Experts say AI-assisted diagnosis of pulmonary nodules is too early to completely replace doctors’ judgment

China News Service, Beijing, September 10 (Reporter Zhao Fangyuan) "AI-assisted diagnosis is indeed making progress, but it is still too early to use it to completely replace doctors' judgment." At the 16th Chest Imaging Academic Conference held in Yan'an recently, Hou Delun, director of the Medical Imaging Center of Beijing Chest Hospital, said that AI currently has high accuracy in screening pulmonary nodules, especially in terms of follow-up analysis of growth status such as size, shape, and density of pulmonary nodules. It is already the most mature AI application scenario.

However, he also emphasized that artificial intelligence is still an auxiliary diagnostic tool in the field of pulmonary nodule screening, and the final accurate diagnostic conclusion must be combined with manual judgment. Radiologists must take the initiative to learn and master the interactive use of artificial intelligence, and conduct big data research with AI in their daily work to improve diagnosis efficiency and service quality.

The 16th Chest Imaging Academic Conference and Chest Imaging Diagnosis and Intervention Advances Study Class is hosted by Beijing Chest Hospital Affiliated to Capital Medical University and is open to doctors, researchers and graduate students in imaging departments, tuberculosis departments, respiratory departments, thoracic surgery departments, infectious diseases departments, oncology departments and other related professions at all levels of medical institutions.

Li Liang, deputy secretary of the Party Committee and President of Beijing Chest Hospital, pointed out that the current role of imaging diagnosis in the field of chest diseases continues to increase, and the industry needs to focus on the extended development of chest imaging, especially the application of artificial intelligence. Imaging practitioners should think and explore how to organically integrate AI technology with work, and how to combine AI with effective diagnosis.

Peng Ning, deputy secretary-general of the Shaanxi Provincial Medical Association, also said that chest imaging is a key support for accurate diagnosis and treatment of respiratory and circulatory system diseases, and plays an irreplaceable role in early screening of lung cancer and other scenarios. In recent years, my country's chest imaging and artificial intelligence technologies have been deeply integrated, and the diagnosis and treatment system has been continuously improved, building a solid line of defense for the prevention and control of major diseases.

Industry insiders pointed out that in grassroots areas where high-quality medical resources are relatively weak, AI-assisted diagnosis is expected to become an important starting point to improve the level of grassroots imaging diagnosis. But the premise is that grassroots doctors need to learn to use AI tools correctly and have the final diagnostic ability. (over)