Kai-Cheng HU

Run Don’t Walk

Ithaca, New York, United States

About

I was fascinated by the idea that signals, essentially just sequences of numbers evolving over time, could reveal something profound about the world. This curiosity drove me to explore their underlying principles and mathematical foundations.

Experience

  • 個人職涯訓練 at Career Break
    Apr 2026 - Present · 4 mos

    Business Operations Trainee at APEX Polytechnic Engineering Co., LTD. Gained hands-on exposure to company operations, including pricing, procurement, supplier coordination, project execution, expense tracking, and cost control.

  • ROC Army (Taiwan) ()
    • Combat Medic
      Feb 2026 - Apr 2026 · 3 mos

      Taiwan's Compulsory Military Service. Direct Supervisor: Captain Yu-Ying Wu Huadong Defense Command, Infantryman -> Hualien Combined Maintenance Depot, Army The 2nd (Eastern) Regional Support Command, Medic Steadfast commitment. Selfless devotion. A mission to safeguard our homeland. Honored to have served.

    • Infantryman
      Jan 2026 - Feb 2026 · 2 mos

  • 研究助理 at 國立臺灣大學電機資訊學院
    Sep 2025 - Jan 2026 · 5 mos

    RA at Biosystem and Medical device Design Lab., supervised by Prof. Hao-Li Liu In our project, which utilizes focused ultrasound to suppress epileptic activity, I was responsible for developing an fMRI-based framework to interpret the resulting neural modulation. I designed a complete preprocessing and analysis pipeline that links raw EPI signals to quantitative measures of activity changes. To improve data reliability, I refined motion correction, segmentation, and filtering steps, and replaced conventional global regressors with component-based nuisance removal derived from non-neural regions. This improved signal specificity and reproducibility across sessions. Building on the cleaned data, I applied a Hilbert-transform analytic signal approach to quantify the amplitude envelope of BOLD fluctuations, capturing how global and regional activity shifted during ultrasound stimulation. This work provided a physically grounded way to describe brain-wide modulation and offered insight into how targeted acoustic interventions reshape large-scale neural dynamics.

  • Research Assistant at Stanford University School of Medicine
    Jun 2025 - Sep 2025 · 4 mos

    Undergraduate Visiting Research Intern at Stanford University, Molecular Imaging Instrumentation Lab, Prof. Craig Levin, under the mentorship of Dr. West Foster. Mail: [email protected] Contributed to the development of a small-animal multi-isotope PET system designed to capture complementary physiological signals from multiple isotopes in a single scan, reducing radiation dose and scan time. Focused on signal processing and quantitative imaging pipeline design, including detector-level energy and depth-of-interaction (DOI) calibration, LUT generation, and image reconstruction. Enhanced triple-coincidence sensitivity through ASIC-based hardware triggering and optimized electronic collimation. Developed an exponential-decay energy fitting model to correct SiPM saturation and nonlinearity, improving energy resolution from 23 % to 13 %. Established a DOI estimation model using the Max/Sum ratio and isotonic regression, achieving ~5 mm DOI resolution, 1.4 ns coincidence time resolution, and refining spatial accuracy. The resulting multi-isotope quantitative imaging pipeline is under review at Physics in Medicine & Biology (PMB), where I am listed as second author.

  • Research Assistant at Stanford Deliberative Democracy Lab
    Jul 2025 - Aug 2025 · 2 mos

    Agentic AI Research and Application Development for Deliberative Democracy. Research Adviser: Prof. Alice Siu