Weakly supervised 3D classification of multi-disease chest CT scans using multi-resolution deep segmentation features via dual-stage CNN architecture (DenseVNet, 3D Residual U-Net).

This Repo contains the updated implementation of our paper "Weakly supervised 3D classification of chest CT using aggregated multi-resolution deep segmentation features", Proc. SPIE 11314, Medical Imaging 2020: Computer-Aided Diagnosis, 1131408 (16 March 2020)

Computes CT contrast phase and GI tract contrast using TotalSegmentator and ML

Single image super resolution algorithm RED+ADMM+De-QuIP

A Novel Multiresolution-Statistical Texture Analysis Architecture: Radiomics-Aided Diagnosis of PDAC Based on Plain CT Images

Képalkotás szenzor adatokból

Implementation of IEC 60601-2-44 Ed.4 Annex GG for objective evaluation of CT Metal Artifact Reduction (MAR) using DDOG CHO and AUC analysis.

Dirichlet VAE trained unsupervised on NIH DeepLesion (32k multi-organ CT patches) to test spontaneous dual-axis disentanglement of organ identity and malignancy. Probe-based evaluation with UMAP, MIG score, and latent traversals. Extension of Keel et al. BMVC 2023.

A Simple Tutorial for 2D and 3D U-Net Training and Test Pipeline in Pytorch

MSc záróvizsga kidolgozás orvosi mérnökinformatika szakirányra

A comprehensive benchmark study evaluating the robustness of Segment Anything Model (SAM) and its medical domain adaptation (MedSAM) under realistic noisy medical imaging conditions.

Command line tool to extract Siemens' AI-Rad Companion data from a DicomConquest Server into a relational SQLite database.

The 3D footplate is a non-commercially viable tool to assist medical imaging of complex hindfoot pathologies. This 'Class I Medical Device' helps to position and fixate the foot to the lower leg, allowing clinical stress tests to be simulated during CT scans. The medical images acquired can provide quantitative displacements of bones.

Adaptive weighted sparse coding with edge-aware transform filtering for clinically reliable low-dose CT denoising

Bias-aware multimodal pancreatic cancer detection using CT imaging and urinary biomarkers, with ResNet50 CT modelling, biomarker MLP classification, and exploratory fusion analysis.

Preprocessing images with focus on segmentation of CT slices