University College Dublin researcher combining CT radiomics and foundation models for renal tumour classification
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-02) 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 snapshot contains no publication counts or productivity metrics beyond the two reviewed 2026 papers, so no output score can be computed. This axis would measure aggregate publication volume, output type and cadence across years.
Impact—
No impact data is available: the snapshot has no citation counts, h-index or altmetrics, so influence cannot be assessed. This axis would measure citations, downstream adoption and clinical uptake of the work once such data exists.
Focus0/100
The focus score is 0 on the 0-100 radar, matching an even 50/50 field distribution between the two cataloged subfields — computational radiology and radiomics plus foundation models. The split reads as method-level breadth rather than topic drift: both halves target renal cell carcinoma classification on CT, one via handcrafted radiomics fused with RenalCLIP or MedVAE representations, the other via comparative encoding studies.
Collaboration22/100
Collaboration scores 22/100, computed from just 2 distinct coauthors in the cataloged record — Yuan Liang and Sourav Bhattacharjee recur across both 2026 papers. The pattern indicates a compact, stable core team rather than a broad network. Campbell holds the last author position on both papers, with corresponding-author status not flagged, consistent with a supervisory role within this small group.
Activity—
Activity cannot be scored: the record spans only the 2026 arXiv entries in Evolve's catalog, with no longitudinal publication history. This axis would measure publishing cadence, venue spread and sustained engagement over time.
Reproducibility55/100
Reproducibility scores 55/100, a mean taken over the 2 reviewed papers, both rated medium readiness. Evolve's verdicts mark the boundaries: the 0.804/0.854 AUC margins over simpler fusion and a strong SVC are small and untested, and the 82.7% KiTS23 result rests on a 60-case test set with no significance testing — plausible claims that still need external validation.
Research summary
Abraham Campbell is a researcher at University College Dublin whose cataloged work sits in medical image analysis, specifically computational radiology and the combination of handcrafted radiomics with foundation-model representations for renal tumor characterization on CT. Evolve's reviewed catalog contains two 2026 arXiv papers, both with Campbell in the last author position. The first conditions RenalCLIP foundation features with radiomics through FiLM modulation for clear cell renal cell carcinoma classification, reporting AUC values of 0.804 and 0.854; Evolve judges the margins over simpler fusion strategies and a strong SVC baseline to be small and untested. The second compares 2D and 3D CT encodings and reports that 3D gated fusion of radiomics with MedVAE representations reaches 82.7% AUC on the KiTS23 renal cell carcinoma cohort, again outperforming either representation alone — a plausible claim that rests on a 60-case test set without significance testing. The bibliometric record is compact: 2 distinct coauthors (Yuan Liang and Sourav Bhattacharjee), a collaboration score of 22, an even 50/50 split across the two cataloged subfields, and a reproducibility score of 55 averaged over the two reviewed papers, both rated medium readiness. Output, impact and activity metrics are not yet available in the snapshot, so this should be read as an early-stage profile, name-resolved from public bibliographic data.
Collaboration pattern
Collaboration is tight: only 2 distinct coauthors appear in the record — Yuan Liang and Sourav Bhattacharjee — and both recur on each 2026 paper. Campbell sits in the last author position on both, with corresponding-author status not flagged, suggesting a supervisory role over a small stable core team. The collaboration score of 22/100 reflects this narrow network rather than a broad set of partners.
Research keywords
Highlights
Radiomics-conditioned FiLM modulation of RenalCLIP features for ccRCC CT classification
Radiomics and foundation representations shown complementary on KiTS23
Repeated last-author positioning across the 2026 renal imaging studies
Both cataloged 2026 papers place Campbell in the last author position with corresponding status not flagged, working with the same two recurring coauthors — a compact, repeatable team arrangement centred on radiomics and foundation-model fusion for renal tumours.
Timeline
- 2026Published the radiomics-conditioned FiLM modulation of RenalCLIP features study for clear cell renal cell carcinoma CT classification on arXiv, in the last author position.
- 2026Published the comparative study of 2D and 3D CT encodings on arXiv, reporting 82.7% AUC for radiomics + MedVAE 3D gated fusion on KiTS23.
- 2026Concentrated both cataloged outputs within a single year, with the same two coauthors (Yuan Liang, Sourav Bhattacharjee) and a shared radiomics and foundation-model fusion theme.
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 signals come from Evolve's reviewed catalog only: two 2026 arXiv papers; no ORCID or country data was provided, and the profile is name-resolved from public bibliographic data.
- Output, impact and activity scores are missing because the snapshot contains no citation counts, publication totals or longitudinal record.
- The reproducibility score (55) is averaged over only 2 reviewed papers, both medium readiness — too small a base for a stable estimate.
- Performance numbers quoted (AUC 0.804/0.854, 82.7% AUC, 60-case test set) are as reported by Evolve's verdicts and have not been independently verified.