COPD is the seventh leading cause of death in Korea. This is higher than the traffic accident (No. 10), and the number of patients continues to increase due to air pollution and aging, including fine dust. In particular, the World Health Organization (WHO) identified COPD as the fourth leading cause of death and predicted that it would rise to the third in 2030.

However, ‘chronic obstructive pulmonary disease’, which is less well-known than asthma, is a chronic respiratory disease that most people are missing out on to understand whether they have their own disease until there are no significant symptoms.

According to data from the Korea Health Insurance Review and Assessment Service, there were only 1.14 million patients who visited hospitals with asthma in 2018, while only 190,000 patients with COPD treatment. It is estimated that there are currently 3 million domestic COPD patients, but the diagnosis rate is only 2.8%. Academia estimates that 14% of the population aged 40 years or older suffer from COPD.

At the same time, the risk of lung cancer is increasing due to the factors causing chronic obstructive pulmonary disease (COPD). The National Health Insurance Corporation’s National Sampling Cohort was a follow-up of 33,548 patients aged 40 to 84 who had no history of developing lung cancer.

On April 27, the research team of Samsung Medical Center O-Jung Kwon, Professor Hye-yoon Kwon, Professor Joo-hee Jo and Kang Dan-bi of the Center for Clinical Epidemiology announced that they have published the paper in a recent issue of Thorax (IF=10.307).

According to the research team, 1834 cases of lung cancer were identified during follow-up, and the risk of developing lung cancer in COPD patients was 3.12 times higher than in non-COPD patients.

Among them, a Korean medical imaging solution development company unveiled the world’s first AVIEW COPD, which automates various image analysis algorithms required to diagnose COPD using artificial intelligence (AI) technology.

This is the first solution presented at RSNA 2018, the largest radiological society in North America, and is characterized by fully automating the segmentation of the bronchial and lung lobes, which have been a challenge for a long time.

In particular, as a result of comparing the results of 200 cases processed by the research team with 7-year-experienced experts, the accuracy of the final quantitative index analysis reached 96%. On the other hand, the total time including rework was 13man-hour, which was only 6%.

COPD requires subtype classification through CT for effective treatment even after diagnosis, and the treatment effect is periodically monitored through CT examination. However, in order to quantitatively read through CT images, there was a difficulty in taking a long time for pre-processing and analysis, so there was a limit to staying in qualitative reading depending on the experience of reading.

In this situation, a solution that automates the image analysis algorithm of COPD diagnosis is causing changes in the diagnosis method and environment in the future. Currently, A-View CPD has been confirmed by National Taiwan University Hospital, and has been in trial operation at Leuven Hospital in Belgium, Posh Hospital in France, and University Hospital in Hokkaido, Japan.

As the majority of deaths from corona19 are known to have underlying diseases, attention and interest in the vulnerable population is increasing. In particular, coronavirus is an important cause of exacerbation of respiratory diseases, and the medical community emphasizes prevention of infection among patients with chronic respiratory diseases such as asthma and chronic obstructive pulmonary disease (COPD).

CORELINE SOFT, which developed AVIEW COPD, developed the national lung cancer screening software and created the first case in which the lung cancer screening process was provided in the cloud and verified on a large scale worldwide. This is the reason why AVIEW COPD released in such an environment is expected to grow and spread.

/ Reporter Seo Jae-chang (prmoed@hellot.net)