Patents


UNet-based Hydrocephalus Diagnosis Support System

Chengzhen Liu, Boxuan Hu, Lifan Gu
Computer Software Copyright, 2024

The diagnosis of pediatric hydrocephalus traditionally relies on radiological examinations, with Evans' Index as a key severity indicator. To address the time-consuming and subjective nature of manual measurements, we propose an automated diagnostic approach using deep learning. Our enhanced U-Net model performs lateral ventricle segmentation and Evans' Index measurement from brain imaging, incorporating dense connectivity and attention mechanisms. Results demonstrate high segmentation accuracy and strong correlation with manual measurements, validated through clinical cross-validation. This automated solution offers efficient and objective diagnostic support, showing promise for broader medical imaging applications.

Open Sources


SkyPilot
MultiCloud Systems and AI Infrastructure
Achievement
[SkyServe Sync Logs]   [SkySys Lint]
IntelliDoc
Cloud Computing and AI-based systems
Achievement
[Code] [Poster]
The Second Persona
2D Game Development with Unity
Achievement
Published on itch.io [Link]
BrainFlow Vision
Focusing on Neuroscience and Deep Learning
Achievement
Obtained Computer Software Copyright Registration Certificate [PDF]

Hobby Projects


OpenSat: Open5gs Simulator for Satellite Networks Enhancement Star

- Satellite Network Simulator for 5G CoreNet Roaming 🛰️
- Selected to Awesome-Telco even without notification :) 🤔
[wiki] [repo]

Hypatia Impl: Simulator for Satellite Networks Enhancement Star

Reproduction and enhancement of IMC 20' Exploring the "Internet from space" with Hypatia 🚀
[paper] [repo]

Awesome Satellite Instances Original Star

Instances of wireless and space networks, and summarizing them categorically 📡
[repo]

Mobility Management Simulator Original Star

Partial implementation of RFC 5944 Mobile IP framework for Linux Ubuntu 22.04LTS, written in Python 3.10 🚑
[repo]

UTS: Unifying the Sky Original Star

One Claude/Codex/Gemini Deployment, Multiple Backends Support 🤖
[repo]