A highly effective artificial intelligence (AI) tool has demonstrated superior performance in screening for lung cancer compared to humans. The tool, powered by advanced machine learning, shows enhanced sensitivity, reducing the likelihood of overlooking cases that demand urgent intervention.
The tool enhances radiologists’ tumor detection on chest X-rays and promotes greater acceptance of AI in healthcare. Recent studies highlight the significant assistance AI provides to doctors in medical diagnoses.
The study’s lead author, Dr. Chang Min Park, suggests that the large sample size in the study may have increased readers’ confidence in the AI’s suggestions. The study observes that humans are more likely to trust AI with high diagnostic performance, highlighting the issue of human trust in AI.
According to the study, there was a significant increase in the successful identification of cancerous lung nodules on chest X-rays, reaching 94% with the use of computer data. While these growths usually result from infections, they can occasionally signal the presence of cancer.
The study involved 20 experienced thoracic radiologists and 10 students analyzing 120 chest X-rays without AI assistance. Half of the X-rays were from male lung cancer patients from South Korea, and the other half were controls. In a follow-up session, the groups re-evaluated the X-rays with either high- or low-accuracy AI. The study emphasizes the importance of using high diagnostic performance AI, but acknowledges that definitions may vary based on the task and clinical context.
The practicality of an AI model capable of identifying every abnormality may be limited in reducing workload during mass screening. Dr. Park’s study proposes that effective use of AI in clinical settings necessitates both high-performance AI models for specific tasks and consideration of the relevant clinical context in which the AI will be utilized.
Researchers plan to extend their work on human-AI collaboration to other abnormalities detected on chest X-rays and CT images. Lung cancer is often diagnosed too late due to the lack of initial symptoms. Previous studies have demonstrated AI’s ability to predict colon diseases, dementia, Alzheimer’s, and heart attacks.