Name: Xin Tie
Type: User
Company: University of Wisconsin-Madison
Bio: Graduate research assistant in Department of Radiology, University of Wisconsin Madison. Fascinated by CV in Medical Imaging, NLP in Radiology and Multi-Modal
Location: Madison
Blog: https://www.linkedin.com/in/xintie8254b6157/
Xin Tie's Projects
CT Recon Toolbox
A collection of resources on applications of multi-modal learning in medical imaging.
Slides for Conference Presentation
Code for CS771 Learning-based Computer Vision Course Project: An Integrated AI Framework for COVID-19 Classification and Detection Using Chest Radiographs
Code for Paper: A Quality‐checked and Physics‐constrained DL Method to Estimate Material Basis Images from Single‐kV Contrast‐enhanced Chest CT scans; TensorFlow Op for CT Reconstruction
🦜🔗 Build context-aware reasoning applications
Longitudinally-aware Segmentation Network for interim-therapy PET segmentation in pediatric Hodgkin lymphoma
Homework for CS 760 Machine Learning
GPU-accelerated Monte Carlo x-ray transport code to simulate medical x-ray imaging devices.
AI Toolkit for Healthcare Imaging
PET report summarization using Pointer Generator Network
This is the webpage we designed for physicians to review the test cases in our study
Code for Paper: Automatic Personalized Impression Generation for PET Reports Using Large Language Models
Code for CS769 Advanced NLP Course Project: Automatic Impression Generation for PET Reports using Lightweight Adaptation of Pretrained LLMs
Seq2seq model training without built-in trainer function in the transformer library
Code for CS766 Computer Vision Course Project: Sparse View CT Reconstruction Based on a Dual-domain Convolutional Neural Network
TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients.
Datasets, Transforms and Models specific to Computer Vision
GitHub Pages template for academic personal websites