Neural decoding researcher working across brain-computer interfaces, geometric data analysis, and privacy-preserving ML
2
papers here
1
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
10
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—
Data insufficient: no scored value yet. This axis would measure total publication volume, output composition (articles, preprints, venues), and its trajectory over time. The snapshot's raw year counts hint at activity, but output-completeness data is not available for scoring.
Impact—
Data insufficient: no scored value yet. This axis would measure citation-based influence — citations per paper, field-weighted impact, and uptake over time. The snapshot reports h-index 10 and i10-index 10, but the underlying citation record is flagged insufficient for scoring.
Focus100/100
Scores 100/100, the maximum, reflecting a field distribution concentrated entirely in one field: computational neuroscience (brain-machine interfaces). The indexed topic clusters reinforce this — EEG and brain-computer interfaces leads with 10 works, followed by privacy-preserving technologies (6), topological and geometric data analysis (6), emotion recognition (5), and advanced neural network applications (3) — all orbiting one core problem: extracting usable signals from neural data.
Collaboration41/100
Scores 41/100, computed from 7 distinct coauthors in the snapshot — a moderate footprint suggesting small-team work rather than large consortia. The frequent-coauthor list names seven people: Canyang Zhao, Bolin Peng, J. Patrick Mayo, Bing Liu, Antoine Collas, Florent Bouchard, and Bertrand Thirion. In the Evolve catalog, this researcher is a middle author on the latent-alignment BCI paper and first author on the SPD-matrix fMRI benchmark.
Activity70/100
Scores 70/100. The year-by-year counts run 2 (2018), 6 (2019), and 7 (2020) — an early peak — then drop to 1-2 works per year through 2021-2025 before rebounding sharply to 8 works in 2026. The recent three-year window holds 10 works, so the record currently sits in its most active phase since 2020.
Reproducibility100/100
Scores 100/100, but the evidence base is one paper (reproMean 100 across reproPapers 1). The reviewed brain-machine interface study carries a high repro-readiness rating in the Evolve catalog, consistent with the perfect score. With a single assessed paper, however, this maximum should be read as promising rather than established; further reviews could move it.
Research summary
Ce Ju is a researcher affiliated with the Chinese Academy of Sciences whose profile, name-resolved from an OpenAlex bibliographic snapshot, sits in computational neuroscience (brain-machine interfaces) — the sole field in the field distribution — while extending into adjacent methodology: topological and geometric data analysis, privacy-preserving technologies, emotion recognition, and advanced neural network applications. The topic profile is led by EEG and brain-computer interfaces (10 works), followed by privacy-preserving technologies in data (6), topological and geometric data analysis (6), emotion and mood recognition (5), and advanced neural network applications (3). Year-by-year counts show 2 works in 2018, rising to 6 in 2019 and 7 in 2020, dipping to 1-2 per year through 2021-2025, and rebounding sharply to 8 works in 2026, with 10 works in the most recent three-year window. The Evolve catalog holds two 2026 papers: a middle-authorship study on task-conditioned latent alignment for stable cross-session neural decoding in brain-machine interfaces (rated high for repro-readiness), and a first-authorship benchmark of SPD matrix learning for resting-state fMRI connectome prediction, not yet reviewed. Collaboration scores a moderate 41/100 on 7 distinct coauthors, while focus and reproducibility both reach 100/100; the output and impact axes remain unscored pending more complete data.
Collaboration pattern
Collaboration scores 41/100 from 7 distinct coauthors — a moderate footprint consistent with small-team work. The frequent-coauthor list names Canyang Zhao, Bolin Peng, J. Patrick Mayo, Bing Liu, Antoine Collas, Florent Bouchard, and Bertrand Thirion. In the Evolve catalog, Ce Ju appears as a middle author on the latent-alignment BCI paper and first author on the SPD-matrix fMRI benchmark, neither flagged corresponding — suggesting contribution through methodological roles rather than sole-correspondence leadership.
Research keywords
Highlights
Task-conditioned latent alignment stabilizes cross-session neural decoding
First-author benchmark of SPD matrix learning for fMRI connectome prediction
A 2026 arXiv paper benchmarks the external generalization of SPD matrix learning for resting-state fMRI connectome prediction, with Ce Ju in the first author position. It is not yet Evolve-reviewed, so no verdict, keywords, or repro-readiness rating exists for it.
A methods portfolio spanning privacy and topology around the BCI core
Beyond the brain-machine interface core, the topic profile records 6 works on privacy-preserving technologies in data and 6 on topological and geometric data analysis, plus 5 on emotion and mood recognition — a broader methods agenda supporting the neural-decoding work. No catalog papers cover these clusters yet.
Timeline
- 2018Earliest works in the snapshot's year-by-year counts: 2 indexed works appear this year.
- 2019-2020Early peak: 6 works in 2019 and 7 in 2020 — the highest annual totals before 2026.
- 2021-2025Quieter stretch, with 1-2 works per year sustained across 2021-2025.
- 2026Sharp rebound to 8 works, including both Evolve-catalog papers: the task-conditioned latent alignment study for brain-machine interfaces and the first-author benchmark of SPD matrix learning on resting-state fMRI connectomes.
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 are computed from an OpenAlex-sourced snapshot (identity: openalex); the output and impact axes are flagged missing and are not scored.
- The snapshot lists h-index 10 and i10-index 10, but the underlying citation data is insufficient for the impact axis; treat these numbers as indicative only.
- Reproducibility rests on a single assessed paper (reproPapers: 1); only one catalog paper has an Evolve verdict, keywords, and repro-readiness rating.
- Affiliation metadata lists country 'FR' while the institution reads Chinese Academy of Sciences; this profile is name-resolved from public bibliographic data (ORCID 0000-0002-0753-2179).