Machine Learning in Nuclear Medicine: Part 1—Introduction

Abstract

Learning Objectives: On successful completion of this activity, participants should be able to (1) provide an introduction to machine learning, neural networks, and deep learning; (2) discuss common machine learning algorithms with illustrative examples and figures; and (3) compare machine learning algorithms and provide guidance on selection for a given application. Financial Disclosure: Sandra E. Black received in-kind funding to her institution from GE Healthcare and Avid Pharmaceuticals. The authors of this article have indicated no other relevant relationships that could be perceived as a real or apparent conflict of interest. CME Credit: SNMMI is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to sponsor continuing education for physicians. SNMMI designates each JNM continuing education article for a maximum of 2.0 AMA PRA Category 1 Credits. Physicians should claim only credit commensurate with the extent of their participation in the activity. For CE credit, SAM, and other credit types, participants can access this activity through the SNMMI website (http://www.snmmilearningcenter.org) through April 2022. This article, the first in a 2-part series, provides an introduction to machine learning (ML) in a nuclear medicine context. This part addresses the history of ML and describes common algorithms, with illustrations of when they can be helpful in nuclear medicine. Part 2 focuses on current contributions of ML to our field, addresses future expectations and limitations, and provides a critical appraisal of what ML can and cannot do.

Introduction to Radiomics

“…Here, a hold-out set of data that is only touched after algorithm finalization during a final validation is best practice. For further information about machine learning, please refer to the recent article by Uribe et al (11).…”

“…Radiomic features have also been suggested to predict clinical endpoints such as survival and treatment response, and, to be directly linked to genomic, transcriptomic, or proteomic characteristics (1,2,9). While even individual radiomic features may correlate with genomic data or clinical outcomes, the impact of radiomics is increased when the wealth of information that it provides -typically hundreds of features, a fraction of which will contribute to a disease-specific "radiomic signature"-is processed using machine learning techniques (10,11).…”

Artificial intelligence and radiomics in nuclear medicine: potentials and challenges

“…Recent studies reported the use of machine learning techniques in the prediction of diagnosis on CT, MR, SPECT, and PET in a limited number of cases; these results are yet to be validated in a larger data set obtained in clinical setting [20][21][22][23]. While machine reporting sounds to be a threat to imaging professionals (replacement of physicians by AI tools), there are several technical, organizational, financial, ethical, and administrative issues to be resolved before it is implemented in practice.…”

Intelligent Imaging: Anatomy of Machine Learning and Deep Learning

“…ML algorithms, including ANNs, have 3 key components (6,8). The first is the mathematical model that is used to describe or explain the relationships within the data; specifically, the relationships between inputs (features) and outputs (outcomes).…”

“…2). This phase predicts the accuracy of the ANN when used clinically or in research (5,6). That degree of accuracy can then be expected in the application phase, in which the neural network makes inferences about images without a grounded truth (Supplemental Fig.…”