Senior (last) author on generative AI for 3D geological microstructures and few-shot whole-slide image analysis
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Insufficient data: this axis would measure aggregate publication volume and productivity over time (paper counts by year and venue), but the bibliometric record for this profile contains no such counts yet, so no score is given.
Impact—
No impact data available. This axis would capture citation-based influence — citations, h-index, and citing works — none of which are present in the current record, so the dimension is intentionally left unscored.
Focus100/100
Focus scores 100 because the attested field distribution contains exactly one field — computational materials science (porous media / digital rock) — with count 1. Both cataloged 2026 papers also sit in senior last-author positions on generative-model applications, from 3D geological microstructure synthesis to whole-slide image analysis under few-shot annotation, so the profile reads as sharply concentrated on generative modeling applied to scientific imaging.
Collaboration41/100
The radar collaboration dimension scores 41, built on a raw count of 7 distinct co-authors in the current snapshot. The recurring set spans Ali Aouf, Eric Laloy, Bart Rogiers, Tiffanie Godelaine, Maxime Zanella, Karim El Khoury, and Benoit Macq, indicating a stable but comparatively compact working circle spanning both the porous-media generation work and the whole-slide imaging study; wider network breadth cannot be assessed from the present data.
Activity—
Data insufficient. This axis would track recent publishing activity — annual output trends and recency of contributions — but the snapshot provides no per-year productivity series beyond the two cataloged 2026 papers, so the dimension stays unscored.
Reproducibility55/100
Reproducibility scores 55, computed as the mean readiness rating (reproMean = 55) across 1 reviewed paper (reproPapers = 1). That paper, DimExDAM, is rated medium: the verdict finds its results across five porous media convincing but flags thin ablations and diversity tests, so the score reflects credible core results with incomplete experimental validation rather than fully verified methodology.
Research summary
Christophe De Vleeschouwer is a senior researcher whose cataloged record positions him as last author on applied generative-modeling studies. The Evolve-reviewed paper DimExDAM (Generating Heterogeneous 3D Geological Microstructures from 2D Images via a Stable Diffusion-Adversarial Model, arXiv, 2026) proposes a hybrid denoising-diffusion and adversarial pipeline that reconstructs heterogeneous three-dimensional microstructures from two-dimensional slices, evaluated across five porous-media cases including Boom Clay; Evolve finds the results convincing but notes that ablation and diversity testing is thin, and rates reproducibility readiness as medium. A second 2026 arXiv paper, Whole-Slide Image Analysis under Realistic Few-Shot Annotation Protocols, extends his activity into computational-pathology imaging, again from the last author position. The current bibliometric snapshot records 7 distinct co-authors — Ali Aouf, Eric Laloy, Bart Rogiers, Tiffanie Godelaine, Maxime Zanella, Karim El Khoury, and Benoit Macq — and all attested field activity concentrates in computational materials science for porous media and digital rock. Radar dimensions give focus 100 and reproducibility 55; output, impact, and activity axes lack sufficient data and are left unscored. No institutional affiliation or country is present in the source record.
Collaboration pattern
The collaboration dimension (41) rests on 7 distinct co-authors in the current snapshot: Ali Aouf, Eric Laloy, Bart Rogiers, Tiffanie Godelaine, Maxime Zanella, Karim El Khoury, and Benoit Macq. This is a compact, apparently stable circle pairing machine-learning contributors with porous-media and geological-modeling specialists, consistent with a senior researcher supervising applied generative-modeling projects rather than operating a large, rotating network.
Research keywords
Highlights
DimExDAM: stable diffusion-adversarial hybrid for 2D-to-3D geological microstructure generation
Senior-author entry into computational pathology imaging
A 2026 arXiv study on whole-slide image analysis under realistic few-shot annotation protocols, taken from the last author position; Evolve review metadata (verdict, keywords, reproducibility rating) is not yet available.
Consistent last-author leadership across both cataloged 2026 papers
Both records in the Evolve catalog list De Vleeschouwer in the last author position with no corresponding-author flag, suggesting supervision-level oversight of generative-model application work rather than hands-on first-author experimentation.
Timeline
- 2026arXiv preprint on generating heterogeneous 3D geological microstructures from 2D images via a stable diffusion-adversarial model (DimExDAM), with De Vleeschouwer as last author.
- 2026arXiv preprint on whole-slide image analysis under realistic few-shot annotation protocols, again from the last author position.
- 2026Both cataloged outputs land in 2026, concentrating the visible record in a single year and in generative modeling applied to scientific imaging.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric dimensions are computed from OpenAlex-derived records only; citation, output-volume, and activity series are absent, so the output, impact, and activity axes are unscored.
- Institution and country are not present in the source record, so no affiliation, geography, or career-history claim is made; the profile resolves to ORCID 0000-0001-5049-2929.
- The whole-slide imaging paper carries no Evolve verdict, keywords, field tag, slug, or reproducibility rating yet; it is assessed only at title level.
- The reproducibility score (55) averages a single reviewed paper (reproPapers = 1), so it is indicative rather than an established track record.