LoSA-Net: Localized and Scale-Adaptive Network for Boundary-Sensitive Prediction of Perineural Invasion in 3D MRI

Jul 16, 2026

Upload by Eden

Perineural invasion (PNI) signals tumor aggressiveness and can change surgical planning, yet MRI shows faint, elongated or subtle nerve-sheath cues that closely resemble nearby anatomy and are often weakened by routine downsampling or overly global feature aggregation. LoSA-Net introduces a localized, scale-adaptive 3D architecture for boundary-sensitive PNI prediction. Talking Neighborhood Attention preserves nerve-aligned detail, Scale-Adaptive Feature Mixing adjusts receptive fields via multiscale depthwise processing, and Cross-Scale Refinement and Alignment maintains stage-wise consistency between semantic context and fine boundaries. On contrast-enhanced MRI from 168 cholangiocarcinoma patients, LoSA-Net reaches an AUC of 0.7567 and surpasses matched convolutional and transformer baselines.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages5

Reading Time13 min

Abstract

Introduction

Methodology

LOSA-NET: A LOCALIZED AND SCALE-ADAPTIVE NETWORK FOR BOUNDARY-SENSITIVE PREDICTION OF PERINEURAL INVASION IN 3D MRI Youngung Han 1 ,2, Hyunsu Go 1, Kyeonghun Kim2, Induk Um3, Junga Kim 1, Jaewon Jung 1, Woo Kyoung Jeong4, Won Jae Lee5, Pa Hong5, Ken Ying-Kai Liao6, Hyuk-Jae Lee 1, Nam-Joon Kim 1 ,† 1 Seoul National University, Seoul, Republic of Korea 2 OUTTA, Seoul, Republic of Korea 3 Chung-Ang University, Seoul, Republic of Korea 4 Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 5 Samsung Changwon Hospital, Changwon, Republic of Korea 6NVIDIA AI Technology Center, Taipei, Taiwan †Corresponding author: [knj01@snu.ac.kr](knj01@snu.ac.kr) arXiv :2607 . 10992v1 [ cs .CV] 13 Jul 2026 ABSTRACT Perineural invasion (PNI) is a clinically relevant indicator of tumor aggressiveness and can influence surgical decision-making, motivating interest in reliable preoperative assessment. The subtle MRI features of PNI, however, often resemble nearby anatomy, complicating noninvasive prediction. These fine perineural cues are easily attenuated by routine downsampling or overly global feature aggregation, reducing the effectiveness of conventional volumetric models. We present LoSA-Net, a localized and scale-adaptive architecture for boundary-sensitive PNI prediction in 3D MRI. Talking Neighborhood Attention (TNA) preserves nerve-aligned detail through localized self-attention with head-wise mixing, and Scale-Adaptive Feature Mixing (SAFM) modulates the receptive field using multiscale depthwise processing. Cross-Scale Refinement and Alignment (CSRA) maintains consistency between semantic context and highresolution boundaries across stages. In contrast-enhanced MRI scans from 168 patients with cholangiocarcinoma, LoSA-Net achieves an AUC of 0.7567 and outperforms representative convolutional and transformer baselines under matched preprocessing and optimization settings. Index Terms— PNI, Boundary-aware learning, Multi-scale feature fusion, Medical image analysis 1. INTRODUCTION Perineural invasion (PNI) is the spread of tumor cells along or within the nerve sheath and is associated with pain, neurologic deficits, and poor survival across multiple malignancies [1, 2] . PNI may warrant more aggressive margin considerations during surgery [3, 4], suggesting that reliable noninvasive preoperative identification could help avoid inappropriate treatment and support oncologically safer, tailored care. Despite its clinical relevance, PNI typically appears on MRI as faint, elongated signal changes or subtle perineural thickening near nerve-caliber structures, often resembling adjacent vessels or ducts and sometimes falling below the effective resolution of routine protocols [5, 6] . Consequently, preoperative detection remains challenging even for expert readers, and confirmation frequently relies on postoperative pathology. Radiomics-based approaches have demonstrated that MRI carries useful information for noninvasive PNI prediction [3, 7, 8], but their hand-crafted features are sensitive to acquisition and scanner variability, which hampers reproducibility and external generalization [9, 10, 11] . These challenges motivate end-to-end volumetric models that learn features directly from the data. Generic 3D convolutional and transformer backbones, however, have inductive biases that are not particularly aligned with the characteristics of PNI. Convolutional encoders [12, 13, 14] achieve invariance by striding and pooling, which can diminish the thin perineural edges that differentiate PNI from look-alike anatomy. Studies of U-Net variants and recent segmentation models highlight consistent difficulties at low-contrast, fine boundaries, reinforcing the need for boundary-aware, multi-scale designs [15, 16] . Vision transformers provide long-range context but lack built-in locality [17, 18, 19] . Hierarchical or windowed designs [20, 21] recover local structure, although cros

FAQs

What makes preoperative perineural invasion (PNI) prediction difficult on 3D MRI?

PNI presents as faint, elongated signals or subtle perineural thickening that can resemble adjacent vessels or ducts, and routine acquisition downsampling can reduce these boundary cues below effective resolution.

How does LoSA-Net address the loss of fine boundary information in conventional models?

LoSA-Net uses Talking Neighborhood Attention to preserve localized nerve-aligned detail, Scale-Adaptive Feature Mixing to modulate receptive fields with multiscale depthwise processing, and Cross-Scale Refinement and Alignment to keep semantic context consistent with high-resolution boundaries across stages.

How well does LoSA-Net perform compared with baseline models?

On contrast-enhanced MRI from 168 cholangiocarcinoma patients, LoSA-Net achieves an AUC of 0.7567 and outperforms representative convolutional and transformer baselines under matched preprocessing and optimization settings.

Jul 16, 2026

Upload by Eden