pet-imaging

FALCON is a Python-based software application designed to facilitate PET motion correction, both for head and total-body scans. Our program is built around the fast 'greedy' registration toolkit, which serves as the registration engine. With FALCON, users can enjoy a streamlined experience for implementing motion correction.

Quality enhancement of ultra-low-dose PET images using 3D-UNet on wavelet domain.

CPU-first framework for protocol-robust dynamic PET quantification, temporal recovery, kinetic connectivity, and radiomics

How To Segment in 3D Using 2D Models: Automated 3D Segmentation of Prostate Cancer Metastatic Lesions on PET Volumes Using Multi-Angle Maximum Intensity Projections and Diffusion Models

Képalkotás szenzor adatokból

🏥 Wavelet-domain deep learning framework (U-Net based) for ultra-low-dose PET image denoising. Optimizes anatomical structure preservation and inference latency using 2D/3D Discrete Wavelet Transforms.

GATE PET model for voxel-wise Dosimetry in Radiouclide Therapy namely in Radioembolization

R-based neuroimaging analytics pipeline built at the Northwestern University Mesulam Center. Includes automated multi-modal PET/MRI reporting (R Markdown + cron), amyloid SUVr/visual read QC, and Tau PET discrepancy analysis via REDCap API.

🏆 MedAI Hackathon 2026 — 3rd Place · 3D CNN (MedicalNet ResNet-34 + FiLM) for amyloid PET Centiloid prediction · Boston University · PyTorch

Lightweight box-prompted OCT, ultrasound and PET segmentation with reproducible 2D/3D evaluation

Benchmark of 14 deep learning models for temporal frame prediction in simulated dynamic PET. Leave-one-tumour-out cross-validation.

Laptop-sized teaching re-implementation of SiPM signal multiplexing in a monolithic PET detector (Subedi et al. 2025, Radiation Measurements 182:107399). CPU and Apple-GPU versions.

Threshold-specific accuracy for Alzheimer PET Centiloid prediction. Proposes TSA reporting standard. ADNI + OASIS-3. RO-2026-002.

Deep learning for early Alzheimer's detection from PET + MRI. 1st place, Genç Beyinler Yeni Fikirler 2026. Research showcase — source code not public.

Cortical amyloid-β PET components from the AMYPAD consortium, with a Python tool to register PET to MNI and quantify component loadings by spatial regression.

A PET MOVIE for Interpretable Synthesis of Late-Frame [11C]-PiB PET Images from Early-Frame Counterparts

Nuclide: total information dominance on radiopharmaceuticals. An open, cited knowledge graph of theranostics and diagnostic nuclear medicine: 811 records, static JSON API, MCP server and CLI. Forked from OnCo.