고려대학교 의과대학 및 의과학연구지원센터 구성원이 아닌 경우에는 본인인증을 통하여 글을 작성할 수 있습니다.
- 본인인증 후 글을 작성할 수 있습니다.
내 명의의 휴대전화로 비밀번호를 재설정 할 수 있습니다.
Korea University Ansan Hospital Research Team Led by Professors Ki-Sun Lee, Won Suk Choi, and Ki Yeol Lee Developed Deep Learning-based Algorithm that Assists COVID-19 Diagnosis based on Chest X-ray
<(From Left) Professor Ki-Sun Lee of Medical Science Research Center,Won Suk Choi Professor of Department of Infectious Diseases, and Professor Ki Yeol Lee of Department of Radiology>
A Korea University Ansan Hospital research team led by professor Ki-Sun Lee of the Precision Medicine Center (headed by Nan Hee Kim) also serving as a professor of dentistry at the Medical Science Research Center, Won Suk Choi (Department of Infectious Diseases), and Ki Yeol Lee (Department of Radiology) came up with an algorithm that would help diagnose COVID-19 using a technology based on deep learning combined with chest X-ray.
The deep learning-based algorithm developed by the research team can assess the chest X-ray to classify its finding into three categories: normal, pneumonia and COVID-19. Taking advantage of explainable deep learning, this algorithm is able to distinguish COVID-19-driven pneumonia from the pneumonia resulting from other causes to demonstrate differentiation accuracy as high as 95%.
<Deep learning-based decision map for healthy lung, lung with pneumonia or COVID-19>
The multi-disciplinary research team consists of professor Ki-Sun Lee of the KU Ansan Hospital Precision Medicine Center, professor Jae Young Kim and researcher Eun-tae Jeon of the Medical Science Research Center, professor Won Suk Choi of the Department of Infectious Diseases, and professor Ki Yeol Lee of the Department of Radiology. The study is published in the Journal of Personalized Medicine, one of the top 10% journal in the field of JCR Health Care Sciences (LNIK:https://doi.org/10.3390/jpm10040213).
The first author of the paper, professor Ki-Sun Lee who developed the artificial intelligence program said, "There are many other researches already done by medical AI companies or other organizations on the development of chest X-ray-based algorithms to help better diagnose COVID-19. This means that our KU Ansan Hospital is not the pioneer for this. However, we have made our basic research data open to others, which can help improve the accuracy of deep learning and identify the reason for any possible classification errors. So we hope that our study could be of help to other researches which are to be launched or currently under way."
Professor Lee is a former software developer at Samsung SDS and dentist. Early this year, he developed an osteoporosis screening algorithm based on deep learning using dental X-ray.
Professor Ki-Sun Lee also said that the study team plans to further improve the algorithm and collaborate with other countries or medical organizations which do not have enough radiologists or medical resources so that they can take the full advantage of it in practical use.
The research was done with the support from Ansan Hospital and National Research Foundation of Korea's Basic Research in Science & Engineering.
The 3rd Korea University International Medical Student Research Conference
On December 22, 2020, Korea University College of Medicine (KUCM) successfully held the third Korea University International Medical Student Research Conference.
KUCM offers various programs in an attempt to help its students better prepared for the rapid changes in the medical environment with the fourth industrial revolution under way. Student-led research activities are part of such schemes launched in 2010 in an effort to expose students to medical researches so that they can grow into motivated researchers.
With the support of the designated professors, students are encouraged to be engaged in various research activities for one year and the conference held in November offer them a chance to share the outcome of their studies. KUCM has expanded the event to invite participants from abroad and began in 2018 to host the Korea University International Medical Student Research Conference, the first of its kind in Korea. Consisting of 22 research teams of 49 students, the 12th student research organization was launched in December 2019.
The year 2020 saw its conference held on-line due to the COVID-19 pandemic and it was attended by more than 150 medical students and professors of different countries. The event was preceded by registration and poster presentation. On the date of the event, an opening speech was delivered by Young-Wook Yoon, dean of College of Medicine, which was followed by a keynote lecture by professor Gou Young Koh of the KAIST Institute for Basic Science (IBS) Center for Vascular Research. Conference also had three-minute speech sessions, and oral presentations made by students from home (KUCM) and abroad. And then, there came the session to present rewards and all the event came to close with a closing remark.
KUCM contributed to the event by offering 22 poster presentations and 11 three-minute speeches. From abroad, six teams consisting of 12 students from the University of Munich, Nagoya University, Sapporo University, National University of Singapore, Chinese University of Hong Kong, and University of Bologna joined the event.
A freshman of KUCM who participated in the event for the first time said, “It is exciting that they conducted researches on such various themes. It definitely inspired me. It was also very meaningful in that we could meet and discuss with students of other countries. Such chances are very rare and valuable. I now plan to design and develop a research project of my own. I'd love to join the event next year again and win a prize too.”
Young-Wook Yoon, dean of KUCM said, “I am very happy to see the outcome of such promising researches, which are sure to develop further in the future. This conference offers a valuable chance for participants to share their ideas and it is sure to add value to their career path. I want to thank the students and professors for taking time out of their busy schedule to be engaged in researches and hope that the students grow into good medical scientists.”
Research Team Led by Professor Seong-Mi Park of Department of Cardiology Proves Young People with Hypertension, Diabetes, Cardiovascular Diseases are More Vulnerable to Fatal Outcome in COVID-19 Infection
Meta-analysis of 51 Studies on COVID-19 by Korea University Anam Hospital Research Team of Department of Cardiology Published in an International Journal
A Korea University Anam Hospital research team led by professors Seong-Mi Park and SungA Bae of the Department of Cardiology reported the relative risk of fatal outcome in young patients with any of cardiovascular risk factors or heart disease was much greater than expected.
It has been reported that the major risk factors for fatal outcome of severe COVID-19 and death in COVID-19 include age greater than 60 years, male sex, and the presence of comorbidities. A number of studies reported the elderly patients are vulnerable to fatal outcome in COVID-19 infection as they are more likely to have underlying diseases such as hypertension, diabetes, or cardiovascular disease whose prevalence tends to increase with age. However, there have not been many studies done on the risk of the young population with cardiovascular risk factors in relation to the clinical outcome of COVID-19. Therefore, we conducted a systematic review and comparison to see the risk of severity and death from COVID-19 in different age groups.
A meta-analysis was performed on 51 studies collected from PubMed and Embase databases with a total of 48,317 patients with confirmed COVID-19 infection. Overall, the relative risk of developing severe COVID-19 or death was significantly higher in patients with hypertension, diabetes, or CVD across all age groups. Although young patients had lower prevalence rates of cardiovascular comorbidities than elderly patients, the relative risk of fatal outcome in young patients with hypertension, diabetes, and CVD was more than two times higher than in elderly patients.
The findings of this study demonstrated the correlation between COVID-19 infection and cardiovascular risk factors and heart diseases among different age groups. It showed that special attention needs to be paid to those young COVID-19 patients with underlying cardiovascular risk factors as they are also at risk of developing severe COVID-19 or death.
Professor Seong-Mi Park said, ”This study showed that not only the elderly but also the young people in COVID-19 infection are very vulnerable to fatal outcomes such as need of ICU care and ventilator or death when they have underlying cardiovascular risk factors" For years, professor Park has studied hypertension and metabolic syndrome in young patients and has demonstrated they have subclinical but considerable myocardial damage even in young "Serum Aldosterone Is Related to Left Ventricular Geometry and Function in Young Adults with Never-Treated Primary Hypertension" J Clin Med. 2019;8:1045, and "Association between epicardial adipose tissue, high-sensitivity C-reactive protein and myocardial dysfunction in middle-aged men with suspected metabolic syndrome" Cardiovascular Diabetology 2018;17:95 and more). Young patients tend to pay less attention to manage their medical illness even they have myocardial and vascular damage, which she predicted would leave them very vulnerable to fatal outcome of COVID-19 infection.
Professor SungA Bae added, “I hope that this study would help young people appreciate the importance of early diagnosis and management of metabolic syndrome, hypertension, diabetes and other chronic diseases, which are increasing among the young and closely associated with obesity and lack of physical exercise.”
The study titled, "Impact of cardiovascular disease and risk factors on fatal outcomes in patients with COVID-19 according to age: a systematic review and meta-analysis" is published in the internationally renowned Heart (2020 Dec 17:heartjnl-2020-317901. doi: 10.1136/heartjnl-2020-317901. Online ahead of print.).
“Early Detection of Lung Cancer Made Possible with Nano Technology”
One drop of blood leads to diagnosis of lung cancer
AI assesses exosomes in blood
Stage 1 lung cancer can be detected
Study published in ACS Nano
One drop of blood is enough for the diagnosis of lung cancer and it can be done in about half an hour. Such a new approach was developed by a Korean research team and it is published in May issue of ACS Nano (IF:14.5), a renowned international journal in chemistry and nano technology.
A joint research team led by professor Yeonho Choi of biomedical engineering at Korea University and professor Hyun Koo Kim of thoracic surgery at Korea University Guro Hospital developed a model based on nano technology and artificial intelligence, which can classify exosomes derived from normal and lung cancer cell lines with an accuracy of 95%. Exosomes, nanosized extracellular vesicles found in blood are used as a biomarker for cancer diagnosis.
This technology would allow even patients with stage 1 lung cancer, who have had difficulty in early diagnosis, to be checked for lung cancer in about 30 minutes with a drop of blood. This is sure to make significant contribution to much better survival of the lung cancer patients.
▲ From left to right: professor Yeonho Choi (corresponding author, Department of Biomedical Engineering at Korea University), professor Hyun Koo Kim (corresponding author, Department of Thoracic Surgery at Korea University Guro Hospital), and Hyunku Shin (first author, Department of Bio-convergence Engineering at Korea University)
Blood sample has already been used to diagnose lung cancer, but it could detect lung cancer in only about 50% of the patients. The new approach improved it up to 84%. It can also predict how the tumors would progress.
Many of the lung cancer patients are diagnosed only after they have advanced to stage 3 or above, making it difficult to get curative treatment. As one of the cancers with a high mortality, its early detection can improve the survival significantly. It is no wonder there have been a number of active researches conducted in order to find ways to detect lung cancer in early stages, and exosomes, nanosized extracellular vesicles found in blood, have been proposed as promising biomarkers for liquid biopsy.
The researchers isolated exosomes from the cell culture of 20 healthy subjects and 43 patients with stages 1 and 2 of non-small cell lung cancer. They detected more than 2,000 Raman spectroscopy signals using nanotechnology based on surface-enhanced Raman spectroscopy. Using the signal, the researchers trained a deep learning-based artificial intelligence model and succeeded in sorting healthy cells and lung cancer cell exosomes with 95 percent accuracy. They also succeeded in classifying lung cancer patients' exosomes to about 84-percent sensitivity and 85-percent specificity compared to exosomes derived from lung cancer cells.
Dr. Choi said, “This research proved the effectiveness of the diagnostic approach using the exosome assessment and deep-learning artificial intelligence in early stage lung cancer.” He explained, “It can not only detect stage 1 lung cancer but also diagnose lung cancers in the later stages with much better accuracy. This means that it may even be possible to use it as a test to monitor tumors for changes that indicate when they are progressing.”
“By using this technology, hospitals can conduct a blood test to pre-screen patients who are likely to have lung cancer and perform CT tests only when necessary as such tests may cause radiation exposure,” Professor Kim said. “The new method is significant because it can detect stage 1 lung cancer patients with relatively good accuracy. This early detection is expected to result in improved survival of those patients diagnosed with lung cancer.”
The study, Early-Stage Lung Cancer Diagnosis by Deep Learning-Based Spectroscopic Analysis of Circulating Exosomes, was carried out as part of the basic research project strategically implemented by the National Research Foundation of Korea and sponsored by the Korean Health Industry Development Institute.
In the research, technology developed by Exopert, a medical technology holding company of Korea University, for which professor Yeonho Choi serves as its CEO was used. Exopert is dedicated to the development of kits which can quickly isolate exosomes in high purity from the blood sample of cancer patients plus cancer diagnostic tools using exosomes. It plans to launch a multi-center study involving five hospitals including the Korea University Guro Hospital in an attempt to demonstrate the reliability of its technology and facilitate its commercialization. About 400 healthy participants and patients will be recruited for the study.