Using a Novel AI Tool for Cancer Detection on Whole-Body PET/CT Scans

Get Permission

A novel artificial intelligence (AI) tool may accurately detect six different types of cancers on whole-body positron-emission tomography/computed tomography (PET/CT) scans and automatically quantify tumor burden to assess patient risk, treatment response, and survival, according to new findings presented by Leung et al at the 2024 Society of Nuclear Medicine and Molecular Imaging (SNMMI) Annual Meeting (Abstract 241979).


“Automatic detection and characterization of cancer are important clinical needs to enable early treatment,” explained lead study author Kevin H. Leung, PhD, a research associate at Johns Hopkins University School of Medicine. “Most AI models that aim to detect cancer are built on small to moderately sized data sets that usually encompass a single malignancy and/or radiotracer. This represents a critical bottleneck in the current training and evaluation paradigm for AI applications in medical imaging and radiology,” he added.

Study Methods and Results

In the new study, researchers used 611 fluorodeoxyglucose PET/CT scans of patients with lung cancer, melanoma, lymphoma, head and neck cancer, and breast cancer as well as 408 prostate-specific membrane antigen PET/CT scans of patients with prostate cancer to develop a deep transfer learning approach for fully automated, whole-body tumor segmentation and prognosis on PET/CT scans.

The researchers found that the novel AI tool automatically extracted radiomic features and whole-body imaging measures from the predicted tumor segmentations to quantify molecular tumor burden and uptake across all cancer types. They then used quantitative features and imaging measures to build models capable of demonstrating prognostic value for risk stratification, survival estimation, and prediction of treatment response in patients with cancer.


The researchers indicated that generalizable, fully automated AI tools could play a major role at imaging centers in the future by aiding in the interpretation of PET/CT scans of patients with cancer. The novel deep learning approach may lead to the discovery of important molecular insights about the underlying biological processes that may be currently understudied in large-scale patient populations.

“In addition to performing cancer prognosis, the approach provides a framework that will help improve patient outcomes and survival by identifying robust predictive biomarkers, characterizing tumor subtypes, and enabling the early detection and treatment of cancer,” underscored Dr. Leung. “The approach may also assist in the early management of patients with advanced, end-stage disease by identifying appropriate treatment regimens and predicting response to therapies such as radiopharmaceutical therapy,” he concluded.

Disclosure: For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.