Load a CT DICOM series, window it and resample to isotropic voxels with every plane previewing, then run an ONNX model that paints each organ into the 3D viewer. QA the surfaces by hand and export an annotated volume plus a per-organ volumetrics report — built for research and model QA, not diagnosis.
The usual organ-segmentation run is a pydicom loader, a SimpleITK resample, an inference script, and a pile of matplotlib slices you squint at afterward. Four tools, three intermediate arrays, and no way to see a bad window until the overlay already looks wrong. On the canvas the same chain is a handful of live nodes, and every one previews. The expensive parts — reading the series, resampling to isotropic, scoring the volume — become cached intermediates, so changing the window at the end costs seconds, not another full pass.
An ONNX segmentation model trained on your organ scheme and a CT series from any scanner that writes standard DICOM. LSD reads spacing and orientation straight from the tags, so nothing needs converting to NIfTI first. Multi-organ or single-structure models load the same way — the viewer colors whatever label set the model emits.
This pipeline is built for reproducible research and model QA — comparing a segmentation to last month's, checking a model against a held-out study, measuring volumes for a cohort paper. It is not a diagnostic device and makes no clinical claim. The value is that every number carries its provenance: the window, the voxel spacing, and the exact model version travel with the export, so a result from six months ago rebuilds identically.
Drag the study folder onto the canvas — a few hundred axial slices land as one @load node keyed by series UID. LSD reads slice spacing, orientation, and rescale tags without decoding every image, so the volume previews in seconds. Scroll the slice slider to confirm the series loaded top-to-bottom in the right order.
Add a @shape windowing node and pick a soft-tissue preset — level 40, width 400. The preview re-renders as you drag the sliders, so organ boundaries snap into contrast instead of washing out. Switch to a bone or lung preset any time; the node remembers each setting and the cached raw volume stays untouched.
CT slices are usually thicker than they are wide, which skews any 3D measurement. Chain a resample @shape node set to 1 mm isotropic voxels and watch the axial, coronal, and sagittal previews line up to true proportions. The resampled volume caches, so the model step downstream never recomputes it.
Drop your ONNX organ model onto the canvas and wire the resampled volume into the @deliver node it becomes. Labels stream back a slice block at a time, and the viewer paints each organ its own color as they arrive. A flipped orientation or wrong label index shows up in seconds — long before the full volume finishes.
Open the 3D viewer and turn each predicted organ into a shaded surface floating inside the windowed volume. Rotate it to catch the classic failures — a liver that leaks into the ribcage, a kidney with a hole punched through it. Toggle organs on and off and scrub the slice overlay to confirm the surface tracks the boundary on every plane.
Finish with a @deliver export node. Write the masks back into a DICOM-compatible segmentation volume so downstream tools read it natively, and emit a per-organ volumetrics CSV — millilitres per structure, plus the window, spacing, and model version behind each number. Re-running on the next study reuses every cached setting.
Index a full axial series with slice spacing, orientation, and rescale tags intact.
Apply an intensity window and resample to isotropic voxels, previewing each plane as you go.
Run ONNX segmentation, QA the organs in 3D, then export an annotated volume and a volumetrics report.
Screen 400 MRI studies before analysis: harmonize metadata, check orientation and spacing, flag motion, and export a pass/fail QA table and exclusion list.
Model post-closure pit-lake rebound in LSD — krige borehole heads against a NetCDF recharge grid and deliver a live 3D drawdown model plus report.
Validate 63 LAS drillhole logs against collar surveys, krige a lithology block model, and export one glTF the whole team can open.
Build churn features from Postgres events and Parquet snapshots, catch a leaky window visually, and score the holdout with ONNX — on one canvas.
Score a million-row Parquet batch with an ONNX model — training-identical normalization, live previews, scored Parquet and metrics out.
Turn raw LAS/LAZ sweeps from a mobile robot into a nav-ready occupancy grid and 3D map — registration to export with zero scripts.
Sync CSV sensor logs with live REST telemetry, flag outliers and dropouts, and score every robot pass/fail before the field trial.
Turn folders of line-scan images and a label CSV into an ONNX defect-segmentation pipeline with overlay QA and a per-coil defect report.
Turn GeoTIFF orthomosaics into a stitched land-cover raster with per-class area stats, using tiled ONNX segmentation you can preview at every step.
Align CMM measurement points to a nominal CAD mesh, compute point-to-surface deviation, and export a color-mapped glTF model with a pass/fail report.
Resample a week of machine sensor logs, baseline each tag, and flag drift and step changes on a live dashboard with a ranked maintenance shortlist.
Join a live meter feed to 85 campus building meshes, aggregate it hourly, and paint a 3D heat-map that repaints as energy use shifts, plus a weekly report.
Drag the study folder onto the canvas — a few hundred axial slices land as one @load node keyed by series UID. LSD reads slice spacing, orientation, and rescale tags without decoding every image, so the volume previews in seconds. Scroll the slice slider to confirm the series loaded top-to-bottom in the right order.
CT slices are usually thicker than they are wide, which skews any 3D measurement. Chain a resample @shape node set to 1 mm isotropic voxels and watch the axial, coronal, and sagittal previews line up to true proportions. The resampled volume caches, so the model step downstream never recomputes it.
Drop your ONNX organ model onto the canvas and wire the resampled volume into the @deliver node it becomes. Labels stream back a slice block at a time, and the viewer paints each organ its own color as they arrive. A flipped orientation or wrong label index shows up in seconds — long before the full volume finishes.
Open the 3D viewer and turn each predicted organ into a shaded surface floating inside the windowed volume. Rotate it to catch the classic failures — a liver that leaks into the ribcage, a kidney with a hole punched through it. Toggle organs on and off and scrub the slice overlay to confirm the surface tracks the boundary on every plane.