PPubMed16 Oct 2019Cited 71×

The Current State of Artificial Intelligence in Medical Imaging and Nuclear Medicine The Current State of AI in Medical Imaging and Nuclear Medicine Lee <i> et al </i>

Louise I T Lee, Senthooran Kanthasamy, Radha S Ayyalaraju, Rakesh Ganatra

Abstract

'Rakesh Ganatra'] The last decade has seen a huge surge in interest surrounding artificial intelligence (AI). AI has been around since the 1950s, although technological limitations in the early days meant performance was initially inferior compared to humans.1 With rapid progression of algorithm design, growth of vast digital datasets and development of powerful computing power, AI now has the capability to outperform humans. Consequently, the integration of AI into the modern world is skyrocketing. This review article will give an overview of the use of AI in the modern world and discuss current and potential uses in healthcare, with a particular focus on its applications and likely impact in medical imaging. We will discuss the consequences and challenges of AI integration into healthcare.

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Answer

Lee et al. review how AI is reshaping medical imaging and nuclear medicine, highlighting clinical successes, workflow gains, and the need for regulation.

What’s new

i.Comprehensive review of AI applications in image interpretation, workflow optimization, and hybrid imaging for medical imaging and nuclear medicine
ii.Summarized high‑performance AI studies, e.g., deep‑learning diagnosis of Parkinson's disease with 94‑96% sensitivity and MRI acceleration using 29% of k‑space data
iii.Outlined the UK RCR framework, regulatory gaps, and national AI projects and industry efforts
iv.Identified practical challenges such as data access, validation, and ethical considerations for clinical translation

How they did it

The authors performed a narrative literature review of recent peer‑reviewed studies, conference reports, and policy documents related to AI in radiology and nuclear medicine. They extracted examples of clinical performance, workflow impact, and regulatory initiatives, and organized them into thematic sections covering imaging modalities, AI‑driven workflow tools, and implementation frameworks.

What they found

Deep‑learning models achieved 96% sensitivity for Parkinson's disease diagnosis from DAT‑SPECT scans (Kim et al.)
Another SPECT AI system reached 94% sensitivity, reducing inter‑observer variability (Choi et al.)
AI‑triage software (Viz.ai) saved an average of 52 minutes in >95% of acute stroke cases
MRI acquisition time was cut by 71% using a deep‑learning method that required only 29% of k‑space data
UK imaging demand rose 30% in five years, highlighting the need for AI‑driven work‑list prioritization

Caveats

✕Absence of a unified regulatory framework for AI deployment in clinical practice
✕Limited large‑scale, prospective validation of many reported AI algorithms
✕Data privacy, security, and access to ‘big data’ hinder broader AI development
✕Smaller patient volumes in nuclear medicine reduce available training datasets

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Similar papers in all four archives

The same idea searched in arXiv, PubMed, bioRxiv and chemRxiv at once, so a method can lead you to where it is being applied.

AarXiv4 found

PPubMed4 found

BbioRxiv4 found

CchemRxiv3 found

Searching the other archives…