---
title: "A VTK-based, CUDA-optimized non-parametric vessel detection method"
authors:
  - "Levent Alpoge"
  - "Alark Joshi"
  - "Dustin Scheinost"
  - "John Onofrey"
  - "Xiaoning Qian"
  - "Xenios Papademetris"
year: 2010
venue: "The VTK Journal"
arxiv: null
doi: "10.54294/z1w0uu"
source: "https://doi.org/10.54294/z1w0uu"
topics:
  - "number-theory"
seed_rank: 1219
seed_batch: "alpoge-math-2026-08-24"
reviewed: "2026-08-24"
pool: "maths-foundations"
relevance_score: 8
lineage: arithmetic-geometry
cites:
  []
see:
  - "1197-the-3-4-infinity-modular-family-of-2-tori-completed-at-its-thr"
---

# A VTK-based, CUDA-optimized non-parametric vessel detection method

## One-sentence takeaway

Intel STS / high-school paper: non-parametric vessel detection in medical images, implemented in VTK and optimized in then-new CUDA — the "CUDA OG" artifact, and the only CS paper he still links.

## Why it's lovely

Homepage: "my first ever work as a kid introduced a brand new language called cuda for gaming hardware to a medical computer vision problem." Hundred gigaflops, hand over rmem/smem, no ICI.

## Problem

Detect vessels in medical volumes without a parametric tube model, fast enough to be useful.

## Design

VTK pipeline plus CUDA kernels for the non-parametric detector.

## Evidence

The VTK Journal (2010), DOI 10.54294/z1w0uu. Intel STS finalist. No arXiv.

## Limitations

Not mathematics. He says none of the later CS work is listed. Still a sourced research paper.

## Implications for Broadside

Biographical root of the ingest. Do not expand into a medical-imaging shelf.

## Bottom line

Mint the VTK Journal paper.

## Links

- DOI: [10.54294/z1w0uu](https://doi.org/10.54294/z1w0uu)