AI Models Improve Survival Predictions for Lung Radiation Patients
Researchers have developed a machine-learning approach that combines medical imaging data with clinical records to better predict survival outcomes for patients undergoing stereotactic body radiotherapy for lung tumors.
Key points
- Machine learning models improved survival prediction accuracy by combining clinical data with CT scan features.
- The study involved 158 patients across six different medical centers in France and Canada.
- Predictive performance was strongest for one-year survival but declined for three-year outcomes.
A multicenter study published in Nature Health Sciences suggests that machine-learning models incorporating imaging data can more accurately predict survival outcomes for lung cancer patients receiving stereotactic body radiotherapy (SBRT). Researchers from institutions including Institut Godinot and Polytechnique Montreal analyzed data from 158 patients across six centers to determine if radiomics—the extraction of large amounts of data from medical images—could enhance traditional clinical predictions.
The research team developed models using 180 distinct features, including clinical history, radiation dosage metrics, and 124 radiomic markers extracted from planning CT scans. By combining these data points, the researchers found that the integrated model outperformed those relying solely on clinical information. Specifically, the addition of imaging and dosage data increased the model's discriminative power by a margin of 0.12, according to the concordance index.
The predictive accuracy was most effective at the one-year mark following treatment, achieving a time-dependent area under the curve (AUC) of 0.69. However, the models were less reliable for longer-term three-year predictions and struggled to accurately forecast secondary outcomes like cancer recurrence or the spread of the disease. The study authors noted that while these results are promising, the findings are based on a retrospective analysis of a relatively small patient group and require further validation in larger, external populations.
What to know: This study from Nature Health Sciences highlights the potential for AI to assist oncologists in personalizing follow-up care, though it is not yet a tool for clinical decision-making. Patients should consult their medical team to discuss individual treatment plans and prognosis.
Source Nature Health Sciences — nature.com/articles/s41598-026-77575-6 ↗
Read next
🏥Public HealthPreventive Services Task Force Resumes Meetings Under New Leadership
The U.S. Preventive Services Task Force is set to reconvene this week for the first time in over a year following a significant membership overhaul by Health Secretary Ro…
❤️Conditions & CareBlood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma
Researchers have identified a specific RNA signature in blood plasma that may help predict how well lymphoma patients respond to CAR T-cell therapy.
🧠Mind & Mental HealthAddressing Body Image and the Reality of Gluteal Anatomy
A personal reflection on the social and psychological impact of body shape, specifically regarding the gluteal region, and the limitations of consumer solutions for perce…
❤️Conditions & CareMigraine and Smoking Linked to Increased Early Stroke Risk in Women
A large-scale study suggests that people who have given birth and experience migraines while also smoking face a significantly higher risk of stroke before age 60.
🔬ResearchCBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities
A double-blind clinical trial published in JAMA Pediatrics indicates that CBD oil provides no significant behavioral improvement for children and teenagers with intellect…
🏃FitnessExpert Advice on Hydration Gear Sparks Debate Among Recreational Runners
Olympic medalist Josh Kerr recently suggested that runners may not require hydration packs for distances under 12 miles, igniting a broader conversation about gear necess…