NTUST researcher applying deep learning to ultrasound-guided intervention and LLM-based clinical decision support.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
2
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—
Output cannot be scored yet. The OpenAlex record shows only sparse per-year counts (one work in 2025, six in 2026, plus isolated earlier entries), too thin to summarize productivity; this axis would measure total works, publication venues, and authorship roles.
Impact—
Impact cannot be scored yet. Available citation indicators are minimal (h-index 2, i10-index 0) and both catalog papers are recent 2026 preprints, so citation evidence has not accumulated; this axis would measure citations, usage, and downstream scholarly uptake.
Focus100/100
The most decisive signal in this profile. OpenAlex topic assignment concentrates almost entirely on health-facing AI: Artificial Intelligence in Healthcare and Education (4 works), Multimodal Machine Learning Applications (3), and Machine Learning in Healthcare (3), plus Topic Modeling (2) and Medical Image Segmentation Techniques (1). The reviewed field list contains a single area — ultrasound-guided intervention with computer vision — supporting the maximum focus score of 100.
Collaboration53/100
Collaboration sits at 53. The recent window lists 8 works with 13 distinct coauthors, and 12 recurring collaborators appear in the profile, including Yung Wei Shueh, Zhi-Jie Chen, and Donghua Zhang. Both cataloged 2026 papers place Hsu in a middle, non-corresponding position, suggesting contribution within larger team efforts rather than an independent group leadership role so far.
Activity64/100
Activity scores 64, driven by recency: countsByYear shows 2 works in 2025 rising to 6 in 2026, and recent3yWorks totals 8 — meaning most recorded output lands in the current year. Two isolated entries in 2002 and 2017 also appear in the record and are more plausibly bibliographic noise than sustained earlier activity.
Reproducibility55/100
Reproducibility scores 55 from a single scored paper (reproPapers 1, reproMean 55), so the signal is narrow. Evolve judged STUNet-Fusion medium readiness — promising phantom results but best-case reporting without significance tests — and DIASENTINEL low: honest reporting, yet private data, no code release, and a weaker AUPRC than an XGBoost baseline.
Research summary
Chia-Hsuan Hsu is an early-career researcher at National Taiwan University of Science and Technology; this profile is name-resolved from OpenAlex bibliographic data. The topic record clusters tightly around applied AI for health: Artificial Intelligence in Healthcare and Education (4 works), Multimodal Machine Learning Applications (3), and Machine Learning in Healthcare (3), with smaller traces of Topic Modeling (2) and Medical Image Segmentation Techniques (1). Recent activity is accelerating — 8 works in the last three years, six of them recorded in 2026. Two Evolve-reviewed arXiv preprints show the range: STUNet-Fusion (2026), a ConvLSTM/U-Net multi-channel motion-fusion network for needle-tip localization in ultrasound video, and DIASENTINEL (2026), an auditable multi-agent LLM system for guideline-grounded type 2 diabetes risk screening on electronic health records. In both, Hsu is a middle, non-corresponding author. Evolve's verdicts are cautiously positive: promising phantom results but best-case reporting for the first; honest reporting, though private data, no code, and an XGBoost baseline beating it on AUPRC for the second. Indicators remain thin — h-index 2, i10-index 0, reproducibility mean 55 from one scored paper, and no computable output or impact scores.
Collaboration pattern
Hsu works inside multi-author teams: both 2026 catalog papers are middle-author, non-corresponding contributions. The recent window spans 8 works with 13 distinct coauthors, and 12 frequent collaborators recur — including Yung Wei Shueh, Zhi-Jie Chen, Hsin-Ling Hsu, Donghua Zhang, and Chenwei Wu — consistent with project-based teamwork across ultrasound intervention and clinical NLP rather than a single long-standing lab line. No solo-authored works appear in the reviewed set.
Research keywords
Highlights
Spatiotemporal needle-tip localization in ultrasound video (STUNet-Fusion)
Auditable multi-agent LLM system for diabetes risk screening
Timeline
- 2025OpenAlex records 2 works in 2025, the earliest recent activity in this profile.
- 2026STUNet-Fusion, on spatiotemporal needle-tip localization in ultrasound video, appears as an arXiv preprint with Hsu as a middle author.
- 2026DIASENTINEL, an auditable multi-agent system for guideline-grounded diabetes risk screening, is released as a second 2026 arXiv preprint; OpenAlex records 6 works in this year.
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 drawn from OpenAlex only (identity: openalex); no other citation aggregator or publisher data is included.
- Profile is name-resolved from public bibliographic data; 'Chia-Hsuan Hsu' is a common Taiwanese romanization, and the isolated 2002 and 2017 countsByYear entries may reflect same-name records rather than this researcher.
- Both Evolve-reviewed papers are 2026 arXiv preprints; peer-reviewed venue status is not established in the provided data.
- Citation indicators are minimal (h-index 2, i10-index 0), and the output and impact dimensions lack sufficient data to score.