Li Guo Hung 李國弘

Data Engineer

Data Scientist

AI Engineer

Backend Engineer

  Taiwan

7 years background of Medical Engineering

5 years experience of Machine Learning and Deep Learning (including structure data, image (computer vision) and NLP

Have experience of full stack development

2 years experience as a Data Engineer  AESOP Technology

Experiences of building recommendation systems and data integration platforms


[email protected] 

  https://www.linkedin.com/in/%E5%9C%8B%E5%BC%98-%E6%9D%8E-4792a410b/ 

  https://github.com/MasterLKH180cm

Project Lists:

(Data Science)

https://hackmd.io/@JqKPghXeQSGi1_hMmxtK7g/HJGTei7J9

(Backend)

https://hackmd.io/@JqKPghXeQSGi1_hMmxtK7g/Sk7RgiQ19

Work Experience 工作經歷 

Data Scientist/ AI Engineer  •  AHEAD Intelligence Ltd. 先勁智能有限公司

Mar 2023 - Present

  • Hands on data processing, statistical analysis, algorithm development, and deployment in research and production
  • Data analysis, visualization and generating reports
  • Performing machine learning model training experiments and other exploratory work for research and development
  • Data management – curating & annote data, upload data to database
  • Management of the database itself – efficiently locating and retrieving required data from the storage facility and constantly be on the alert for incursions or attempts to breach cybersecurity
  • Pipeline maintenance 
  • Working with cross-functional medical, business, and technical teams to optimize data inputs according to model needs and outcomes

Data Engineer 資料工程師  •  AESOP Technology 醫守科技股份有限公司

Feb 2021 - Mar 2023

  • Mayo Clinic Project
    • Python, Bigquery, SQL
    • Using American data
    • Build association models for medicines, diagnoses, procedures, surgery, lab test, diagnoses extracted by NLP
  • Prescription Error Detection System
    • SQL 
    • Billions of prescription data
    • Detection of wrong prescribed medicines 
  • Health Insurance Declaration Recommendation System
    • Pytorch
    • GRU-based deep learning model, co-occurrence association statistic modal 
    • Recommend  the proper diagnosis codes for each prescription
  • Missing Diagnosis Detection System
    • Natural Language Processing
    • Using NLP to extract diagnosis terms and translate them to a diagnosis code and combine it with usage frequency
    • Web-based Internal Tool Development
  • Clinical Data Cleansing and Validation 

Research Assistant 研究助理   •  國立臺灣大學 

Jul 2020 - Oct 2020

  • Medical AI Research with Professors and Doctors
    • ETL, Modeling, Analysis
    • Journal publication
  • Personnel Training
    • Lecture
    • Small talk

Part-Time Software Engineer 兼職軟體工程師  •  National Taiwan University Hospital 臺大醫院

Sep 2018 - Dec 2019 

  • Medical Data Delivering using SQL

Education 學歷

2018 - 2020

National Taiwan University(國立台灣大學)

Biomedical Electronics and Bioinformatics

生醫電子與資訊學研究所

2014 - 2018

National Yang Ming University(國立陽明大學)

Biomedical Engineering

生物醫學工程系

2014 - 2018

National Yang Ming University(國立陽明大學)

Biophotonics and Nano Science

生醫光電暨奈米科學學士學位學程

Skills 技能

Machine/Deep Learning


  • Structure data, computer vision, natural language processing 
  • ETL
  • Modeling
  • Visualization
  • Analysis
  • Performance Evaluation
  • Deployment

System Design/Web Develop


  • React.js
  • Node.js
  • Django
  • postgreSQL, MSSQL
  • Data Integration Platform, Dashboard

Language


  • Python
  • SQL
  • Bash
  • Javascript
  • Golang
  • Matlab
  • SAS
  • C/C++

Big Data


  • Bigquery
  • Spark
  • PowerBI

Tools


  • Github
  • Docker

Cloud


  • AWS
  • GCP


Publication 期刊發表

2020

ACUTE EXACERBATION OF A CHRONIC OBSTRUCTIVE PULMONARY DISEASE PREDICTION SYSTEM USING WEARABLE DEVICE DATA, MACHINE LEARNING, AND DEEP LEARNING: DEVELOPMENT AND COHORT STUDY 

Li, Guo-Hung, Wu, Chia-Tung, Huang, Chun-Ta, Lai, Feipei, Kuo, Lu-Ch eng, Chien, Jung-Yien

JMIR mHealth and uHealth (SCI, IF=4.31) / Accept 

2021

DEEPDRG: PERFORMANCE OF ARTIFICIAL INTELLIGENCE MODEL FOR REAL-TIME PREDICTION OF DIAGNOSIS-RELATED GROUPS

Md. Mohaimenul Islam, Guo-Hung Li, Tahmina Nasrin Poly, Yu-Chuan (Jack) Li

Healthcare (SCI, IF=2.65) / Accept 

2022

PREDICTING POSTOPERATIVE MORTALITY WITH DEEP NEURAL NETWORKS AND NATURAL LANGUAGE PROCESSING: MODEL DEVELOPMENT AND VALIDATION 

Chen P, Chen L, Lin Y, Li G, Lai F, Lu C, Yang C, Chen K, Lin T 

JMIR Med Inform (SCI, IF=3.23)/ Accept 

Li Guo Hung has over 7 years background of biomedical engineering, over 5 years experience of machine learning and deep learning, including structure data, image (computer vision), and NLP with lots of projects and 3 journal publications, and 2 years of experience in web-based system development. He is also familiar with using cloud platforms, including GCP and AWS.


Li Guo Hung is currently working for AESOP technology as a data engineer for Medical Big Data Product and he has participated in 3 main products about “Prescription Error” and “Health Insurance Declaration”. He used to process billions of prescription data and generate robust association models of medication, diagnosis, procedure, lab test, and surgery. He is the first person to use the GRU-based deep learning model to suggest the proper diagnosis code in AESOP, which improved over 5% in accuracy compared with the original method. He is also the first person to use natural language processing to deal with free text medical records and find the forgotten diagnosis codes which should be listed in the problem list but are not, and this method served as a new function of the existing system. With his help, AESOP technology has been accepted into Mayo Clinic Platform_Accelerate, a 20-week program that helps early-stage health tech AI startups get market-ready.


Li Guo Hung is a talented and hard-working engineer who devoted himself to data science and computer science. He has lots of practical experience in the medical informatics system using python, JavaScript and Golang. He is not only an experienced data science worker and a full-stack software engineer, but also a good communicator. He is used to being a translator between technical domains and people. He is good at helping people to understand profound knowledge.