한빛사 인터뷰
- 등록2023-05-15
- 조회918
1.1 Can you please briefly summarize the paper?
In this manuscript, we used machine-learning to identify any medication classes that might be associated with improved outcomes in patients with monoclonal gammopathy of undetermined significance (MGUS). Specifically, we used XGBoost to perform survival analysis on a large, retrospective cohort of MGUS patients who had medication data available. We then used bootstrapping and examined Shapley values to determine which medication classes were significantly associated with longer MGUS time to disease progression. Our study found that multivitamins, immunosuppression, non-coronary NSAIDS, proton pump inhibitors, vitamin D supplementation, opioids, statins and beta-blockers, were associated with a lower hazard ratio for MGUS progression.
1.2 Can you please tell us the main difficulties you had in the laboratory work and how you overcame them?
The main challenges for this study were obtaining sufficient data with documented outcomes for MGUS patients, and obtaining quality medications data. Fortunately, Dr. Kumar (co-author) maintains a large and complete database of MGUS patients. Our medication data was unfortunately limited to only the presence or absence of medication classes, as we did not have reliable information on medication doses or durations. Merging electronic health record data with insurance claims data might have given us more complete medication data.
2. Please introduce your laboratory, university, or organization to bio-researchers in Korea.
Mayo Clinic is a large, academic, tertiary referral center and the #1 Hospital in the United States, as ranked by USNews and World Report. Mayo Clinic has three main campuses across the US, along with several smaller regional sites, and several international sites, as well. Mayo Clinic was founded in 1864 and has an affiliated medical school, the Mayo Clinic Alix School of Medicine.
3. Please tell us your experiences and your thoughts related to research activities abroad.
As I am early in my career, I have had limited opportunities to collaborate on research abroad, but have always admired the Korean ethos of innovation and rising through challenges, and would be eager to collaborate with Korean researchers or institutions.
4. Can you provide some advice for younger scientists who have plans to study abroad?
- Find a mentor who cares about your professional success and personal wellbeing
- Say “yes” to projects broadly, at least initially, and establish a reputation for excellence. This will lead to more opportunities coming your way.
- Find a community and support system to help you navigate a new environment
- Share your culture and heritage with your colleagues, who will most likely be very interested to learn more!
5. Future plan?
We intend to continue similar work applying machine-learning to various disease states, both for drug repurposing hypotheses and for better outcomes predictions. Ideally, this work could inform prospective studies and potentially decision-support tools in our electronic health records.
6. How do you think this research will benefit the tomato farming or agricultural industry in Korea?...
While this research was focused more on machine-learning and human healthcare, similar machine-learning methods could be applied to agricultural data, and we have provided a code appendix for reference.
#Artificial Intelligence
# Machine learning
# Hospital Medicine
댓글 작성 로그인
팝업 닫기관련 링크
