuracy or resolution of CT scan is not always necessary.
— The time between the acquisition of the patient images and the initiation of image-based modelling
shall be minimized.
Other factors which can influence the quality of the final scanned images are as follows.
— Possible patient motion during the scanning process and its implications on the imaging accuracy.
Even breathing can cause errors in scanning in cardiovascular applications, for example.
— The use of contrast media during the scanning process to highlight blood or other liquids through
various tracers.
— Any digital filtering techniques applied in the data processing at the scanning stage to produce the
DICOM data used later for segmentation and reconstruction.
6.3 Segmentation
Image segmentation divides the image into meaningful regions. Segmentation usually extracts one or
more subsets of the data from the whole dataset, such that each subset would represent an anatomical
part, or tissue of the same characteristics [(e.g. bone having high density versus soft tissue having lower
density so a simple threshold can be established based on Hounsfield unit (HU)] etc. Segmentation in
medical imaging is generally considered a difficult problem, mainly because of the sheer size of the
datasets coupled with the complexity and variability of the anatomic organs.
The situation is worsened by the shortcomings of imaging modalities (such as sampling artifacts,
noise, low contrast, etc.) that can cause the boundaries of anatomical structures to be indistinct and
disconnected. The segmentation process becomes challenging in the absence of clear distinction in the
characteristics desired, such as density ranges overlapping (e.g. very soft bone indistinguishable from
calcified cartilage), or blood vessel walls not sufficiently distinguishable from surrounding muscle
tissue, etc. The challenge is greater when the target tissue is complex (intermingled or touching) in
its location (e.g. small diameter nerves around the orbit and vessels and nerves through skull base
foramens, anterior of a distal femur in a highly arthritic patient appearing to be totally connected to
their patella, or small complex branched blood vessels.).
Thus, the main challenge of segmentation algorithms is to accurately extract the boundary of the solid
organ or region of interest (ROI) and separate it from the rest of the dataset. The notion of boundary
starts in a 2D context of an image slice. Boundaries from all images (slices) can help build whole models
after segmentation.
For the segmentation of bony structures from CT scans the following relevant segmentation
technologies can be selected or combined:
— intensity-based segmentation: thresholding, edge-based (e.g. canny edge detector, sobel edge
detector), region-based (e.g. region growing, region splitting and merging), hybrid-based (e.g.
watershed transformation);
9
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 14 ----------------------
ISO/IEC 3532-1:2023(E)
— deformable model-based segmentation: active contour model, level-set method;
— atlas-based segmentation: active shape model, active appearance models, label fusion;
— machine learning-based segmentation: supervised (e.g. ANN, SVM), unsupervised (e.g. K-means
clustering, Fuzzy C-means algorithm).
Segmentation of bony structures from CT scanned image data has therefore benefited from multiple
software algorithmic techniques. Several software advancements have emerged stretching from purely
image intensity (thresholding) as described above, into more intelligent deformable model versions
where a shape is assumed to have been approximately recognized but adjusted based on the data.
Even more advanced techniques include the shaper perception to relay on multiple preconceived shape
models (from an atlas) and all the way to using t
...

FINAL
INTERNATIONAL ISO/IEC
DRAFT
STANDARD FDIS
3532-1
ISO/IEC JTC 1
Information technology — Medical
Secretariat: ANSI
image-based modelling for 3D
Voting begins on:
2023-02-22 printing —
Voting terminates on:
Part 1:
2023-04-19
General requirements
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 SUPPOR TING
DOCUMENTATION.
IN ADDITION TO THEIR EVALUATION AS
Reference number
BEING ACCEPTABLE FOR INDUSTRIAL, TECHNO-
ISO/IEC FDIS 3532-1:2023(E)
LOGICAL, COMMERCIAL AND USER PURPOSES,
DRAFT INTERNATIONAL STANDARDS MAY ON
OCCASION HAVE TO BE CONSIDERED IN THE
LIGHT OF THEIR POTENTIAL TO BECOME STAN-
DARDS TO WHICH REFERENCE MAY BE MADE IN
NATIONAL REGULATIONS. © ISO/IEC 2023

---------------------- Page: 1 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
FINAL
INTERNATIONAL ISO/IEC
DRAFT
STANDARD FDIS
3532-1
ISO/IEC JTC 1
Information technology — Medical
Secretariat: ANSI
image-based modelling for 3D
Voting begins on:
printing —
Voting terminates on:
Part 1:
General requirements
COPYRIGHT PROTECTED DOCUMENT
© ISO/IEC 2023
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.
RECIPIENTS OF THIS DRAFT ARE INVITED TO
ISO copyright office
SUBMIT, WITH THEIR COMMENTS, NOTIFICATION
OF ANY RELEVANT PATENT RIGHTS OF WHICH
CP 401 • Ch. de Blandonnet 8
THEY ARE AWARE AND TO PROVIDE SUPPOR TING
CH-1214 Vernier, Geneva
DOCUMENTATION.
Phone: [phone]
IN ADDITION TO THEIR EVALUATION AS
Reference number
Email: [e-mail]
BEING ACCEPTABLE FOR INDUSTRIAL, TECHNO­
ISO/IEC FDIS 3532­1:2023(E)
Website: www.iso.org
LOGICAL, COMMERCIAL AND USER PURPOSES,
DRAFT INTERNATIONAL STANDARDS MAY ON
Published in Switzerland
OCCASION HAVE TO BE CONSIDERED IN THE
LIGHT OF THEIR POTENTIAL TO BECOME STAN­
DARDS TO WHICH REFERENCE MAY BE MADE IN
ii
  © ISO/IEC 2023 – All rights reserved
NATIONAL REGULATIONS. © ISO/IEC 2023

---------------------- Page: 2 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms, definitions and abbreviated terms . 1
3.1 Terms and definitions . 1
3.2 Abbreviated terms . 4
4 Overview of image processing for the medical industry . 5
4.1 Process flow . 5
4.1.1 3D printing process for medical applications . 5
4.1.2 Explanation of a typical use case (cranial implant case) . 5
5 General requirements . 6
6 Requirements of data processing .7
6.1 Medical image data flow . 7
6.2 Medical image acquisition/computed tomography scan . 8
6.3 Segmentation . 9
6.4 3D reconstruction and visualization . 11
6.5 Calibration and validation of 2D and 3D conversion .12
6.6 File format . 13
Annex A (informative) Reporting .14
Bibliography .15
iii
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 3 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical
activity. ISO and IEC technical committees collaborate in fields of mutual interest. Other international
organizations, governmental and non­governmental, in liaison with ISO and IEC, also take part in the
work.
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 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 or
www.iec.ch/members_experts/refdocs).
Attention is drawn to the possibility that some of the elements of this document may be the subject
of patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent
rights. Details of any patent rights identified during the development of the document will be in the
Introduction and/or on the ISO list of patent declarations received (see www.iso.org/patents) or the IEC
list of patent declarations received (see https://patents.iec.ch).
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. In the IEC, see www.iec.ch/understanding­standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology.
A list of all parts in the ISO/IEC 3532 series can be found on the ISO and IEC websites.
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 and
www.iec.ch/national­committees.
iv
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
Introduction
This document was developed in response to the need for customization of 3D scanning and 3D printing
technology within the medical industry, which can be achieved by taking full advantage of information
and communication technology (ICT).
This document addresses the overview of medical image processing and requirements for image-
based modelling. 3D printing technology has caused a revolution in health care delivery. New classes
of medical devices embody the true meaning of personalized medicine. Medical device designers and
practitioners are able to practically and efficiently create devices that were very difficult or impossible
to create before. In addition to using 3D printing technology to create standard medical devices with
features like intricate lattice structures, clinicians and engineers work in conjunction to produce what
are known as patient-specific devices or patient-matched devices. These are medical devices designed
to fit a specific patient’s anatomy, typically using medical imaging from that patient. Anatomically
matched devices have very complex geometrical contours and shapes. Several challenges exist in the
design process between the input data and the final device design. Most of these steps definitely depend
on software­based management of medical images.
Overall, the world revenue from 3D printing technology in the healthcare industry is expected to grow
exponentially, yet very few guides exist for 3D printing for medical practice. Medical images from the
human body are different from solid objects due to the non-geometric nature of the human body. To
perform 3D printing for medical practice, an accurate and consistent approach for image processing and
data creation from medical images is needed. Standardization for 3D printing processes in medicine
is urgently required for education, diagnosis, neurosurgical treatment, developing simulation models,
medical equipment (including surgical guides) and surgical implantable devices in the clinical fields.
Regulatory bodies from several countries (US, Repulic of Korea, etc.) have already published their
own guidelines for approval. However, those guidelines are not specifically designed for 3D printing
technology.
Applications of 3D printing in medicine are booming, such as surgical simulation models, surgical
guides, educational models, surgical implants, etc. Those which are manufactured by 3D printing
technology require patient- and/or procedure-specific data (e.g. planned surgical technique and others)
and medical image data acquisition processing. Most of the processing of medical images for 3D printing
medical devices is software-based. In order to accurately and consistently visualize human body
anatomy, appropriate software-based modelling for 3D printing is needed. This document provides
requirements of software-based medical image processing for the purpose of producing 3D models for
3D printing. Valuable information related to optimized medical image data for additive manufacturing
can be found in ISO/ASTM TR 52916.
v
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 5 ----------------------
FINAL DRAFT INTERNATIONAL STANDARD ISO/IEC FDIS 3532-1:2023(E)
Information technology — Medical image-based modelling
for 3D printing —
Part 1:
General requirements
1 Scope
This document specifies the requirements for medical image-based modelling for 3D printing for
medical applications. It concerns accurate 3D data modelling in the medical field using medical image
data generated from computed tomography (CT) devices. It also specifies the principal considerations
for the general procedures of medical image-based modelling. It excludes soft tissue modelling from
magnetic resonance image (MRI).
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/IEC 2382, Information technology — Vocabulary
ISO/ASTM 52900, Additive manufacturing — General principles — Fundamentals and Vocabulary
3 Terms, definitions and abbreviated terms
For the purposes of this document, the terms and definitions given in ISO/IEC 2382, ISO/ASTM 52900
and the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1 Terms and definitions
3.1.1
image acquisition
scanning of the structure of interest using computed tomography (CT), magnetic resonance imaging or
other three-dimensional imaging technology
3.1.2
slice distance
slice spacing
distance between the centre of the slices, which is calculated by the difference in the slice locations of
two adjacent slices
3.1.3
hard tissue
tissue which is mineralized and has a firm intercellular matrix (such as bone, tooth enamel, dentin and
cementum)
1
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 6 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
3.1.4
soft tissue
tissue that connects, supports or surrounds other structures and organs of the body, excluding hard
tissue (3.1.3)
3.1.5
solid organ
organ which has firm tissue consistency such as the heart, kidney, liver, lungs, pancreas, etc., excluding
hollow organs (such as the organs of the gastrointestinal tract) and tissue with liquid consistency (such
as blood)
3.1.6
pixel
picture element
smallest two-dimensional element of a display image that can be independently assigned attributes
such as color and intensity
[SOURCE: ISO/IEC 2382:2015, 2125999, modified — Notes to entry have been removed.]
3.1.7
voxel
volume element
smallest three-dimensional element in volume or volumetric (solid) modeling that can be independently
assigned attributes such as colour and intensity
[SOURCE: ISO/IEC 2382:2015, 2126000, modified — Notes to entry have been removed; "solid" has
been replaced by "volume or volumetric (solid)".]
3.1.8
vector data
vector image
vector model
digital description of 2D image or 3D model stored as a series of points and mathematical functions to
describe the geometric figure
[SOURCE: ISO 12651-1:2012, 4.139, modified — "image" has been replaced by "2D image or 3D model”.]
3.1.9
raster data
raster image
raster model
bitmap data
bitmap image
bitmap model
2D image or 3D model data formed by a set of picture elements (3.1.6) or volume elements (3.1.7)
arranged in a grid pattern
3.1.10
volume model
solid model
three-dimensional geometric model which deals with the solid characteristics of an object in order to
represent its internal structure as well as its external shapes
Note 1 to entry: See ISO/IEC 2382 for definitions of volume modeling and solid modeling.
Note 2 to entry: Volume model can be represented with raster model (3.1.9) or vector model (3.1.8).
2
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
3.1.11
surface model
boundary model
data set of a model which represents the surfaces of objects
Note 1 to entry: See ISO/IEC 2382 for definitions of surfacing and surface modeling.
3.1.12
facet model
faceted model
surface model (3.1.11) of which surfaces consist of group of polygons
Note 1 to entry: A triangle is widely used as a polygon.
3.1.13
segmentation
process of separating the objects of interest from their surroundings
Note 1 to entry: Segmentation can be applicable to 2D, 3D, raster or vector data (3.1.8).
3.1.14
3D visualization
presentation intended for human viewing of a scene on a flat display surface, using graphics techniques
to convey depth information and knowledge of the arrangement and shapes of the visualized scene in a
three­dimensional space
Note 1 to entry: The graphics techniques can include use of perspective, occlusion, stereoscopy, lighting and
environmental effects, and ability to navigate the viewpoint to alternate positions and orientations.
3.1.15
3D modelling
activity intended to create a digital representation of the form and arrangement of one or more 3D
objects in a three-dimensional space.
Note 1 to entry: 3D models can contain geometric information such as mesh vertices, appearance, lighting, and
animation information. The created representation is a prerequisite to creating a 3D visualization (3.1.14) of the
modelled objects.
3.1.16
maximum intensity projection
MIP
scientific visualization method for 3D data that projects in the visualization plane the voxels with
maximum intensity that fall in the way of parallel rays traced from the viewpoint to the plane of
projection.
3.1.17
minimum intensity projection
MinIP
data visualization method that enables detection of low-density structures in a given volume
Note 1 to entry: The algorithm uses all the data in a volume of interest to generate a single two-dimensional
image. In other words, it consists of projecting the voxel with the lowest attenuation value on every view
throughout the volume onto a 2D image.
3
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 8 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
3.1.18
Hounsfield value
Hounsfield unit
integer representing the intensity of the image at each image point [pixel (3.1.6)] which originates from
the x-ray scanning process and in turn represents the image intensity which in turn depends on the
density of the tissue at that location
Note 1 to entry: Hounsfield values rise monotonically with tissue density but are not linearly proportional to
density.
Note 2 to entry: The highest range of biological tissue Hounsfield values is for cortical bone, and they can go even
higher for image artefacts such as metallic implants, metallic sections of a hospital bed included in the image, etc.
3.1.19
multiplanar reformation
MPR
two-dimensional reformatted images that are reconstructed secondarily in arbitrary planes from the
stack of axial image data.
Note 1 to entry: Multiplanar reformation (MPR) allows images to be created from the original axial plane in
either the coronal, sagittal or oblique plane.
3.1.20
volume rendering
set of techniques used to display a 2D projection of a 3D discretely sampled data set, typically a 3D
scalar field
3.2 Abbreviated terms
2D two­dimensional
3D three­dimensional
AM additive manufacturing
AMF additive manufacturing file format
ANN artificial neural network
CAD computer aided design
CT computed tomography
DICOM digital imaging and communications in medicine
HU Hounsfield unit
PACS picture archiving communication system
QC quality control
ROI region of interest
STL stereolithography
SVM support vector machine
4
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
4 Overview of image processing for the medical industry
4.1 Process flow
4.1.1 3D printing process for medical applications
In general, the medical 3D printing processing flow can be divided into eight phases, as shown in
Figure 1.
1) Image acquisition phase
In the image acquisition phase, medical images are acquired from medical imaging devices such as CT.
2) Segmentation phase
In the segmentation phase, the acquired medical images are segmented to fit the design purpose and
are processed to be divided (segmented) to extract a subset that would represent the part(s) of the
anatomy under consideration.
3) 3D modelling phase
In the 3D modelling phase, the segmented data representing the human tissue is converted
(reconstructed) into a 3D model optimized for 3D printing.
4) 3D printing phase
In the 3D printing phase, 3D printing is performed using the 3D model designed. For this phase 3D
model is processed for 3D printing by slicing, assigning build parameters, being oriented and placed
within the build space, and can have support structures generated.
5) Post-processing phase
In the post-processing phase, the 3D printed part is post-processed to become fit for actual medical use.
6) Quality control (QC) phase
In the QC phase, the 3D printed part is finally verified to meet all requirements (user/design/quality/
risk).
7) Clinical application and review phase
In the clinical application and review phase, the 3D printed part is reviewed as applicable to clinical
application by the healthcare practitioner.
8) Post-market phase
In the post­marketing stage, the 3D printed part is managed based on the post­sale market management
policy according to product life cycle issues such as tracking management/recall.
4.1.2 Explanation of a typical use case (cranial implant case)
Computed tomography (CT) is a common imaging modality for medical applications. For instance, for
patients with a skull defect visiting a neurosurgical clinic, CT has been known as the gold standard
for investigating bone­related problems. Figure 1 shows that the CT images are initially transferred to
the PACS server in DICOM file format. DICOM images have been used to reconstruct 3D image through
segmentation and 3D modelling by certain software. This 3D modelled image is transformed and
exported to design software as a stereolithography (STL) file. After completion and confirmation of
3D cranial implant by designing software, a metal AM machine builds this implant as designed. Post-
processing such as heat treatment, machining, cleaning and sanding is performed. Reverse engineering
is performed to confirm the completeness of the implant before delivery by 3D scanning and matching
to the original digital blueprint. After QC, the implant is packed, sterilized and delivered. An operation
5
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 10 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
is performed to cover the defect with the 3D printed cranial implant. For this medical 3D printing
process, accuracy and reproducibility should be considered. The accuracy and reproducibility of the
parts (anatomical model, surgical guides, implant, etc.) from medical 3D printed parts are affected
by the sum of errors introduced in each step during data flow. These steps can be image acquisition,
segmentation and any subsequent post-processing of the segmented images. This document covers
processes 1, 2 and 3 as shown in Figure 1, ending with a 3D model of the relevant patient anatomy for
use in multiple other later processes. Activities related to items for processes 4 - 8 are addressed by
ISO/TC 261.
Figure 1 — Typical workflow of medical 3D printing (example: cranial implant case)
5 General requirements
To conform to this document, all of the following items shall be considered and relevant information
shall be documented.
— The medical image acquisition protocol by the CT scanner.
— The clinical purpose (bone/hard tissue) of image­based modelling.
— The segmentation method and associated parameters.
— The processes and parameters for 3D reconstruction.
Major parameters, settings and descriptions of methods used in the processes above shall be recorded.
6
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 11 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
6 Requirements of data processing
6.1 Medical image data flow
There are usually two medical image data flows involved in data processing: example flow and direct
flow. MIP, MinIP, MPRs and volume rendering are used before transporting the medical image to the
PACS server. Typically, DICOM files are used to make 3D images. However, many PACS companies
provide plugged­in 3D visualization software to reform raw data to 3D images and transport 3D images
directly to the PACS server as captured images. These 3D visualizations on the PACS server are 2D
projections of a 3D object and are not suitable for 3D modelling. The 2D printers for films prints out 2D
images (X ray radiograph, CT, MRI, etc.) or 3D-modelled captured images. See Figure 2.
7
© ISO/IEC 2023 – All rights reserved

---------------------- Page: 12 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
Key
Example flow (dotted lines) Direct flow (solid lines)
1 scanned by device A scanned by device
2 convert to DICOM B 3D visualization
3 save to PACS server C save to PACS server
4 3D visualization D (transfer to STL)
5 transfer to STL or other file formats E directs to Figure 1
6 directs to Figure 1
Figure 2 — Data flow of medical images (Example)
6.2 Medical image acquisition/computed tomography scan
To make medical 3D models, sequential 2D images are necessary and should be acquired from sectional
images. Generally, 2D slice images are acquired from a CT scan of the patient's body at regular intervals
depending on the scanning needs. Each CT image set has its own strengths and weaknesses with
respect to the different objectives of the observation. Bones are typically clearly identified. Variabilities
8
  © ISO/IEC 2023 – All rights reserved

---------------------- Page: 13 ----------------------
ISO/IEC FDIS 3532-1:2023(E)
of output depend on factors such as spatial resolution/voxel size of the images, which in turn depends
on the x-ray dosage, the quality of the scanned images, operator capability, and low and high resolution
on 2D to 3D conversion algorithms. Features smaller than 0,3 mm cannot be printed successfully with
some printing processes if smaller features are needed. This will require special considerations for
process selection and post­processing operations. Care should be taken in choosing the appropriate
3D printing technology and the part manufacturers should be requested to consider the required
resolutions.
For medical image acquisition, the typical slice distance of less than 1 mm is sufficient and the following
points shall be addressed.
— Required accuracy and clinical purposes shall be compatible. The CT scanning protocol shall be
specified beforehand to achieve the required accuracy of the final models.
— The highest accuracy or resolution of CT scan is not always necessary.
— The time between the acquisition of the patient images and the initiation of image-based modelling
shall be minimized.
Other factors which can influence the quality of the final scanned images are as follows.
— Possible patient motion during the scanning process and its implications on the imaging accuracy.
Even breathing can cause errors in scanning in cardiovascular applications, for example.
— The use of contrast media during the scanning process to highlight blood or other liquids through
various tracers.
— Any digital filtering techniques applied in the data processing at the scanning stage to produce the
DICOM data used later for segmentation and reconstruction.
6.3 Segmentation
Image segmentation divides the image into meaningful regions. Segmentation usually extracts one or
more subsets of the data from the whole dataset, such that each subset would represent an anatomical
part, or tissue of the same characteristics [(e.g. bone having high density versus soft tissue having lower
density so a simple threshold can be established based on Hounsfield unit (HU)] etc. Segmentation in
medical imaging is generally considered a difficult problem, mainly because of the sheer size of the
datasets coupled with the complexity and variability of the anatomic organs.
The situation is worsened by the shortcomings of imaging modalities (such as sampling artifacts,
noise, low contrast, etc.) that can cause the boundaries of anatomical structures to be indistinct and
disconnected. The segmentation process becomes challenging in the absence of clear distinction in the
characteristics desired, such as density ranges overlapping (e.g. very soft bone indistinguishable from
calcified cartilage), or blood vessel walls not sufficiently distinguishable from surrounding muscle
tissue, etc. The challenge is greater when the target tissue is complex (intermingled or touching) in
its location (e.g. small diameter nerves around the orbit and vessels and nerves through skull base
foramens, anterior of a distal femur in a highly arthritic patient appearing to be totally connected to
their patella, or small complex branched blood vessels.).
Thus, the main challenge of segmentation algorithms is to accurately extract the boundary of the solid
organ or region of interest (ROI) and separate it from the rest of the dataset. The notion of boundary
starts in a 2D context of an image slice. Boundaries from all images (slices) can help build whole models
after segmentation.
For the segmentation of bony structures from CT scans the following relevant segmentation
technologies can
...

ISO/IEC JTC 1 -N:2021(E) Style Definition: Heading 1: Indent: Left: 0 cm, First line:
0 cm, Tab stops: Not at 0.76 cm
ISO/IEC DIS FDIS 3532-1:20212023(E)
Style Definition: a2: Outline numbered + Level: 2 +
Numbering Style: 1, 2, 3, … + Start at: 1 + Alignment: Left +
ISO/IEC JTC 1
Aligned at: 0 cm + Tab after: 0.63 cm + Indent at: 0 cm
Date: 2022-09-082023-02
Style Definition: a3: Outline numbered + Level: 3 +
Numbering Style: 1, 2, 3, … + Start at: 1 + Alignment: Left +
Seccretariat: ANSI
Aligned at: 0 cm + Tab after: 1.27 cm + Indent at: 0 cm
Information technology —Medical image-based modelling for 3D printing – Part 1: General
Style Definition: a4: Outline numbered + Level: 4 +
requirements.
Numbering Style: 1, 2, 3, … + Start at: 1 + Alignment: Left +
Aligned at: 0 cm + Tab after: 1.9 cm + Indent at: 0 cm
Style Definition: a5: Outline numbered + Level: 5 +
Numbering Style: 1, 2, 3, … + Start at: 1 + Alignment: Left +
Aligned at: 0 cm + Tab after: 1.9 cm + Indent at: 0 cm
Style Definition: a6: Outline numbered + Level: 6 +
Numbering Style: 1, 2, 3, … + Start at: 1 + Alignment: Left +
Aligned at: 0 cm + Tab after: 2.54 cm + Indent at: 0 cm
Style Definition: ANNEX: Outline numbered + Level: 1 +
Numbering Style: A, B, C, … + Start at: 1 + Alignment: Left +
Aligned at: 0 cm + Indent at: 0 cm
Style Definition: List Bullet: Indent: Left: 0 cm, Hanging:
0.63 cm, No bullets or numbering, Tab stops: 0.63 cm,
List tab
Style Definition: List Bullet 2: Indent: Left: 0.5 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
1.13 cm, List tab
Style Definition: List Bullet 3: Indent: Left: 1 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
1.63 cm, List tab
Style Definition: List Bullet 4: Indent: Left: 1.5 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
2.13 cm, List tab
Style Definition: List Bullet 5: Indent: Left: 2 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
2.63 cm, List tab
Style Definition: List Number: Indent: Left: 0 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
0.63 cm, List tab
Style Definition: List Number 5: Indent: Left: 2 cm,
Hanging: 0.63 cm, No bullets or numbering, Tab stops:
2.63 cm, List tab
Style Definition: 확인되지 않은 멘션1
Formatted: Font: 11.5 pt
Formatted: Font: 11.5 pt, Bold
Formatted: Font: 11.5 pt, Bold
Formatted: Font: Bold
Formatted: Font: Bold

---------------------- Page: 1 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
Formatted
© ISO 20212023, Published in Switzerland
All rights reserved. Unless otherwise specified, 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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
[e-mail]
www.iso.org
iiii © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
Contents
Foreword . iv
Introduction . v
1 Scope . 1
2 Normative references . 1
3 Terms, definitions and abbreviated terms . 1
3.1 Terms and definitions . 1
3.2 Abbreviated terms. 4
4 Overview of image processing for the medical industry . 5
4.1 Process flow . 5
4.1.1 3D printing process for medical applications . 5
4.1.2 Explanation of a typical use case (cranial implant case) . 6
5 General requirements . 6
6 Requirements of data processing. 7
6.1 Medical image data flow . 7
6.2 Medical image acquisition/computed tomography scan . 8
6.3 Segmentation . 9
6.4 3D reconstruction and visualization . 10
6.5 Calibration and validation of 2D and 3D conversion . 12
6.6 File format . 12
Annex A (Informative) Reporting . 13
Bibliography . 14

Foreword . v
Introduction . vi
1 Scope . 1
2 Normative references . 1
3 Terms, definitions and abbreviated terms . 1
3.1 Terms and definitions . 1
3.2 Abbreviated terms. 4
4 Overview of image processing for the medical industry . 5
4.1 Process flow . 5
4.1.1 3D printing process for medical applications . 5
4.1.2 Explanation of a typical use case (cranial implant case) . 6
5 General requirements . 7
6 Requirements of data processing. 7
6.1 Medical image data flow . 7
6.2 Medical image acquisition/computed tomography scan . 10
6.3 Segmentation . 10
6.4 3D reconstruction and visualization . 12
6.5 Calibration and validation of 2D and 3D conversion . 13
6.6 File format . 14
Annex A (informative) Reporting . 15
© ISO/IEC 20212023 – All rights reserved iiiiii

---------------------- Page: 3 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
Bibliography . 16
iviv © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
Formatted: Tab stops: Not at 14.94 cm
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical activity.
ISO and IEC technical committees collaborate in fields of mutual interest. Other international
organizations, governmental and non-governmental, in liaison with ISO and IEC, also take part in the
work.
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 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 or
www.iec.ch/members_experts/refdocs).
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent rights. Details
of any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www.iso.org/patents) or the IEC list of patent
declarations received (see https://patents.iec.ch).
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. In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology.
A list of all parts in the ISO/IEC 3532 series can be found on the ISO and IEC websites.
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 and www.iec.ch/national-
committees.

© ISO/IEC 20212023 – All rights reserved vv

---------------------- Page: 5 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
Introduction
This document was developed in response to the need for customization of 3D scanning and 3D printing
technology within the medical industry, which can be achieved by taking full advantage of information
and communication technology (ICT).
This document addresses the overview of medical image processing and requirements for image-based
modelling. 3D printing technology has caused a revolution in health care delivery. New classes of medical
devices embody the true meaning of personalized medicine. Medical device designers and practitioners
are able to practically and efficiently create devices that were very difficult or impossible to create before.
In addition to using 3D printing technology to create standard medical devices with features like intricate
lattice structures, clinicians and engineers work in conjunction to produce what are known as patient-
specific devices or patient-matched devices. These are medical devices designed to fit a specific patient’s
anatomy, typically using medical imaging from that patient. Anatomically matched devices have very
complex geometrical contours and shapes. Several challenges exist in the design process between the
input data and the final device design. Most of these steps definitely depend on software-based
management of medical images.
Overall, the world revenue from 3D printing technology in the healthcare industry is expected to grow
exponentially, yet very few guides exist for 3D printing for medical practice. Medical images from the
human body are different from solid objects due to the non-geometric nature of the human body. To
perform 3D printing for medical practice, an accurate and consistent approach for image processing and
data creation from medical images is needed. Standardization for 3D printing processes in medicine is
urgently required for education, diagnosis, neurosurgical treatment, developing simulation models,
medical equipment (including surgical guides) and surgical implantable devices in the clinical fields.
Regulatory bodies from several countries (US, Korea, Repulic of Korea, etc.) have already published their
own guidelines for approval. However, those guidelines are not specifically designed for 3D printing
technology.
Applications of 3D printing in medicine are booming, such as surgical simulation models, surgical guides,
educational models, surgical implants, etc. Those which are manufactured by 3D printing technology
require patient- and/or procedure-specific data (e.g. planned surgical technique and others) and medical
image data acquisition processing. Most of the processing of medical images for 3D printing medical
devices is software-based. In order to accurately and consistently visualize human body anatomy,
appropriate software-based modelling for 3D printing is needed. This document provides  requirements
of software-based medical image processing for the purpose of producing 3D models for 3D printing.
Valuable information related to optimized medical image data for additive manufacturing can be found
in ISO/ASTM TR 52916. Formatted: std_publisher
Formatted: std_documentType
Formatted: std_docNumber
vivi © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 6 ----------------------
FINAL DRAFT INTERNATIONAL STANDARD ISO/IEC DISFDIS 3532-1:20212023(E)
Formatted: Font: 11.5 pt

Information Technology —technology — Medical Image-Based
Modellingimage-based modelling for 3D printing– — Part 1:
General Requirementsrequirements
1 Scope
This document specifies the requirements for medical image-based modelling for 3D printing for
medical applications. It concerns accurate 3D data modelling in the medical field using medical image
data generated from computed tomography (CT) devices. It also specifies the principal considerations for
the general procedures of medical image-based modelling. It excludes soft tissue modelling from
magnetic resonance image (MRI).
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/IEC 2382:2015, Information technology — Vocabulary
ISO/ASTM/ 52900:2021, Additive manufacturing –— General principles –— Fundamentals and
Vocabulary
3 Terms, definitions and abbreviated terms
For the purposes of this document, the terms and definitions given in ISO/IEC 2382:2015, ISO/ASTM Formatted: std_publisher
52900:2021, and the following apply.
Formatted: std_docNumber
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
Formatted: std_publisher
Formatted: std_docNumber
— ISO Online browsing platform: available at https://www.iso.org/obp
Formatted: std_docNumber
— IEC Electropedia: available at https://www.electropedia.org/
Formatted: Default Paragraph Font, English (United
Kingdom)
3.1 3.1 Terms and definitions
Formatted: Default Paragraph Font, Font: Times New
Roman
3.1.1
Formatted: No bullets or numbering
image acquisition
scanning of the structure of interest using computed tomography (CT), magnetic resonance imaging or
other three-dimensional imaging technology
[SOURCE: ISO 21227-1:2003(en), 3.4]
3.1.2
slice distance
slice spacing
distance between the centre of the slices, which is calculated by the difference in the slice locations of two
adjacent slices
© ISO/IEC 20212023 – All rights reserved 11

---------------------- Page: 7 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
3.1.3
hard tissue
tissue which is mineralized and has a firm intercellular matrix (such as bone, tooth enamel, dentin and
cementum)
3.1.4
soft tissue
tissue that connects, supports or surrounds other structures and organs of the body, excluding hard tissue
Formatted: Font: Italic
(3.1.3)
3.1.5
solid organ
organ which has firm tissue consistency such as the heart, kidney, liver, lungs, and pancreas, etc.,
excluding hollow organs (such as the organs of the gastrointestinal tract) and tissue with liquid
consistency (such as blood)
3.1.6
pixel
picture element
smallest two-dimensional element of a display image that can be independently assigned attributes such
as color and intensity
[SOURCE: ISO/IEC 2382:2015, 13.03.08]2125999, modified — Notes to entry have been removed.] Formatted: English (United States)
Formatted: Default Paragraph Font, English (United
3.1.7
States)
voxel
Formatted: English (United States)
volume element
smallest three-dimensional element in volume or volumetric (solid) modeling that can be independently
assigned attributes such as colour and intensity
[SOURCE: ISO/IEC 2382:2015, 13.03.09]2126000, modified — Notes to entry have been removed; "solid" Formatted: Default Paragraph Font
has been replaced by "volume or volumetric (solid)".]
3.1.8
vector data
vector image
vector model
digital description of 2D image or 3D model stored as a series of points and mathematical functions to
describe the geometric figure
[SOURCE: ISO 12651-1:2012, 4.139, modified — “3D model” has been added, points"image" has been
replaced by “vertices”]"2D image or 3D model”.]
3.1.9
raster data
raster image
raster model
bitmap data
bitmap image
bitmap model
2D image or 3D model data formed by a set of picture elements (3.1.6) or volume elements (3.1.7) arranged Formatted: Font: Italic
in a grid pattern
Formatted: Font: Italic
[SOURCE: ISO 12651-1:2012, 4.18, modified — “3D model” and “volume elements” has been added]
22 © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 8 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
3.1.10
volume model
solid model
three-dimensional geometric model which deals with the solid characteristics of an object in order to
represent its internal structure as well as its external shapes
Note 1 to entry: Solid modeling;See ISO/IEC 2382 for definitions of volume modeling: terms and definition
standardized by ISO/IEC [ISO/IEC 2382-13:1996; ISO/IEC 2382-24:1995].solid modeling.
Note 2 to entry: Volume model can be represented with raster model (3.1.9) or vector model. (3.1.8).
Formatted: Font: Italic
Formatted: Font: Italic
3.1.11
surface model
boundary model
data set of a model which represents the surfaces of objects
Note 1 to entry: See ISO/IEC 2382 for definitions of surfacing; and surface modeling: terms and definition
standardized by ISO/IEC [ISO/IEC 2382-13:1996; ISO/IEC 2382-24:1995].
3.1.12
facet model
faceted model
surface model (3.1.11) of which surfaces consist of group of polygons
Note 1 to entry: A triangle is widely used as a polygon.
3.1.13
Formatted: Don't adjust space between Latin and Asian
text, Don't adjust space between Asian text and numbers
segmentation
process of separating the objects of interest from their surroundings
Note 1 to entry: Segmentation can be applicable to 2D, 3D, raster or vector data. (3.1.8).
Formatted: Font: Italic
3.1.14
3D visualization
presentation intended for human viewing of a scene on a flat display surface, using graphics techniques
to convey depth information and knowledge of the arrangement and shapes of the visualized scene in a
three-dimensional space
Note 1 to entry: The graphics techniques can include use of perspective, occlusion, stereoscopy, lighting and
environmental effects, and ability to navigate the viewpoint to alternate positions and orientations.
3.1.15
3D modelling
activity intended to create a digital representation of the form and arrangement of one or more 3D objects
in a three-dimensional space.
Note 1 to entry: 3D Models maymodels can contain geometric information such as mesh vertices, appearance,
lighting, and animation information. The created representation is a prerequisite to creating a 3D visualization
Formatted: Font: Italic
(3.1.14) of the modelled objects.
3.1.16
maximum intensity projection
MIP
© ISO/IEC 20212023 – All rights reserved 33

---------------------- Page: 9 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
scientific visualization method for 3D data that projects in the visualization plane the voxels with
maximum intensity that fall in the way of parallel rays traced from the viewpoint to the plane of
projection.
3.1.17
minimum intensity projection
MinIP
data visualization method that enables detection of low-density structures in a given volume.
Note 1 to entry: The algorithm uses all the data in a volume of interest to generate a single two-dimensional
image. In other words, it consists of projecting the voxel with the lowest attenuation value on every view throughout
the volume onto a 2D image.
3.1.18
Hounsfield Valuevalue
Hounsfield Unitunit
an integer representing the intensity of the image at each image point ([pixel) (3.1.6)] which originates Formatted: Font: Italic
from the x-ray scanning process and in turn represents the image intensity which in turn depends on the
density of the tissue at that location.
Note 1 to entry: Hounsfield values rise monotonically with tissue density but are not linearly proportional to
density.
Note 2 to entry: The highest range of biological tissue Hounsfield values is for cortical bone, and they can go even
higher for image artefacts such as metallic implants, metallic sections of a hospital bed included in the image, etc.


3.1.19
multiplanar reformation
MPR
two-dimensional reformatted images that are reconstructed secondarily in arbitrary planes from the
stack of axial image data.
Note 1 to entry: Multiplanar reformation (MPR) allows images to be created from the original axial plane in either
Formatted: Tab stops: 2.8 cm, Left
the coronal, sagittal, or oblique plane.
3.1.20
volume rendering
set of techniques used to display a 2D projection of a 3D discretely sampled data set, typically a 3D scalar
field
3.2 3.2 Abbreviated terms
Formatted: No bullets or numbering
2D two-dimensional
3D three-dimensional
AM additive manufacturing
AMF additive manufacturing file format
ANN artificial neural network
CAD computer aided design
CT computed tomography
44 © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 10 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
DICOM digital imaging and communications in medicine
HU Hounsfield unit
PACS picture archiving communication system
QC quality control
ROI region of interest
STL stereolithography
SVM support vector machine
4 Overview of image processing for the medical industry
4.1 4.1 Process flow
Formatted: No bullets or numbering
4.1.1 4.1.1 3D printing process for medical applications
In general, the medical 3D printing processing flow can be divided into eight phases, as shown in Figure 1. Formatted: cite_fig
Formatted: cite_fig
1) Image acquisition phase
Formatted: Font: Bold
In the image acquisition phase, medical images are acquired from medical imaging devices such as CT.
2) Segmentation phase Formatted: Font: Bold
In the segmentation phase, the acquired medical images are segmented to fit the design purpose and are
processed to be divided (segmented) to extract a subset that would represent the part(s) of the anatomy
under consideration.
3) 3D modelling phase
Formatted: Font: Bold
In the 3D modelling phase, the segmented data representing the human tissue is converted Formatted: Body Text, Tab stops: Not at 1.4 cm + 2.1 cm
+ 2.8 cm + 3.5 cm + 4.2 cm + 4.9 cm + 5.6 cm + 6.3 cm +
(reconstructed) into a 3D model optimized for 3D printing.
7 cm
4) 3D printing phase
Formatted: Font: Bold
In the 3D printing phase, 3D printing is performed using the 3D model designed. For this phase 3D model
is processed for 3D printing by slicing, assigning build parameters, being oriented and placed within the
build space, and can have support structures generated.
5) Post-processing phase
Formatted: Font: Bold
In the post-processing phase, the 3D printed part is post-processed to become fit for actual medical use.
6) Quality control (QC) phase Formatted: Font: Bold
Formatted: Font: Bold
In the QC phase, the 3D printed part is finally verified to meet all requirements
Formatted: Font: Bold
(user/design/quality/risk).
7) Clinical application and review phase Formatted: Font: Bold
In the clinical application and review phase, the 3D printed part is reviewed as applicable to clinical
application by the healthcare practitioner.
8) Post-market phase
Formatted: Font: Bold
In the post-marketing stage, the 3D printed part is managed based on the post-sale market management
policy according to product life cycle issues such as tracking management/recall.
© ISO/IEC 20212023 – All rights reserved 55

---------------------- Page: 11 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)
4.1.2 4.1.2 Explanation of a typical use case (cranial implant case)
Formatted: No bullets or numbering
Computed tomography (CT) is a common imaging modality for medical applications. For instance, for
patients with a skull defect visiting a neurosurgical clinic, CT has been known as the gold standard for
investigating bone-related problems. Figure 1 shows that the CT images are initially transferred to the
PACS server asin DICOM file format. DICOM images have been used to reconstruct 3D image through
segmentation and 3D modelling by certain software. This 3D modelled image is transformed and
exported to design software as a stereolithography (STL) file. After completion and confirmation of 3D
cranial implant by designing software, a metal AM machine builds this implant as designed. Post-
processing such as heat treatment, machining, cleaning and sanding is performed. Reverse engineering
is performed to confirm the completeness of the implant before delivery by 3D scanning and matching to
the original digital blueprint. After quality controlQC, the implant is packed, sterilized, and delivered. An
operation is performed to cover the defect with the 3D printed cranial implant. For this medical 3D
printing process, accuracy and reproducibility should be considered. The accuracy and reproducibility of
the parts (anatomical model, surgical guides, implant, etc.) from medical 3D printed parts are affected by
the sum of errors introduced in each step during data flow. These steps can be image acquisition,
segmentation, and any subsequent post-processing of the segmented images. This document scope
covers the processes 1, 2 and 3 of Fig. as shown in Figure 1, ending with a 3D model of the relevant patient Formatted: cite_fig
anatomy for use in multiple other later processes. Activities related to items for processes 4, 5, 6, 7 and -
8 are addressed by ISO/TC 261 standards. .

66 © ISO/IEC 20212023 – All rights reserved

---------------------- Page: 12 ----------------------
ISO/IEC DISFDIS 3532-1:20212023(E)

Figure 1 — Typical workflow of medical 3D printing (example: cranial implant case)
5 General requirements
To comply withconform to this document, all of the following items shall be considered and relevant
selectioninformation shall be documented.
— — The medical image acquisition protocol by the CT scanner. Formatted: Indent: Left: 0 cm, Hanging: 1.34 cm, No
bullets or numbering, Tab stops: 1.4 cm, Left
— The clinical purpose (bone/hard tissue) of image-based modelling.
— T
...