RESEARCH

研究

Research

我的研究圍繞腦波訊號本身:怎麼量得準、怎麼把雜訊和訊號分開,以及量到的東西能不能說明認知歷程的個別差異。

My work centres on the EEG signal itself: how to measure it reliably, how to separate noise from signal, and whether what we measure can speak to individual differences in cognition.

PROJECTS

研究計畫

Projects

腦波雜訊特徵萃取方法之開發EEG Artifact Feature Extraction

進行中Ongoing

與工研院及鯨揚公司的產學合作計畫。蒐集各類腦波雜訊資料,並把它建成可參數化調整的雜訊模型,讓後續的去雜訊演算法有可控的測試素材。An industry project with ITRI and a partner company. We collect EEG artifact data and turn it into parameterised noise models, so that denoising algorithms have controllable material to be tested against.

陽明交大 神經工程實驗室NYCU Neural Engineering Lab指導 柯立偉Advisor Li-Wei Ko2026–2026–

使用編序教學法進行提示工程研究Programmed Instruction as Prompt Engineering

已結案Completed

國科會大專學生研究計畫。把編序教學法拆解步驟、逐步提示的邏輯套用到大型語言模型的提示設計,研究結果投稿至清華大學教育創新研討會。An NSTC Undergraduate Research Grant. I applied the step-wise logic of programmed instruction to prompt design for large language models; the results were submitted to the NTHU Conference on Educational Innovation.

國科會大專生計畫NSTC Grant指導 莊鈞翔Advisor Chun-Hsiang Chuang2024–20252024–2025

EXPERIENCE

研究經歷

Research Experience

  • 2025.06–2026.06

    研究助理Research Assistant

    國立清華大學 許慧玉教授實驗室 · 指導:許慧玉教授、姚在府教授Prof. Hui-Yu Hsu's Lab, NTHU · Advisors: Hui-Yu Hsu, Zai-Fu Yao

    每週固定前往台大醫院新竹分院進行腦波研究收案,對象為阿茲海默症或帕金森氏症之長者。熟悉無線乾式電極腦波帽的操作,並負責對其他收案人員進行教學。Travelled weekly to NTUH Hsinchu Branch to collect EEG data from older adults with Alzheimer's or Parkinson's disease. Experienced with wireless dry-electrode EEG headsets, and responsible for training other data collectors.

TRAINING

訓練與獎項

Training and Awards

  • 2025.08

    MRI 安全講習MRI Safety Training

    國立成功大學National Cheng Kung University

    通過 MRI 安全講習考試並取得證書,具操作 MRI 實驗電腦資格。Passed the MRI safety certification exam; qualified to operate MRI experiment workstations.

  • 2023.12

    教育大數據分析競賽Educational Big Data Analytics Competition

    教育部 · 全國賽入圍Ministry of Education · National finalist

METHODS

方法與工具

Methods and Tools

訊號處理以 MATLAB 的 EEGLAB 與 ERPLAB 為主,分析與視覺化則在 MATLAB 與 Python 之間來回。相較於把個別差異當成需要平均掉的雜訊,我比較傾向把它當成訊號本身。

硬體上熟悉無線乾式電極腦波帽在非實驗室環境的操作——醫院診間、教室這類地方的訊號品質問題,是我研究雜訊的起點。

Signal processing is mostly done in MATLAB with EEGLAB and ERPLAB; analysis and visualisation move between MATLAB and Python. Rather than treating individual differences as noise to be averaged away, I tend to treat them as the signal.

On the hardware side, I am used to running wireless dry-electrode EEG headsets outside the lab. The signal-quality problems of clinics and classrooms are where my interest in artifacts began.