wk10 AI

A set of vocabulary flashcards covering terminology, applications, and professional roles regarding Artificial Intelligence in Nuclear Medicine.

Artificial Intelligence (AI)

“the capability of a computer program to think, learn, react, adapt to solve problems, and perform reasoning like a human”.

Machine Learning (ML)

A branch of AI defined as “a system that has the capacity to improve and learn to recognise patterns of disease features”; it finds patterns in data to make predictions without being specifically programmed.

Deep Learning (DL)

A branch of machine learning that uses deep neural networks to perform more sophisticated tasks with less human involvement than standard machine learning.

Radiomics

“extracting clinically meaningful quantitative features from medical images” to enhance data and clinical decision-making.

Large Language Models (LLMs)

Advanced AI systems, such as ChatGPT, trained on large amounts of text data to understand and generate human-like language.

Investment in medical imaging AI research (2024)

Estimated to be approximately US$3-6$ billion, reflecting an increase from US$80$ million in 2016.

Pre-appointment screening

A nuclear medicine application of AI focusing on scheduling and determining the amount of radioisotopes needed for each day.

Theranostics

An area of nuclear medicine where AI enables more precise dosimetry, which is critical for treatment planning.

Multimodal AI models

Future AI directions that combine data from diverse modalities including PET, SPECT, CT, MRI, genomics, and clinical records to create comprehensive patient profiles.

Imaging biobanks

The databases that feed AI; building quality versions of these is a key role for medical radiation practitioners.

Emotional intelligence

A human attribute that will always be at the heart of patient care, distinguishing the role of a human technologist from a robot.