Chapter 8. Artificial Intelligence-based Software
116 Regulatory Affairs Professionals Society (RAPS)
Table 8-1. Terms, Abbreviations and Definitions Related to AI and ML Technologies
Term Abbreviation Definition
Artificial intelligence AI A branch of computer science, statistics, and engineering that uses
algorithms or models to perform tasks and exhibit behaviors that mimics
humans in making decisions and predictions. The FDA interprets AI as
algorithms or mathematical models that can analyze and interpret data9
Computer-assisted triage CADt CADt is computerized image processing device intended to aid in pri-
oritization and triage of time sensitive patient detection and diagnosis
based on the analysis of medical images acquired from radiological signal
acquisition systems10
Computer-assisted
detection
CADe These are computerized systems intended to identify, mark, highlight, or
in any other manner, direct attention to portions of an image, or aspects
of radiology device data, that may reveal specific abnormalities during
interpretation of patient radiology images or patient radiology device data
by the clinician11
Computer-assisted
diagnosis
CADx These are computerized systems intended to provide an assessment of
disease or other conditions in terms of the likelihood of the presence or
absence of disease, or are intended to specify disease type (i.e., spe-
cific diagnosis or differential diagnosis), severity, stage, or intervention
recommended12
Computer-assisted
detection and diagnosis
CADe/x These are computerized systems intended to provide both CADe and
CADx features12
Digital Imaging and
Communications in
Medicine
DICOM It is an international standard that specifies the protocols used to facili-
tate the exchange of communication and management of medical image
information and data13
Medical image manage-
ment and processing
system
MIMPS A medical image management and processing system is a device that
provides one or more capabilities relating to the review and digital pro-
cessing of medical images for the purposes of interpretation by a trained
practitioner of disease detection, diagnosis, or patient management12
Machine learning ML The subset of AI known as ML consists of algorithms that can learn
from large data sets and make predictions without being explicitly
programmed14
Intended use – The intended use describes the general purpose of the device or its func-
tion and encompasses the indications for use15
Indications for use IFU The IFU of a device is defined in 21 CFR §814.20(b)(3)(i) as “the disease
or condition the device will diagnose, treat, prevent, cure or mitigate,
including a description of the patient population for which the device is
intended”16
Software as a medical
device
SaMD Software intended to be used for one or more medical purposes that per-
form these purposes without being part of a hardware medical device17
Software in a medical
device
SiMD Software embedded in hardware or that is an integral part of a medical
device18
Supervised machine
learning
– Machine learning that makes use of labeled data during training. ML mod-
els are trained with training data that includes a known or determined
output or target variable19
Unsupervised machine
learning
– Machine learning that makes use of unlabeled data during training18
Training data – A set of data used my manufacturers in procedures and machine learning
training to build an ML model20
116 Regulatory Affairs Professionals Society (RAPS)
Table 8-1. Terms, Abbreviations and Definitions Related to AI and ML Technologies
Term Abbreviation Definition
Artificial intelligence AI A branch of computer science, statistics, and engineering that uses
algorithms or models to perform tasks and exhibit behaviors that mimics
humans in making decisions and predictions. The FDA interprets AI as
algorithms or mathematical models that can analyze and interpret data9
Computer-assisted triage CADt CADt is computerized image processing device intended to aid in pri-
oritization and triage of time sensitive patient detection and diagnosis
based on the analysis of medical images acquired from radiological signal
acquisition systems10
Computer-assisted
detection
CADe These are computerized systems intended to identify, mark, highlight, or
in any other manner, direct attention to portions of an image, or aspects
of radiology device data, that may reveal specific abnormalities during
interpretation of patient radiology images or patient radiology device data
by the clinician11
Computer-assisted
diagnosis
CADx These are computerized systems intended to provide an assessment of
disease or other conditions in terms of the likelihood of the presence or
absence of disease, or are intended to specify disease type (i.e., spe-
cific diagnosis or differential diagnosis), severity, stage, or intervention
recommended12
Computer-assisted
detection and diagnosis
CADe/x These are computerized systems intended to provide both CADe and
CADx features12
Digital Imaging and
Communications in
Medicine
DICOM It is an international standard that specifies the protocols used to facili-
tate the exchange of communication and management of medical image
information and data13
Medical image manage-
ment and processing
system
MIMPS A medical image management and processing system is a device that
provides one or more capabilities relating to the review and digital pro-
cessing of medical images for the purposes of interpretation by a trained
practitioner of disease detection, diagnosis, or patient management12
Machine learning ML The subset of AI known as ML consists of algorithms that can learn
from large data sets and make predictions without being explicitly
programmed14
Intended use – The intended use describes the general purpose of the device or its func-
tion and encompasses the indications for use15
Indications for use IFU The IFU of a device is defined in 21 CFR §814.20(b)(3)(i) as “the disease
or condition the device will diagnose, treat, prevent, cure or mitigate,
including a description of the patient population for which the device is
intended”16
Software as a medical
device
SaMD Software intended to be used for one or more medical purposes that per-
form these purposes without being part of a hardware medical device17
Software in a medical
device
SiMD Software embedded in hardware or that is an integral part of a medical
device18
Supervised machine
learning
– Machine learning that makes use of labeled data during training. ML mod-
els are trained with training data that includes a known or determined
output or target variable19
Unsupervised machine
learning
– Machine learning that makes use of unlabeled data during training18
Training data – A set of data used my manufacturers in procedures and machine learning
training to build an ML model20