(Press-News.org) OAK BROOK, Ill. – An artificial intelligence (AI) deep learning tool that estimates the malignancy risk of lung nodules achieved high cancer detection rates while significantly reducing false-positive results. Results of the study, which used data from large, multi-site lung cancer screening trials, were published today in Radiology, a journal of the Radiological Society of North America (RSNA).
Lung cancer remains a significant global health issue, causing the most cancer-related deaths worldwide. Screening high-risk individuals with low-dose chest CT has been shown to reduce lung cancer mortality. However, early screening trials have reported high false-positive rates, leading to unnecessary follow-up procedures, increased patient anxiety and health care costs.
Pulmonary nodules—small round or oval growths in the lungs—are common, and identifying which are malignant is challenging in lung cancer screening.
“Deep learning offers promising solutions, but robust validation is essential,” said Noa Antonissen, M.D., lead researcher and Ph.D. candidate at Radboud University Medical Center, Nijmegen, the Netherlands. “AI accounts for factors that we might not even see on the CT scan to further assess a nodule as likely to be malignant.”
Most current lung cancer screening protocols rely on nodule size, type and growth to estimate malignancy risk. The Pan-Canadian Early Detection of Lung Cancer (PanCan) model, which estimates nodule malignancy risk through a combination of patient and nodule characteristics, illustrates how probability-based tools can refine risk assessment. Such probability-based risk thresholds are increasingly used to guide management protocols. Deep learning offers a promising alternative by enabling fully data-driven predictions, but more evidence is needed before it can be adopted in clinical practice.
In the retrospective study, the researchers trained their in-house developed deep learning algorithm to estimate the risk for malignancy for lung nodules using data from the National Lung Screening Trial which included 16,077 nodules (1,249 malignant).
External testing was conducted using baseline CT scans from the Danish Lung Cancer Screening Trial, the Multicentric Italian Lung Detection trial and the Dutch–Belgian NELSON trial. The pooled cohort included 4,146 participants (median age 58 years, 78% male, median smoking history 38 pack-years) with 7,614 benign and 180 malignant nodules.
The researchers assessed the algorithm’s performance for the pooled cohort and two subsets: indeterminate nodules (5-15 mm) and malignant nodules that were size-matched to benign nodules.
“We selected nodules sized 5–15 mm, due to their diagnostic challenges and frequent need for short-term follow-up,” Dr. Antonissen said. “Accurate risk classification of these nodules could reduce unnecessary procedures.”
For comparison, the algorithm’s performance was evaluated against the PanCan model at nodule and participant levels using the area under the receiver operating characteristic curve (AUC), among other parameters. AUC sums up how well a model can produce relative scores to discriminate between positive or negative instances across all classification thresholds.
In the pooled cohort, the deep learning model achieved AUCs of 0.98, 0.96, and 0.94 for cancers diagnosed within one year, two years, and throughout screening, respectively, compared to PanCan at 0.98, 0.94, and 0.93.
For indeterminate nodules (129 malignant, 2,086 benign), the deep learning model significantly outperformed PanCan across all timeframes with AUCs of 0.95, 0.94, 0.90 vs. 0.91, 0.88, 0.86. For the cancers size-matched to benign nodules, (180 malignant, 360 benign), the deep learning model’s AUC was 0.79 versus PanCan at 0.60.
At 100% sensitivity for cancers diagnosed within 1 year, the deep learning model classified 68.1% of benign cases as low risk compared to 47.4% using the PanCan model, representing a 39.4% relative reduction in false positives.
“Deep learning algorithms can assist radiologists in deciding whether follow-up imaging is needed, but prospective validation is required to determine the clinical applicability of these tools and to guide their implementation in practice,” Dr. Antonissen said. “Reducing false positive results will make lung cancer screening more feasible."
###
“External Test of a Deep Learning Algorithm for Pulmonary Nodule Malignancy Risk Stratification Using European Screening Data.” Collaborating with Dr. Antonissen, were Kiran Vaidhya Venkadesh, Ph.D., Renate Dinnessen, M.Sc., Ernst Th. Scholten, M.D., Ph.D., Zaigham Saghir, M.D., Ph.D., Mario Silva, M.D., Ph.D., Ugo Pastorino, M.D., Grigory Sidorenkov, Ph.D., Marjolein A. Heuvelmans, M.D., Ph.D., Geertruida H. de Bock, Ph.D., Firdaus A. A. Mohamed Hoesein, M.D., Ph.D., Pim A. de Jong, M.D., Ph.D., Harry J. M. Groen, M.D., Ph.D., Rozemarijn Vliegenthart, M.D., Ph.D., Hester A Gietema, M.D., Ph.D., Mathias Prokop, M.D., Ph.D., Cornelia Schaefer-Prokop, M.D., Ph.D., and Colin Jacobs, Ph.D., for the NELSON-POP consortium. This study received funding from the Dutch Cancer Society (KWF Kankerbestrijding, project number 9037) and Siemens Healthineers.
Radiology is edited by co-interim editors Vicky Goh, M.B.B.Ch., King’s College London, U.K., and Kathryn Fowler, M.D., University of California San Diego, California, and owned and published by the Radiological Society of North America, Inc. (https://pubs.rsna.org/journal/radiology)
RSNA is an association of radiologists, radiation oncologists, medical physicists and related scientists promoting excellence in patient care and health care delivery through education, research and technologic innovation. The Society is based in Oak Brook, Illinois. (RSNA.org)
For patient-friendly information on lung cancer screening, visit RadiologyInfo.org.
END
ITHACA, NY—A new study published today in the journal Biological Conservation finds that different communication approaches can influence whether people take action to prevent birds from colliding with windows, a leading cause of bird mortality that kills over 1 billion birds annually in the United States and Canada.
Researchers from the Cornell Lab of Ornithology tested different message framing approaches among bird enthusiasts and the general public to understand what motivates people to make their windows safer for birds. “If we want people to take action to reduce bird and window collisions, we really have to understand how to communicate with ...
SAN ANTONIO — September 16, 2025 — Southwest Research Institute (SwRI) and The University of Texas at San Antonio (UT San Antonio) are collaborating to create a detection system to identify pre-ignition in hydrogen internal combustion engines (H2-ICE). Researchers will combine machine learning (ML) algorithms and artificial intelligence with onboard sensors to help detect pre-ignitions based on their tell-tale signs.
Pre-ignition occurs when unprompted combustion happens inside an engine before the prescribed spark timing. These abnormal, uncontrolled ...
Ammonia is used in fertilizer and many industrial processes. It is also seen as a promising way to store and transport energy, as it is safer and easier to handle ammonia than hydrogen gas. Using plasma, the fourth state of matter, scientists have created a material that boosts ammonia production.
“If one needs industrial hydrogen someplace else than where it is made, it will be easier and safer to transport hydrogen as ammonia and store it until it is needed. Ideally then one would decompose the ...
The International Data Corporation projects that artificial intelligence will add $19.9 trillion to the global economy by 2030, yet educators are still defining how students should learn to use the technology responsibly.
To better equip AI educators and to foster a sense of community among those in the field, Kennesaw State University Department Chair and Professor of Information Technology (IT) Shaoen Wu, along with assistant professors Seyedamin Pouriyeh and Chloe “Yixin” Xie, were recently awarded two National Science Foundation ...
Cardiac arrest happens when the heart stops pumping blood properly, cutting off oxygen to the brain and other vital organs. In these emergencies, cardiopulmonary resuscitation (CPR) can help keep blood and oxygen moving until medical help arrives. For children, CPR usually needs both chest compressions and rescue breaths, because many cases are caused by breathing problems such as drowning, choking, or serious illness.
During the COVID-19 pandemic, fear of infection changed the way CPR was practiced. Public health guidelines for adults recommended chest compression-only CPR (CO-CPR) to minimize the risk of ...
In markets where producers drive the creative process, high-status producers can opt for more radical changes for their artists’ image, whereas middle-status producers are likely to shift their artists’ image to follow popular categories. The new research, published in Strategic Management Journal, uses Korean pop music — or K-pop — to demonstrate how the status of entertainment agencies affects how idol groups shift categories, an effect that is also limited by the artists’ gender.
The research — from Heeyon Kim of Cornell University, Yoonjeoung Heo of Xi'an Jiaotong-Liverpool University, and Chi-Nien Chung of Hong ...
Dr. Juchan Yang’s research team at the Hydrogen & Battery Materials Center, from the Energy & Environment Materials Research Division of the Korea Institute of Materials Science (KIMS), has developed a composite catalyst using the novel material MXene that suppresses the generation of chloride ions-one of the key challenges in seawater electrolysis. This research outcome is expected to accelerate the practical application of seawater electrolysis technology by enabling stable hydrogen production even in seawater.
Hydrogen is gaining attention as an eco-friendly energy source that emits no carbon. However, conventional water electrolysis ...
Gene therapy offers the possibility of addressing cancer at its molecular roots by targeting disease-causing genes rather than relying solely on surgery, chemotherapy, or radiotherapy. Among RNA-based strategies, microRNA (miRNA) sponges and short-hairpin RNAs (shRNAs) have emerged as promising tools to silence oncogenes and restore tumor-suppressor pathways. However, clinical application remains constrained by delivery inefficiency, instability, and off-target toxicity. Viral and non-viral vectors each present strengths and limitations: while viral vectors provide high transfection efficiency, they often cause immune responses and mutagenesis risks; non-viral ...
In November 2016, India’s government abruptly invalidated its two highest-value banknotes, wiping out about 86 percent of the nation’s cash supply overnight. Known as the Great Indian Demonetization, the move was intended to curb corruption and encourage the adoption of digital payments.
The sudden shift sparked chaos. Long lines formed outside banks as people scrambled to exchange their money before it became worthless. Small business owners and street vendors, many of whom relied almost ...
An estimated 7.2 million Americans over age 65 currently live with Alzheimer’s disease (AD). That number is expected to nearly double to 13.8 million by 2060. These increases reflect more than demographic shifts; they point to a growing public health crisis that requires a new, proactive approach. While chronological age is the strongest known risk factor for cognitive decline, losing cognitive function is not an inevitable part of aging.
As AD and other forms of cognitive decline continue to rise at an alarming rate, researchers from Florida Atlantic University’s Charles E. Schmidt College of Medicine, highlight a powerful ...
