Hi, I'm Qian NIU
Researcher
Medical AI, Infodemiology
PhD research focuses on the intersection of Natural Language Processing (NLP) and Health. Using social media and searching data, my works provide real-world evidence to understand public opinions and sentiments toward COVID-19 and HPV vaccination, and online information-seeking behaviors for infectious diseases. Currently exploring the impact of generative AI on the research pipeline and the potential health applications.
PROFILE SUMMARY
SKILLS
Python, R, SQL, AWS, GCP, Machine Learning, Deep Learning, Statistics, Tableau, SPSS, PyTorch, TensorFlow, Fine-tuning LLM, Langchain
Languages: English (Fluent, TOEIC 860), Japanese (JLPT N1), Chinese (Native)
Achievements and Qualifications
Kyoto University President's Award 2023, Academic Division
Read more
Google PhD Fellowship
2022-2024
61 global recipients, 2 in Japan
Read more
JST SPRING Fellowship
2021-2023
Read more
Awards and Fellowships
Kaggle Competition Achievements
HMS - Harmful Brain Activity Classification
Top 3%, Silver Medal
Read more
Kaggle - LLM Science Exam
Top 8%, Bronze Medal
Read more
Professional Certifications
Education
and Experience
Education
Ph.D. in Health Informatics
Kyoto University, Japan
2020-2024
MS in Health Sciences
International University of Health and Welfare, Japan
2018-2020
BS in Rehabilitation Medicine
Capital Medical University, China
2013-2017
Experience
Matsuo-Iwasawa Lab U-Tokyo
Research Intern
Nov 2024 - present
Google PhD Fellowship
Fellow
Sep 2022 - Aug 2024
JST SPRING Fellowship
Fellow
Oct 2021 - Mar 2023
Teaching Assistant Kyoto University
Part-time
Oct 2021 - Mar 2022
Clinical Trainee
Beijing Tsinghua Changgung Hospital
Permanent
Jul 2017 - Mar 2018
Clinical Trainee
China Rehabilitation Research Center
Internship
Mar 2015 - Jul 2017
Volunteer
Capital Medical University Red Cross Society
Sep 2013 - Jul 2015
Main Research Projects
This paper explores the dynamics influencing rapid vaccination rates in Japan, a country with historically low vaccine confidence. The study employs a retrospective analysis of tweets related to COVID-19 in Japan from February 1 to September 30, 2021. It examines the correlation between public sentiment on Twitter, particularly fear of infection, and the vaccination rates using methods such as unigram and bigram token analysis, sentiment analysis, and topic modeling.
Key findings include:
Despite low vaccine confidence, a significant correlation was observed between tweets containing keywords related to vaccine reservations and venues, and the actual vaccination rates.
Negative sentiments predominated, with fear being the most common emotion throughout the period studied.
The study suggests that high awareness of COVID-19 risks and efficient dissemination of vaccine reservation information were critical in prompting public willingness to get vaccinated.
This paper explores the public response to COVID-19 vaccinations in Japan through a Twitter analysis, focusing on sentiment and opinion from August 1, 2020, to June 30, 2021. It investigates the relationship between public sentiment on Twitter and the slow vaccination rollout in a setting of low vaccine confidence. The study uses techniques including sentiment analysis, topic modeling, and correlation analysis of tweets containing COVID-19 vaccine-related keywords.
Key findings include:
Tweet Growth: Vaccine-related tweet volume increased after large-scale vaccinations began.
Sentiment: 85% of tweets were neutral; negative sentiment dominated the rest.
Negative Sentiment Sources:
Vaccine safety concerns.
Slow vaccination progress.
Issues with the reservation system and scams.
Vaccine Brand Sentiment: most positive toward Moderna, most negative toward AstraZeneca due to blood clot concerns, and focused on the effectiveness of Pfizer.
"The Effect of Fear of Infection and Sufficient Vaccine Reservation Information on Rapid COVID-19 Vaccination in Japan: Evidence from a Retrospective Twitter Analysis."
Journal of Medical Internet Research, 24(6), e37466, 2022.
Full Text
"Public opinion and sentiment before and at the beginning of COVID-19 vaccinations in Japan: Twitter analysis.''
JMIR infodemiology, 2(1), e32335, 2022.
Full Text
Kyoto University x G4A Tokyo Digital Health Symposium, 2021.
“Public Opinion and Sentiment Before and at the Beginning of COVID-19 Vaccinations in Japan: Twitter Analysis.”
Full Poster
Informatics in Biology, Medicine and Pharmacology 2022.
“Fear of Infection and Sufficient Vaccine Reservation Information Might Drive Rapid COVID-19 Vaccination in Japan.”
Full Poster
The paper focused on the changes in public awareness of hand, foot, and mouth disease (HFMD) in Japan by analyzing Google Trends data from 2009 to 2021, especially before and during the COVID-19 pandemic. We employed correlation and regression analyses to identify significant terms and gauge the timing and volume of searches in relation to HFMD cases reported by the National Institute of Infectious Diseases.
Key findings include:
HFMD cases and search volume typically peaked in July, except in 2020 and 2021 when the patterns were disrupted by the COVID-19 pandemic.
Search volumes in 2020 showed weak correlations with HFMD cases, suggesting changes in public interest or behavior due to the pandemic.
The regression analysis found that different search terms became significant in explaining HFMD cases during the pandemic, implying a shift in public focus and possibly improved awareness due to overlapping symptoms and prevention measures between HFMD and COVID-19.
"Explanation of Hand, Foot, and Mouth Disease in Japan Using Google Trends: Infodemiology Study.''
BMC infectious diseases, 22(1), 1-12, 2022.
Full Text
Community Services
Peer-reviewer
Journal of Medical Internet Research,
Frontiers in Public Health, PloS One, JMIR Cancer
Youtuber
明日のテクちゃん
@ashitanoteku
This channel offers clear and concise explanations of cutting-edge tech developments, covering AI, robotics, mobile technology, cloud computing, and more.