W. J. Zhang

—

Unclaimed · AI-structured profileOpenAlex

Studies surface-EMG gesture decoding with small calibration budgets; last author on 2026 transfer-learning preprints.

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-28) 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—

Output measures cumulative publication volume and productivity over time. This axis is not yet scoreable: the snapshot records only 3 works, all dated 2026 — too thin to characterise a productivity trajectory or compare against field norms.

Impact—

Impact measures citation-derived influence such as h-index and i10-index, which this snapshot shows as 0. With so few citing works on a nascent record, there is not enough evidence to score influence, so the axis stays unscored.

Focus—

Focus measures topical concentration and consistency across a career. The snapshot's topic labels — muscle activation and electromyography, sensor and energy-harvesting materials — hint at an EMG-centred agenda, but the data are too sparse to compute a reliable score.

Collaboration14/100

The collaboration axis scores 14/100 on a sparse OpenAlex record that lists exactly 1 distinct co-author. Jethro Odeyemi recurs as the frequent collaborator, and W. J. Zhang occupies the last author position on both Evolve-reviewed 2026 preprints — a pattern consistent with a compact, stable two-person team rather than a broad network. Because the record is thin, the score may understate collaborations not captured in the index.

Activity40/100

Activity scores 40/100. The raw snapshot counts 3 works in 2026 (countsByYear), matching recent3yWorks = 3, so all recorded output falls inside the recent three-year window and clusters in a single year. Two of those works are the Evolve-reviewed arXiv preprints on surface-EMG gesture decoding. A 2026-only record suggests an active but just-emerging publishing rhythm; multi-year counts would be needed to confirm sustained productivity.

Reproducibility—

Reproducibility would measure code, data, and protocol sharing. Evolve's reviews note no released code and medium reproducibility readiness for both papers, but no systematic index exists yet, so this axis remains unscored rather than estimated.

Research summary

W. J. Zhang works on surface electromyography (EMG) decoding for gesture recognition and myoelectric control, with a distinctive emphasis on making cross-user models usable under small calibration budgets. The Evolve catalog holds two 2026 arXiv preprints with Zhang in the last author position, neither flagged as corresponding author. The first studies intact-to-amputee transfer in surface-EMG gesture decoding across the DB2 and DB3 databases, comparing training sources and calibration budgets; Evolve's review calls it a well-matched study in which intact-body pretraining beats scarce amputee data once calibrated, while flagging an 11-participant sample, a pool-size confound, and no released code. The second personalises a cross-user surface-EMG encoder on NinaPro, reporting that full fine-tuning wins at every calibration budget while gradient-free prototypical approaches recover 52–78% of its gain at near-zero cost. The OpenAlex snapshot records 3 works in 2026 (h-index 0, i10-index 0), with topics led by advanced sensor and energy-harvesting materials and muscle-activation/electromyography studies, plus balance and falls prevention, EEG-based brain–computer interfaces, and motor control. Radar scores collaboration at 14 and activity at 40, reflecting a sparse record with one distinct co-author; output, impact, focus, and reproducibility lack sufficient data to score. Because the name is initial-based and common, this profile is name-resolved from public bibliographic data, with no institution, country, or ORCID available.

Collaboration pattern

The record shows an extremely compact collaboration footprint: exactly 1 distinct co-author (raw distinctCoauthors), with Jethro Odeyemi as the recurring partner. On both Evolve-reviewed 2026 preprints W. J. Zhang sits in the last author position without being the corresponding author, a pattern consistent with a small, stable two-person team in which the last author provides senior guidance. The collaboration radar score of 14 reflects this sparsity and may understate collaborations not captured by OpenAlex.

Research keywords

surface electromyographygesture recognitionmyoelectric controltransfer learningcalibration budgetfew-shot adaptationprototypical networksprosthetic controltransradial amputationcross-user encoder personalization

Highlights

Intact-to-amputee transfer works — with calibration
Across the DB2→DB3 setting, Evolve's review finds that intact-body pretraining outperforms scarce amputee data once calibrated, while honestly noting an n=11 sample, a pool-size confound, and no released code.
Fine-tuning wins every calibration budget
On NinaPro, full fine-tuning of a cross-user surface-EMG encoder beat every alternative at each calibration budget; gradient-free prototypes recovered 52–78% of its gain at near-zero cost.
A consistent calibration-budget research line
Both 2026 preprints frame the same question — how much per-user data does myoelectric decoding need? — with W. J. Zhang as last author on each, consistent with a guiding role in the compact team with Jethro Odeyemi.

Timeline

2026arXiv preprint on intact-to-amputee transfer in surface-EMG gesture decoding (DB2→DB3, training source and calibration budget) enters the Evolve catalog, with W. J. Zhang as last author.
2026arXiv preprint on personalising a cross-user surface-EMG encoder under a small calibration budget (NinaPro) is reviewed by Evolve, again with W. J. Zhang as last author.
2026OpenAlex snapshot records 3 works in 2026 (recent3yWorks = 3), all clustered in this single year — an output record that is currently active but just emerging.

Papers in the Evolve catalog

Frequent co-authors

Jethro Odeyemi×2

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.

Identity is name-resolved from OpenAlex bibliographic records; no ORCID, institution, or country is available in the input data.
Bibliometrics come from OpenAlex only; this snapshot shows h-index 0 and i10-index 0, and the citation-impact axis is left unscored.
The Evolve catalog holds only 2 reviewed arXiv preprints, both dated 2026; no peer-reviewed venues or earlier career history are captured.
Radar collaboration (14) and activity (40) scores derive from a sparse record with 1 distinct co-author and may understate the researcher's actual footprint.