Publications (6)
SpikeDS: Dual Sparsity Spikformer for Perineural Invasion Prediction in 3D MRI
Induk Um, Youngung Han, Kyeonghun Kim +14
The paper introduces SpikeDS, a spiking neural network with dual sparsity mechanisms to efficiently predict perineural invasion in cholangiocarcinoma from 3D MRI scans, achieving h…
Adaptive Routing for Efficient Diffusion Transformer-Based PNI Prediction
Youngung Han, Dohyun Kweon, Kyeonghun Kim +15
The paper proposes a diffusion‑based transformer model with adaptive routing to predict perineural invasion in cholangiocarcinoma from MRI scans, achieving competitive accuracy whi…
Anatomy-Privileged Distillation with Token Routing for MRI-Based Prediction of Perineural Invasion
Hyunsu Go, Youngung Han, Kyeonghun Kim +15
The paper introduces a teacher‑student deep learning framework that uses anatomy‑aware token routing to predict perineural invasion from T2‑weighted MRI without needing segmentatio…
MMA-Former: Multi-Window Mixture-of-Head Attention Transformer for Adaptive PNI Prediction in 3D MRI
Youngung Han, Induk Um, Kyeonghun Kim +9
The paper introduces MMA-Former, a 3D transformer model with a multi-window mixture-of-head attention mechanism, to predict perineural invasion from T1-weighted MRI scans.
NeoNet: An End-to-End 3D MRI-Based Deep Learning Framework for Non-Invasive Prediction of Perineural Invasion via Generation-Driven Classification
Youngung Han, Minkyung Cha, Kyeonghun Kim +12
Minimizing invasive diagnostic procedures to reduce the risk of patient injury and infection is a central goal in medical imaging. And yet, noninvasive diagnosis of perineural inva…
LoSA-Net: A Localized and Scale-Adaptive Network for Boundary-Sensitive Prediction of Perineural Invasion in 3D MRI
Youngung Han, Hyunsu Go, Kyeonghun Kim +9
The paper introduces LoSA-Net, a neural network that uses localized self‑attention and scale‑adaptive processing to improve detection of perineural invasion boundaries in 3D contra…