NeuroAI · Computational neuroscience · Connectomics · Biomedical imaging · Deep learning

Joshua Yedam You

Ph.D. Candidate

KAIST School of Electrical Engineering, Daejeon, South Korea

KAIST M.S. in Electrical Engineering at KAIST University of Virginia B.S. in Computer Engineering at the University of Virginia

I am currently a Ph.D. candidate in the KAIST School of Electrical Engineering at the Neuro-Instrumentation & Computational Analysis Lab (NICA), advised by Prof. Young-Gyu Yoon. My research explores how data-driven and interpretable deep learning can transform the analysis of large-scale biomedical imaging and neuroscience data. Building on my background in electrical and computer engineering, I am especially interested in computational approaches that help reveal how the brain's structural and functional organization gives rise to cognition, behavior, and neurological disorders.

Before KAIST, I completed my B.S. in Computer Engineering at the University of Virginia, VA, USA and conducted research at the National Institute of Standards and Technology (NIST), MD, USA through the PREP fellowship.

Google Scholar LinkedIn NICA Lab

News

Research

Current interests
Deep learning for revealing neuron-level functional and structural connectomics

I study computational methods for reconstructing and interpreting neuron-level connectivity from large-scale neural imaging data, spanning structural circuit reconstruction, synapse-level analysis, and functional connectivity inference. A current focus is using explainable deep learning to identify functionally connected neurons and reveal the circuit features that drive model predictions.

Slide: Neural functional connectivity

Related work

  • Manuscript in revision at Nature Communications, 2026 In revision
Deep learning for mapping functional brain organization across subjects and species

I study data-driven approaches for learning functionally meaningful brain parcellations and mapping relationships between brain regions across individuals and species. This work aims to build shared representations of brain organization that enable cross-subject and cross-species comparison without relying only on predefined anatomical labels.

Related work

  • Submitted to NeurIPS 2026 Submitted
Deep learning for modeling latent neural manifolds and decoding population dynamics

I study models that extract low-dimensional latent structure from neural population activity and use these representations to decode behavior, stimuli, or circuit states. My broader goal is to understand how population-level dynamics organize neural computation across recording sessions, animals, and modalities.

Slide: Neural population dynamics modeling

Related work

  • Submitted to NeurIPS 2026 Submitted
  • Manuscript under review at Science Advances, 2026 Under review
Deep learning for modeling disease and abnormalities from functional biomedical imaging data

I study data-driven methods for modeling disease and abnormalities from functional biomedical imaging data.

Slide: Functional disease modeling

Related work

  • Manuscript under review at Nature Communications, 2026 Under review
Deep learning for biomedical image processing & analysis

I study deep learning frameworks for biomedical microscopy and imaging that improve image quality, translate between modalities, automate segmentation, and extract biologically interpretable measurements. These methods aim to turn complex imaging data into scalable, quantitative, and experimentally useful biological insight.

Slide: Biomedical image translation

Related work

Publications

* co-first · ** co-corresponding
Scholar profile

Education

Research Experience

Honors & Awards

Teaching & Mentoring

Invited Talks

Technical Skills

Contact

Open to research conversations and collaboration. For the fastest context, please include the relevant paper, dataset, or research question in your message.

Email
Lab
NICA Lab, KAIST EE
Advisor
Prof. Young-Gyu Yoon
Citizenship
United States
Profiles
Scholar LinkedIn