NavEP · Reading & credits
References
The anatomy, projection conventions and catheter-position sources behind the explorer, and exactly what was derived from them.
Anatomy and method sources
-
BodyParts3D anatomical atlas ↗
The Database Center for Life Science (DBCLS). Archive file
isa_BP3D_4.0_obj_99.zip, compatibility version 4.0, from the official LSDB archive. Supplies every heart, vessel, valve, rib and vertebra surface. Licence: CC BY 4.0 (archive licence updated 2025-02-27). -
DICOM XA Positioner Module ↗
DICOM PS3.3, C.8.7.5. Defines the convention used here: positive angles are LAO and cranial.
-
Virtual Coronary Angiography: references and model notes ↗
The template for this project. NavEP follows its approach: BodyParts3D surfaces transformed together into patient space (+X left, +Y head, +Z anterior), a tilt applied before rotation, a point source with perspective, and its illustrative 600 mm source-to-isocentre, 1000 mm source-to-detector and 240 mm field.
-
Standard diagnostic catheter positions & fluoroscopic landmarks ↗
Open Exam Prep, CCI RCES study guide 6.1. Source for the four-catheter montage (HRA, His, CS, RV) and for the RAO 30° / LAO 40–45° roles quoted in the view notes.
Cited sources
Every clinical statement on NavEP paraphrases one of these sources and carries its number, for example [1]. Tap or click a number to see the full reference and the section cited. No text, tables or figures are reproduced. The numbers are the same on every page.
The electrogram tracings on the workflow page and in the quiz are schematic and drawn for NavEP: they are generated from timed events (js/egm.js), not taken from recordings or copied from any published figure. Their timings are chosen to match the cited rules beside them, such as an AH jump of at least 50 ms or a septal VA under 70 ms, and their signal shapes and activation sequences are modelled on the descriptions and example recordings in Murgatroyd et al. (sections 2.1–2.4 and 4.1–4.2), without copying any of them; the worked-case values are fictional. Real signals are noisier and vary more from beat to beat.
Abbreviations
The first use of each abbreviation on a page can be tapped or hovered to show what it stands for. The full list:
Anatomy attribution
BodyParts3D, © The Database Center for Life Science licensed under CC Attribution 4.0 International.
Changes: the selected OBJ surfaces were converted to indexed web geometry and moved together into patient centimetres with one shared transform, . Heart, valve and vessel surfaces are not simplified. Ribs, vertebrae, sternum and diaphragm are clustered on a 2 mm grid to keep the download small. The converter is scripts/compile-anatomy.py. No DBCLS endorsement is implied.
Software and fonts served with the site, so it works on networks that block public CDNs: three.js 0.160.0 (MIT licence, vendor/three/LICENSE), the Inter and IBM Plex Mono typefaces from Fontsource 5.3.0 (SIL Open Font License 1.1, vendor/fonts), and eight Tabler Icons 3.19.0 on the workflow page, drawn inline (MIT licence, vendor/tabler-icons/LICENSE).
| Mesh | Atlas name | File | FMA |
|---|
How catheter positions were placed
Neither the atlas nor the template contains catheter positions, so each catheter is a smooth curve through landmarks computed from the atlas meshes. None are placed by hand. Every EP catheter enters along the atlas IVC centreline, each in its own lane 3.2 mm from it so the shafts don't overlap. The two pacing leads come down the atlas SVC centreline instead. The atlas has no subclavian or brachiocephalic veins, so they start at the top of the SVC. We checked the finished curves with a point-in-mesh test: each catheter tip sits inside the correct cavity mesh. The only samples that fall outside any mesh are where a catheter crosses the tricuspid valve plane, the small gap between the separately supplied CS and RA surfaces, or, for left-sided targets, the interatrial septum.
Left-sided targets are reached through a transseptal sheath. The atlas has no fossa ovalis, so the crossing point is the thinnest part of the interatrial septum: where the RA and LA cavity meshes come closest (5 mm apart in this atlas, thicker than a real fossa). In this model it lies 18 mm behind and 7 mm below the His landmark and 12 mm above the CS ostium, consistent with the source’s description of the fossa as posterior to the His catheter and the dilator tip as above the CS catheter. Beyond the sheath tip the catheter follows a circular arc that leaves along the sheath’s axis and ends on the target, as a deflected tip would. The right veins open just behind the septum, so reaching them takes a loop of more than 180°. The sheath’s drop from the SVC onto the fossa during technique playback follows the source’s description, but the two jumps are timed, not produced by modelled anatomy.
| Landmark | How it was derived |
|---|
Simulated fluoroscopy: what it is and isn’t
The image is a perspective projection from a point source onto a flat detector. The cardiac and great-vessel shadows are uniform silhouettes. Bone, the diaphragm and catheters darken where their surfaces overlap. It is not a physical radiograph: there is no attenuation physics, scatter, cardiac or respiratory motion, lungs or liver. Because the atlas has no whole liver, the area below the diaphragm is not darkened. Catheter shapes start as smooth curves through the landmarks above. Each route is then relaxed to remove kinks that no catheter could make: bending is smoothed while every tip stays on its landmark, the electrode-bearing ends of the CS catheter and LBBAP lead stay on their landmark course, the part in the access vein stays in its lane, a transseptal crossing stays on the septum, and each shaft keeps at least 0.8 mm clear of the catheters placed before it. This is a display smoothing, not a simulation of real catheter mechanics.
The quiz draws its images from the same model. Its explanations reuse the explorer’s view notes and catheter and target descriptions, so they carry the same sources.
This is one atlas anatomy. Real patients vary: chamber size and orientation, the position of the CS ostium, and annulus geometry all differ. Everything here is for education only and does not replace supervised training.