Lei Yin

Dr. Lei Yin


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 August 2024 in Diagnostics
Verified
Journal Article
Open access
PDF file
MYMingjing Yang
Mingjing Yang
ZWZhicheng Wu
Zhicheng Wu
YZYu han Zheng
Yu han Zheng
LHLiqin Huang
Liqin Huang
WDWangbin Ding
Wangbin Ding
LYLei Yin
Lei Yin
Given the diversity of medical images, traditional image segmentation models face the issue of domain shift. Unsupervised domain adaptation (UDA) methods have emerged as a pivotal strategy for cross modality analysis. These methods typically utilize generative adversarial networks (GANs) for both image-level and feature-level domain adaptation through the transformation and reconstruction of images, assuming the features between domains are well-aligned. However, this assumption falters with significant gaps between different medical image modalities, such as MRI and CT. These gaps hinder the effective training of segmentation networks with cross-modality images and can lead to misleading training guidance and instability. To address these challenges, this paper introduces a novel approach comprising a cross-modality feature alignment sub-network and a cross pseudo supervised dual-stream segmentation sub-network. These components work together to bridge domain discrepancies more effectively and ensure a stable training environment. The feature alignment sub-network is designed for the bidirectional alignment of features between the source and target domains, incorporating a self-attention module to aid in learning structurally consistent and relevant information. The segmentation sub-network leverages an enhanced cross-pseudo-supervised loss to harmonize the output of the two segmentation networks, assessing pseudo-distances between domains to improve the pseudo-label quality and thus enhancing the overall learning efficiency of the framework. This method’s success is demonstrated by notable advancements in segmentation precision across target domains for abdomen and brain tasks.

24 February 2025 in Medical & Biological Engineering & Computing
Verified
Journal Article
Closed access
PDF file
MYMingjing Yang
Mingjing Yang
KYKangwen Yang
Kangwen Yang
MWMengjun Wu
Mengjun Wu
LHLiqin Huang
Liqin Huang
WDWangbin Ding
Wangbin Ding
LYLei Yin
Lei Yin

18 January 2026 in IJC Heart & Vasculature
Verified
Journal Article
Open access
YXYilin Xu
Yilin Xu
RCRui Chen
Rui Chen
XLXinglu Li
Xinglu Li
HLHuihua Lin
Huihua Lin
LLLirong Lin
Lirong Lin
MMMingping Ma
Mingping Ma
To evaluate the utility of cardiovascular magnetic resonance imaging (CMR) in distinguishing Anderson-Fabry disease (AFD) harboring the c.640–801G > A mutation from hypertrophic cardiomyopathy (HCM). We enrolled 30 AFD patients, 30 age- and sex-matched HCM patients, and 30 healthy controls (HC). All participants underwent CMR (3.0 T). Left ventricular hypertrophy (LVH) prevalence was high in both AFD and HCM groups (P = 0.12). AFD patients exhibited significantly greater left ventricular lateral wall thickness compared to both HCM and HC (P < 0.001), and a significantly lower interventricular septal to lateral wall thickness ratio (IVS/LW) than HCM (P < 0.001). Late gadolinium enhancement (LGE) was significantly more frequent in the basal inferolateral and apical segments in AFD compared to HCM (P < 0.001 and P = 0.039, respectively). Native T1 values were significantly lower in AFD than HCM in the global LV, septal LV, and within LGE regions (all P < 0.001). Compared to HC, AFD patients had significantly lower septal native T1 (P < 0.001), but comparable global LV native T1 (P = 0.155). Native T1 cut-offs effectively discriminated AFD from HCM: septal native T1 ≤ 1247 ms, global native T1 ≤ 1256 ms, and LGE region native T1 ≤ 1334 ms. Septal native T1 demonstrated the strongest discriminatory capacity. CMR effectively differentiates AFD patients with the c.640–801G > A mutation from HCM. Key discriminators include lower native T1 values (Caution against pseudo-normalization), higher prevalence of basal inferolateral LGE, and more symmetric LVH in AFD.