Medical AI researcher at Radboud Nijmegen: PET/CT lesion segmentation, radiomics, and radiology report evaluation
2
papers here
1
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
6
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—
Not scored: this snapshot lacks a verified total-works series for a confirmed author identity, so cumulative publication volume and its growth trajectory cannot be measured reliably; the axis stays unscored until the record is disambiguated.
Impact—
Not scored: citation counts are not validated in this snapshot, so citation influence and field-normalized impact cannot be estimated; this axis would measure how widely the work is cited relative to the medical imaging field.
Focus100/100
Maximum concentration: the OpenAlex topic profile is dominated by radiomics and machine learning in medical imaging (20), AI in cancer detection (15), and medical image segmentation techniques (12), with MRI in cancer diagnosis (6) — nearly all indexed work sits in one oncological imaging cluster, with a single side branch into radiology report evaluation.
Collaboration53/100
Mid-range score of 53: the record lists 13 distinct coauthors, and the frequent-cocauthor set is a stable, largely overlapping medical-imaging circle — Tao Tan appears twice in the provided list, alongside Jiaju Huang, Hao Yang, Xinyu Ma, Kunyan Cai, Junqiang Ma and others — indicating repeat collaboration within one group rather than a broad external network.
Activity100/100
Maximum activity score, driven by a steep ramp in the OpenAlex counts-by-year: 1 indexed work in 2019, 4 in 2020, 2 in 2021, 3 in 2024, then 10 in 2025 and 19 in 2026, giving 32 works in the last three years — a sharply accelerating 2025–2026 output cycle.
Reproducibility20/100
Low score of 20, based on a single Evolve-assessed paper (reproMean 20, reproPapers 1): the Jev radiology-report study is rated low (closed proprietary model, small benchmarks, no released code), while the first-author AutoPET V segmentation pipeline is rated medium (cross-validation gains shown, but test placeholders and missing ablations).
Research summary
Xinglong Liang is a medical AI researcher at Radboud University Nijmegen (NL) working at the intersection of medical image analysis and clinical language evaluation. The OpenAlex record behind this profile shows a steep activity ramp: 1 indexed work in 2019, 4 in 2020, 2 in 2021, 3 in 2024, 10 in 2025, and 19 in 2026 — 32 works in the last three years — with an h-index of 6 and i10-index of 2. Topics concentrate tightly on radiomics and machine learning in medical imaging (20), AI in cancer detection (15), and medical image segmentation techniques (12), plus MRI in cancer diagnosis (6), giving a maximum focus score. Two 2026 arXiv papers reviewed by Evolve bracket this range. As first author, Liang led a tracer-aware interactive segmentation pipeline for AutoPET V, built on MAE-pretrained STU-Net with organ priors and scribble refinement (reproducibility medium: cross-validation gains shown, test placeholders). As a middle author, Liang contributed to evaluating Jev, a proprietary LLM judge for radiology report factuality on RadEvalX (reproducibility low: closed model, small benchmarks, no code). Collaboration is mid-range (score 53, 13 distinct coauthors), centered on a recurring imaging team including Tao Tan, Jiaju Huang and Hao Yang.
Collaboration pattern
Collaboration is team-centric: 13 distinct coauthors in the OpenAlex record, with the frequent-coauthor list dominated by a stable medical-imaging circle — Tao Tan appears twice, alongside Jiaju Huang, Hao Yang, Xinyu Ma, Kunyan Cai, Junqiang Ma, Shaobin Chen, Yue Sun, Chunyao Lu, Tianyu Zhang and Yunchao Yin. The mid-range score (53) reflects repeat collaboration inside one group rather than a wide external network, a pattern shared by both reviewed papers.
Research keywords
Highlights
First-author tracer-aware interactive segmentation pipeline for AutoPET V
Evaluating a proprietary LLM judge for radiology report factuality
2026 output surge spanning imaging and clinical NLP
OpenAlex indexes 19 works in 2026 (versus 10 in 2025), and both Evolve-reviewed papers fall in this year, covering PET/CT lesion segmentation and radiology report evaluation — evidence of a fast-widening, still concentrated research program.
Timeline
- 2019Earliest year in the current OpenAlex record: 1 indexed work.
- 20204 indexed works appear in the record.
- 20212 indexed works appear in the record.
- 20243 indexed works; the record shows no entries for 2022–2023.
- 2025Output accelerates to 10 indexed works.
- 202619 indexed works so far, including both Evolve-reviewed arXiv papers: the first-author AutoPET V interactive segmentation pipeline and the Jev radiology-report evaluation study.
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.
- Bibliometrics come from an OpenAlex identity record (identity: openalex); author-name disambiguation was not manually verified, so the profile is name-resolved from public bibliographic data only.
- Output and impact axes are unscored because the snapshot lacks a verified works-total and validated citation counts; h-index 6 and i10-index 2 are quoted verbatim from the source snapshot and may lag.
- Only 2 papers carry Evolve full assessments; the remaining indexed works are counted but not individually reviewed, so the reproducibility score rests on a single assessed paper.
- Both reviewed papers are 2026 arXiv preprints; peer-review status and final venues are not reflected in this data.