Dr. WAN Tsz Kin (温梓健)

As a researcher specializing in applying machine learning techniques in Biomedical Sciences, my focus lies in disease prediction, prevention, and treatment strategies. Throughout my academic journey, extensive research and implementation in machine learning have been my primary areas of expertise. I hold a Doctor of Philosophy in Electrical Engineering from City University of Hong Kong, which was pursued from July 2019 to November 2023. During this time, I actively engaged as an AI Image Research Officer at the Hospital Authority of Hong Kong from May 2022 to January 2024.

I developed a stroke ICD9 code classification model, which effectively categorizes and classifies stroke cases based on the International Classification of Diseases, 9th Edition (ICD9) codes. This model aids in streamlining the diagnosis and treatment process for stroke patients, facilitating more efficient and accurate healthcare interventions.

Furthermore, I have worked on a COVID-19 mortality prediction model. Leveraging machine learning techniques and relevant datasets, this model predicts the likelihood of mortality in individuals diagnosed with COVID-19. By identifying high-risk individuals, healthcare providers can prioritize resources and interventions, potentially improving patient outcomes and allocating healthcare resources more effectively during the ongoing pandemic.

Significantly, my contributions include the development of an AI application that utilizes cutting-edge models such as Yolov5, EfficientDet, and EfficientNet. This application demonstrates remarkable accuracy, achieving 0.98 accuracy and 0.96 F1 score in the detection, localization, and classification of 14 classes of parasites within the biomedical domain. 

In addition, my research encompasses the collection and preprocessing of extensive datasets, ensuring their quality and suitability for thorough analysis. By enabling comprehensive insights and analysis in the biomedical field, my work contributes to the advancement of disease prediction, prevention, and treatment strategies.

1) Machine Learning: Applying machine learning techniques, such as deep learning algorithms or data mining methods, to analyze and extract insights from biomedical data for disease prediction, treatment optimization, or personalized medicine.

2) Disease Prediction and Prevention: Investigating the use of data-driven approaches, including machine learning models, to predict the occurrence and progression of diseases, identify risk factors, and develop preventive strategies.

3) Medical Image Analysis: Developing and applying machine learning algorithms to analyze and interpret medical images, such as X-rays, MRIs, or CT scans, for automated diagnosis, tumor detection, or image segmentation tasks.

4) Public Health and Epidemiology: Exploring the application of machine learning and data analysis techniques to understand patterns of disease spread, assess public health interventions, and develop predictive models for disease outbreaks.