Artificial intelligence in nuclear medicine

Clinical Snapshot

Key Findings

• P Value: Not applicable to this review format

• Effect Size: Not quantified; no pooled effect sizes reported. The review is descriptive and qualitative in nature

• Primary Outcome: Narrative synthesis of AI applications across the nuclear medicine imaging workflow, encompassing image acquisition acceleration, image quality enhancement, advanced image generation, image interpretation assistance, and treatment planning support across oncology, neurology, and radionuclide therapy

• Nnt Or Sensitivity: Not reported. The single quantitative benchmark cited is that approximately 2% of clinical AI applications in general healthcare progress beyond the prototyping phase, suggesting very low clinical translation rates. Specific sensitivity, specificity, or diagnostic accuracy metrics for individual AI applications are not synthesised

• Confidence Interval: Not reported; no meta-analytic synthesis performed

Clinical Application

Clinical implementation of AI in nuclear medicine remains limited globally, with the authors noting fewer than 2% of AI applications reach commercial maturity. Feasibility depends on local computational infrastructure, validated and regulatory-approved software, trained workforce, and institutional governance frameworks for AI deployment. Integration into existing PACS and nuclear medicine information systems presents practical barriers. Workflow redesign and clinical validation in local populations are prerequisites for safe implementation In Australia, nuclear medicine AI tools would require TGA regulatory approval as Software as a Medical Device (SaMD) under the TGA's Digital Health framework before clinical deployment. The TGA adopted the International Medical Device Regulators Forum (IMDRF) framework for SaMD, and AI tools performing diagnostic functions would typically require Class IIb or Class III classification depending on intended use and risk. The RACGP and ANZSNM (Australian and New Zealand Society of Nuclear Medicine) have not yet published specific AI implementation guidelines for nuclear medicine at the time of this appraisal. PBS implications are indirect — improved diagnostic accuracy from AI-assisted nuclear medicine could influence downstream treatment decisions affecting PBS-listed therapies, particularly in oncology and theranostics (e.g., lutetium-177 PSMA therapy for prostate cancer, which has PBS listing). Australian nuclear medicine services are concentrated in metropolitan tertiary centres, making the equity argument for AI-enabled decentralisation particularly relevant to rural and remote populations. Medicare Benefits Schedule (MBS) item numbers for nuclear medicine studies do not currently differentiate AI-assisted from conventional interpretation, creating reimbursement ambiguity. Australian clinicians should await TGA-approved, locally validated AI tools before clinical adoption. Patients undergoing PET/CT, SPECT/CT, and other nuclear medicine investigations across oncology (including theranostics), neurology (dementia, movement disorders), and cardiology in tertiary and secondary care settings. Most relevant to centres with existing nuclear medicine infrastructure considering AI tool adoption or procurement

Abstract

Artificial intelligence (AI) holds great promise for advancing diagnostics and treatment in nuclear medicine. The rapid growth of AI over the past decade has been largely driven by advances in hardware components such as graphics processing units (GPUs) and the introduction of deep learning (DL) and convolutional neural networks (CNN). The integration of AI and medical imaging has the potential to revolutionize nuclear medicine by, for example, accelerating image acquisition, enhancing image quality, enabling advanced image generation, assisting image interpretation, and aiding treatment planning. Clinical applications have been demonstrated for most medical specialties, including oncology, neurology, and radionuclide therapy. The utilization of AI to provide automated, standardized procedures can help bring advanced imaging from major university centres to smaller local clinics, thus benefiting a broader range of patients. Additionally, AI has vast potential for predicting optimal treatment strategies, assessing risk, optimizing patient flow and outcomes, and even improving productivity, but these capabilities have yet to be fully utilized. The fraction of clinical AI applications in general healthcare reaching beyond the prototyping phase is reported to be as low as 2%. Indeed, in nuclear medicine, very few AI developments have reached commercial maturity. Currently, most AI applications in nuclear medicine follow the imaging flow from image acquisition and reconstruction, post-processing and image preparation, image analysis, and decision support for clinical interpretation. Below, we will briefly review selected areas and comment on challenges and opportunities for AI in nuclear medicine, with a special focus on the transition from development to clinical implementation.

References

1. 1.Andersen, F. L., & Hansen, A. E. (2026). Artificial intelligence in nuclear medicine. The British Journal of Radiology. https://doi.org/10.1093/bjr/tqag012