3d-imaging

🔬🖥 Automated Blood Vasculature Analysis of 3D Light-Sheet Image Volumes

Official code repository for the paper: "Mask-ToF: Learning Microlens Masks for Flying Pixel Correction in Time-of-Flight Imaging"

Proposed a monocular 3D polarization imaging approach for non-Lambertian surfaces by investigating the subtle difference in specular and diffuse reflection and then separating them using independent component analysis (ICA).

Application of AI in 3D Medical Imaging in quantifying Alzheimer's disease progression through automated measurement of hippocampal volume

C++, opencv algorithm for generating 3d images of 2d images, generate 3d image from 2d image

Python-based utility for visualizing 3D data (with an additional color channel) using matplotlib. The voxel plotter allows for a customizable display of 3D data with controls for transparency threshold, skew factor, and sampling ratio.

A series of networks to implement photorealistic renderings through the MeVis Path Tracer from: "The SmARTR Pipeline: a modular workflow for the cinematic rendering of 3D scientific imaging data"

A python script for creating directory structure for 3D projects

The final project of "Applying AI to 3D Medical Imaging Data" from "AI for Healthcare" nanodegree - Udacity.

Using deep learning and computer vision techniques to segment tumors in brain MRI images using UNet.

mask2graph converts 2D/3D binary masks into deterministic geometric graphs (nodes, edges, radii, and metadata) for downstream analysis, simulation, and learning systems.

Machine learning pipeline for the Vesuvius Challenge an open research effort to read carbonised Herculaneum scrolls without unrolling them. Uses PyTorch convolutional models on 3D X-ray micro-CT data to segment ink signatures from charred papyrus, contributing to the recovery of ancient texts

High-throughput detection and enumeration of tumor cells in blood using Digital Holographic Microscopy (DHM) and Deep Learning.

PyTorch Lightning implementation of SRGAN for 3D X-ray tomography. Features stable training for Lab-based and Synchrotron datasets. Lead developer: @calum-green.

The project focuses on four heart conditions that may go initially unremarkable, but can eventually be life-threatening. It a 5-class classification problem ‘0’ Healthy controls • ‘1’ Myocardial infarction • ‘2’ Dilated cardiomyopathy • ‘3’ Hypertrophic cardiomyopathy • ‘4’ Abnormal right ventricle

🫀 Explainable-AI harness for an echocardiographic cine loop: PIL frame decoding, an ImageNet-pretrained VGG16/ResNet50 backbone, and Grad-CAM, LIME, SHAP and occlusion-sensitivity maps. No cardiac fine-tuning yet, so predictions remain ImageNet categories.

Neural network tools for automated cell contouring, fluorescent signal detection, and 3D distance measurements in Imaris microscopy scans, with human review and editing

3D Slicer MCP adapter in initial development for the DCC-MCP ecosystem. Planned AI agent integration for 3D imaging workflows; implementation is not yet available.

3D Implementation of the detectron variants.