Seungjae Han

I am a PhD student at Neuro-Instrumentation & Computational Analysis Lab (NICA Lab) at KAIST, advised by Professor Young-Gyu Yoon. I am developing novel methods to acquire and analyze calcium imaging data. Before KAIST, I graduated Yonsei University, advised by Jiwon Seo. I studied positioning during undergraduate. I am a recipient of SfN 2023 Trainee Professional Development Award (TPDA).

How can we efficiently and accurately acquire and analyze neuronal signal?

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn

I'm intereted in optimization, computer vision, and microscopy. Representative papers are highlighted.

• [Denoising] Statistically unbiased prediction enables accurate denoising of voltage imaging data, Nature Methods, 2023
• [Robust PCA] Efficient Neural Network Approximation of Robust PCA for Automated Analysis of Calcium Imaging Data, MICCAI, 2021

Review article providing an overview of how artificial intellignece is transforming fluorescence microscopy.

Self-supervised learning method for removing Poisson-Gaussian noise in voltage imaging data as well as a wide range of microscopy images.

Effective and reproducible protocol for whole-brain imaging of larval zebrafish using three-dimensional fluorescence microscopy.

REALS can align multiple images very fast and accurately, even if there are lots of comtaminations.

Erasureless cyclic imaging method with self-supervised spectral unmixing.

Neural network that models the light propagation law can reconstruct a volume using two adjacent images.

Expansion microscopy (ExM) for whole vertebrates. Demonstrated nanoscale resolution imaging of whole zebrafish larvae and mouse embryos by expanding them fourfold.

Fast computational microscopy for the volumetric imaging of neural activity, using two neural networks which perform sparse decomposition and deconvolution.

Fast and scalable gradient descent based Robust PCA algorithm. Achieves an order of magnitude speed improvement, demonstrated with calcium imaging dataset as large as tens of gigabytes.

Smartphone application that calculates predicted distances to near base stations based on received signal strength. Research for LTE-based positioning where GPS signal is unavailable.

Analyze the walking trajectory of a blindfolded participant receiving hapic feedback from a mobile robot. Developed a mobile robot system with a haptic device and algorithm for appropriate haptic feedback.

Localization of sound source using two mics in the indoor multipath environment.
Wrote two domestic conference papers and applied one patent.

Visited COLORS Lab, advised by Emre Ugur.