2 weeks Ago By Korea IT Times
August Contributors: We would like to express our gratitude to the July contributors as we celebrate the 20th anniversary of Korea IT Times. We are now excited to publish the new contributor's article for August . By Jinkook Kim, PhD, CEO of Coreline Soft In May 2024, the German government published an ordinance in the “Bundesgesetzblatt.
” This ordinance regulates low-dose CT scans for the early detection of lung cancer in smokers. Starting in July 2024, hospitals that want to implement lung cancer screening will have to comply with these regulations. The publication of this ordinance marks the beginning of the rollout of lung cancer screening across Germany, rather than an increase in regulation.
Lung cancer is the leading cause of cancer-related deaths worldwide, and the risks are increasing daily. According to the National Cancer Center, the five-year relative survival rate is 25.1%, but it rises to 78.
5% if the cancer is detected at a localized stage, meaning it has not spread beyond the lungs. Despite the importance of early detection, only 24.6% of lung cancer cases are detected early, which is significantly lower than other cancers.
This underscores the critical need for national programs promoting lung cancer screening, as early-stage lung cancer often presents no symptoms. Attempts to screen for lung cancer using chest X-rays and sputum tests have not effectively reduced mortality rates. However, a randomized controlled trial (RCT) in the United States, the NLST, involved over 50,000 participants and demonstrated in 2011 that low-dose CT screening could reduce lung cancer mortality by more than 20% compared to the control group.
Similarly, in Europe, the NELSON trial showed a 26% reduction in lung cancer deaths among high-risk men after 10 years of follow-up, with results published in 2020. These studies have established low-dose CT scans as the gold standard for early lung cancer detection. This evidence led to the launch of national lung cancer screening programs in the United States in 2015, South Korea in 2019, and Taiwan and the United Kingdom in 2022.
With this ordinance, Germany is now joining these efforts. In my view, the German ordinance is innovative compared to the initial approaches by the U.S.
, South Korea, Taiwan, and the UK. To summarize, the ordinance mandates a second reading if significant findings arise in the first reading, and it requires the use of AI in all screenings. By having second reads performed by hospitals with high expertise in treating lung cancer, the ordinance aims to minimize unnecessary high-risk follow-up tests while preventing missed cases of early-stage lung cancer.
To handle the potential increase in workload, the program mandates AI use in all screenings, which should reduce workload and variation between readers, ensuring consistent quality across the country. Given the high expectations for medical AI and its slow adoption in clinical practice, I see Germany's decision as a bold and innovative move, one that I am confident will succeed. This confidence is supported by the HANSE study, centered at the University Hospital of Hannover, which over the past three years has successfully demonstrated AI-enabled lung cancer screening workflows.
The proven methodology from this study has been incorporated into the ordinance. The HANSE study utilized my company's AI product, AVIEW LCS, which improved the accuracy and efficiency of readings and supported the multidisciplinary team consultation process necessary for second readings. The multidisciplinary team used AI-provided analytics to reduce unnecessary high-risk biopsies, with less than 1% of biopsies returning negative, a significant improvement over the 27% in the original NELSON study.
Despite this experience, Germany's decision reflects the scale of a national lung cancer screening initiative, which entails conducting hundreds of thousands to millions of CT scans annually. With increasingly high case numbers in Korea, it is necessary to involve not only chest radiologists but also those from other specialties. As various hospitals participate, inter-reader variability can increase, making consistent control essential.
Introducing standardized reading systems like Lung RADS has mitigated some challenges but has also increased workload. AI introduction is one proposed solution. AI can identify small nodules within the complex structure of the lungs on a CT image and automatically measure their volume.
If a previous CT scan is available, AI can compare scans to analyze nodule growth rate and categorize it according to standardized criteria, aiding reader decision-making. My company's AVIEW LCS uses AI to enhance accuracy and efficiency, significantly reducing inter-reader variation. It has received positive user ratings and has been recognized for its necessity and performance in Korea, where it is used in over 100 hospitals nationwide through a cloud-based system managed by the National Cancer Center's quality management project.
When South Korea launched its national lung cancer screening program in 2019, it incorporated AI and established an innovative system to monitor screening quality across the country via the cloud. However, regulatory and funding challenges have limited the program's coverage. A new revolution in lung cancer screening with AI is underway.
If successful in Germany, it will significantly impact not only other European countries but also the United States, South Korea, and Taiwan, which have already introduced lung cancer screening programs. I believe that AI-integrated lung cancer screening will be successful and establish a new global paradigm for cancer prevention and management. About the Author Jinkook Kim, PhD, is the founder and CEO of CorelineSoft and an expert in the field of electronics and imaging systems.
Founded in 2012, CorelineSoft is leading the development of AI-based medical imaging solutions, providing world-class technology in the field of lung imaging. Its flagship product, AVIEW LCS, utilizes AI to detect early-stage lung nodules in low-dose chest CT scans, achieving a breakthrough in medical diagnosis. Dr.
Kim received a B.S. in Electronics and Communication Engineering from Hanyang University and a Ph.
D. from KAIST with research interests in electronics, medical imaging, and 3D visualization. CorelineSoft is strengthening its presence in the North American and European markets and expanding its global reach through partnerships with leading hospitals and experts.
.
