Micro-lesion segmentation
Designing centroid-guided learning systems for sub-3 mm brain tumor segmentation and counting in multimodal MRI.
Medical image analysis · Segmentation · Multimodal AI
Zhenye Lou
Computer science undergraduate and medical AI researcher building efficient, clinically motivated systems for micro-lesion segmentation, pathology foundation models, and adaptive medical image understanding.
- Education
- Sichuan University, B.S. Computer Science and Technology
- Current Work
- Research internships and collaborations across medical AI labs
- Base
- Chengdu, Sichuan · Hangzhou, Zhejiang
Research moments
Robust, scalable medical AI
Research
Building medical AI methods that connect precision, scale, and clinical use.
Pathology foundation models
Contributing to whole-slide image representation learning pipelines that move beyond patching, segmentation, and handcrafted feature extraction.
Adaptive segmentation
Developing parameter-efficient adaptation and automatic prompting methods for nuclei segmentation across heterogeneous medical domains.
Multimodal clinical AI
Exploring image-text learning, retrieval workflows, and tool-augmented reasoning for medical decision support.
Publications and Manuscripts
Selected work across segmentation, foundation models, and biomedical analysis.
Expand entries for status, venue, and contribution context.
J1 · Journal NuSegDG: Integration of Heterogeneous Space and Gaussian Kernel for Domain-Generalized Nuclei Segmentation
Co-first author. CAS Q1 TOP journal work on domain-generalized nuclei segmentation using medical-domain adaptation and automatic prompting.
DOI: 10.1016/j.knosys.2025.113641
J2 · Journal HRMedSeg: Unlocking High-resolution Medical Image Segmentation via Memory-efficient Attention Modeling
Co-first author. High-resolution medical segmentation study focused on efficient attention modeling for memory-constrained clinical imaging workloads.
DOI: 10.48550/arXiv.2504.06205
C1 · Conference MicroBT: Centroid-Guided Synergistic Learning for Multi-Modal Micro Brain Tumor Segmentation and Counting
First author. Proposes centroid-guided multi-task learning, volume-calibrated loss, and size-aware curriculum optimization for micro brain tumor analysis.
J3 · Journal Topological Representation Based on Wavelet Transform as a Novel Imaging Biomarker for Tumor Diagnosis in Ultrasound Images
Co-author. Biomedical analysis work studying topological and wavelet-based representation for ultrasound tumor diagnosis.
DOI: 10.1016/j.cmpb.2025.108859
J4 · Journal MambaVesselNet++: A Hybrid CNN-Mamba Architecture for Medical Image Segmentation
Co-author. Hybrid CNN-Mamba segmentation architecture for medical image analysis.
DOI: 10.1145/3757324
C2 · Conference GALOS: Generalizable Nuclei Segmentation via Lightweight Adaptation and Automatic Prompting
First author. Lightweight adaptation and prompting framework for generalizable nuclei segmentation.
C3 · Conference AA-NuSeg: A Fully Automated and Adaptive Framework for Accurate Nuclei Segmentation
First author. Automated adaptive framework targeting accurate nuclei segmentation across biomedical image settings.
Research Experience
Research roles spanning large-scale pathology, brain tumor analysis, and adaptive segmentation.
2025 - 2026
Tsinghua University
Research Intern · Prof. Gao Huang, Dr. Yulin Wang · Beijing
Contributed to pathology foundation model infrastructure, including data auditing, preprocessing standardization, large-scale training support, downstream benchmark design, and research synthesis for whole-slide image learning.
2025 - Present
The Hong Kong Polytechnic University
Research Assistant · Prof. Zhen Chen · Hong Kong SAR
Spearheaded MicroBT, developing centroid-guided segmentation and instance-level localization for multi-modal micro brain tumor analysis on BraTS-Lighthouse 2025.
2024
University of Nottingham
Research Assistant · Prof. Xiangjian He · Nottingham
Led parameter-efficient adaptation for medical foundation models, including HS-Adapter, Gaussian-kernel prompting, and a two-stage mask decoder for nuclei segmentation across public cross-domain datasets.
2025
Zhejiang University RealDoctor Lab
Research Intern · Prof. Jian Wu · Hangzhou
Studied medical AI agents, ophthalmology foundation models, multimodal learning, image-text integration, retrieval, and missing-modality pretraining.
2023 - 2024
West China Biomedical Big Data Center
Research Assistant · Prof. Kang Li · Chengdu
Led biomedical deep learning and fusion strategy projects, contributing to manuscripts on adaptive nuclei segmentation and medical image analysis.
Jul - Sep 2023
West China Hospital SenseTime Joint Laboratory
Research Intern · Chengdu
Worked on biomedical data analysis with machine learning and explored topology-driven feature extraction for ultrasound tumor diagnosis.
Education
Sichuan University
B.S. in Computer Science and Technology · 2022 - 2026
GPA: 3.87 / 4.00. Academic training in computer science with research centered on medical image computing and clinically motivated AI systems.