Pulmonary Embolism Detection from CT Pulmonary Angiography (PWDR) L1-525

Medical ImagingPulmonary vascular contrast-enhanced reconstruction with embolism location + RV-strain severity readoutδ=5 · advancedL_DAG = 7📋 Stub — not mineable

📋

Unclaimed Principle — open for contribution

This Principle is declared in the catalog but has no reference solver, no pinned dataset, and is not registered on-chain. There is no reward pool. Submitting a cert against this Principle today will record the cert for reproducibility but pay zero PWM.

To claim it as a Bounty #7 contribution: open a PR adding (1) a reference solver, (2) ≥1 dataset pinned to IPFS, (3) updates to the L3 manifest with dataset CIDs. After verifier-agent triple-review, the founders' 3-of-5 multisig signs PWMRegistry.register() and the Principle becomes mineable.

Read the contribution guide →Open a claim issue →

Forward model E

S t a g e 1 (L 1 - 029, C T P A) : y_{p} r o j \to m u_{a} tt e n u a t i o n (r); p u l m o na r y_{a} r t er y_{s} e g m e n t a t i o n (r) \to t r e e_{s} t r u c t u r e; co n t r a s t_{f} i l l in g (b r an c h) \to d e f ec t_{i} n d i c a t or . S t a g e 2 : ana t o mi c_{c} l a ss = s t e p (d e f ec t_{l} oc a t i o n_{i} n_{t} r ee); R V / L V_{r} a t i o = ma x_{d} iam e t e r_{R} V / ma x_{d} iam e t e r_{L} V; se v er i t y = E S C 2019 r i s k_{s} t r a t i f i c a t i o n (ana t o mi c_{c} l a ss, R V / L V, t r o p o nin, h e m o d y nami cs) .

Pulmonary Embolism Detection from CTPA (PWDR): wraps L1-029 CT analytical core (in CTPA acquisition mode) with established radiology grading rules. Stage 1 (analytical, from L1-029): from contrast-enhanced volumetric CT acquired during peak pulmonary-arterial enhancement (timing bolus or test bolus protocol), recover voxel-resolved iodinated-contrast distribution in pulmonary vasculature; segment pulmonary arterial tree (down to subsegmental level); detect intraluminal filling defects; measure RV and LV chamber diameters at maximum dimension. Stage 2 (deterministic threshold): classify defect location into anatomic level; compute Mastora / Qanadli obstruction index; assess RV strain via RV/LV ratio; determine severity tier per ESC 2019 / AHA 2011 risk stratification. Difficulty tier delta = 5. Mismatch parameters: contrast_bolus_timing, motion_during_breath_hold, partial_volume_at_subsegmental, beam_hardening_artifact, AI_threshold_calibration, RV_LV_measurement_axis.

L-DAG

L.xray_source -> L.attenuation_projection -> L.dynamic_reconstruction -> L.pulmonary_artery_segmentation -> L.filling_defect_detection -> L.rv_lv_measurement -> L.pe_severity_classifier -> int.spatial

L.xray_sourceL.attenuation_projectionL.dynamic_reconstructionL.pulmonary_artery_segmentationL.filling_defect_detectionL.rv_lv_measurementL.pe_severity_classifierint.spatial

Well-posedness W

Existence:: true
Uniqueness:: conditional
Stability:: conditional
κ:: 100

Existence inherited from L1-029. Uniqueness conditional on adequate contrast bolus timing (peak pulmonary-arterial enhancement). Stability conditional with contrast_bolus_timing dominant for false-negative rate at subsegmental level (~15-25% miss rate for isolated subsegmental PE). Joint Hadamard well-posedness for the CTPA + PE-detection forward established by Wittram 2007 (foundational PE imaging review), Quanadli 2001 (Qanadli obstruction index), Mastora 2003 (Mastora score), Konstantinides 2019 (ESC 2019 PE guidelines), Jaff 2011 (AHA scientific statement), Stein 2006 (PIOPED II).

Solvability C

Solver class:: linear-operator + image-segmentation [L1-029 reconstruction + pulmonary artery tree segmentation + filling defect detection + RV/LV measurement] + categorical-readout [ESC 2019 severity classifier] | end-to-end deep neural [DeepEmbolus, Aidoc PE, RapidAI PE] with explicit physics-informed vascular regularization
Convergence rate q:: 2
Complexity:: O(H * W * Z * log(N)) for stage 1 reconstruction; O(N_arterial_branches) for stage 2 tree analysis

Specs (0)

No L2 specs registered yet for this principle.