totalsegmentator

TotalSegmentator 2D: A Tool for Rapid Anatomical Structure Analysis

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

Swift Horos/OsiriX plugin that exports active CT/MR series, provisions an isolated Python TotalSegmentator environment, runs segmentation, and imports RT-Struct overlays back into the host viewer.

Desktop CT imaging workstation (PySide6/Qt6) with first-principles CT reconstruction, AI organ segmentation, and quantitative validation. Teaching/research only; not a medical device.

Peer-reviewed pipeline for automated CT-based pelvimetry and mid-pelvic workspace quantification in rectal cancer.

Open, modular MCP server for medical image analysis agents: containerized models (MedSAM2, SAM 2.1, TotalSegmentator, lungmask, HD-BET, SynthStrip, nnU-Net, MONAI, TorchXRayVision), preset/custom guidelines, and a code-customizable viewer.

Bone Mineral Density estimation from CT scans using PyTorch

Staged Dice/IoU benchmark of MOOSE, TotalSegmentator & VoxTell on public CT datasets. Reproducible, seed-fixed, auto organ-matching

Automated analysis of lumbar discs, integrating TotalSegmentator for anatomical localization, PyRadiomics for quantitative feature extraction, and machine learning models for clinical classification (Pfirrmann grading).

Batch medical-imaging pipeline that ingests 3D CT/MRI volumes to Backblaze B2, runs TotalSegmentator locally to produce multi-label masks over 100+ anatomical structures plus per-structure volumetrics, and writes masks and stats back to B2. Next.js 16 + FastAPI sample app over the B2 S3-compatible API.

Resumable, config-driven DICOM → NIfTI pipeline with body-region and contrast-phase classification (BOA/TotalSegmentator)

Research tool for continuous NCC-to-LV-myocardium membranous-septum candidate segmentation on cardiac CTA

Independent, contamination-aware, statistically-rigorous leaderboard for medical image segmentation models: every score with a bootstrap confidence interval and a ranking-stability test, sliced by failure mode, reproducible with one command. Powered by segauge.

Rule-based measurement of eight cardiovascular diameters from TotalSegmentator masks on non-ECG-gated contrast-enhanced chest CT

Apple Silicon MacでCTから顎骨・歯の3Dデータを作成する研究・教育用アプリ

Research desktop app for DICOM CT visualization, vertebral segmentation, and interactive pedicle screw planning.

Python package and CLI for automated CT pelvimetry and body-composition analysis with reproducible QC outputs.

Automatic identification of CT contrast subphases. The final pipeline was developed during the master thesis project "Two-Stage CT Subphase Classification Using OrganLevel Features" (El-Zein, Nora, 2026)

How much can organ shape alone reveal? Using 1,228 whole-body CTs (16 organs, intensity discarded), we predict sex (AUC 0.87), age (MAE 9.3 yr), and show shape is near-unique per person. Hip bones carry the strongest signal. Shape outperforms intensity for sex — the signal is geometric. Code + reproducible pipeline.

Agent-based DICOM-to-FEBio pipeline for medical-image segmentation, mesh generation, finite-element model construction, solver execution, and validation.