Physics-informed deep learning across ultrasound, neuro-oncology, and gait biomechanics at Yonsei University
4
papers here
4
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
9
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 yet scored: publication-volume data is insufficient in the current snapshot. This axis would measure total scholarly output — cumulative works, yearly counts, and distribution of authorship roles across the career.
Impact—
Not yet scored: citation data is insufficient on this axis. It would measure aggregate citations, citation velocity, and field-normalized influence of this researcher's publications relative to field peers.
Focus0/100
Scores 0/100, reflecting maximal breadth: the OpenAlex field distribution splits evenly at 25% across scientific ML/computational bioheat transfer, ultrasound deep learning, AI neuro-oncology, and clinical gait biomechanics. Topic counts still cluster — Medical Image Segmentation Techniques (12) and Ultrasound and Hyperthermia Applications (11) lead — so the pattern reads as physics-informed ML exported across clinical problems rather than unfocused drift.
Collaboration36/100
Scores 36/100 on a small-team pattern: 5 distinct coauthors in the last three years, with Kyungho Yoon, Dongyun Lee, Junkyung Ju, Yeonjae Jung, and Nayoung Son recurring. All four Evolve-reviewed papers place this researcher in the last author position, none as corresponding author — consistent with a senior supervisory role within a compact group rather than large-consortium work.
Activity91/100
Scores 91/100, the strongest axis. OpenAlex logs 22 works in the last three years and per-year counts of 10 (2023), 7 (2024), 5 (2025), and 10 in 2026 — with all four Evolve-reviewed arXiv preprints landing in 2026 alone. A marked acceleration follows the quiet 2020–2022 stretch (2, 2, 2 works), indicating a late but steep productivity ramp.
Reproducibility78/100
Scores 78/100, with a reproducibility mean of 78 computed across 4 reviewed papers. Two efforts rate high readiness — the GRF gait Transformer with ε-LRP maps and Blender viewer, and the OASA glioma forecaster — while the k-Wave synthetic-ultrasound and PPIM bioheat papers rate medium, mainly for missing significance tests, baselines, or ablation/variance reporting.
Research summary
Minwoo Shin (Yonsei University, South Korea; ORCID 0000-0002-0093-2866) is profiled here from public OpenAlex bibliographic data and Evolve's own reviews of four 2026 arXiv preprints. The record shows a late-stage productivity surge: 22 works in the last three years, with yearly counts rising from a quiet 2020–2022 stretch (2, 2, 2) to 10 (2023), 7 (2024), 5 (2025), and 10 in 2026 — the year all four reviewed papers appeared. An h-index of 9 and i10-index of 9 indicate an early citation footprint, and the output and impact dimensions are not scored for lack of data. Topical mass sits in Medical Image Segmentation Techniques (12), Ultrasound and Hyperthermia Applications (11), Photoacoustic and Ultrasonic Imaging (9), Advanced Vision and Imaging (7), and Gas Dynamics and Kinetic Theory (6), while the field distribution splits evenly (25% each) across computational bioheat and scientific ML, ultrasound deep learning, AI neuro-oncology, and clinical gait biomechanics — hence a focus score of 0. All four reviewed papers carry last-author placement, none corresponding, within a five-person coauthor circle; collaboration scores 36. Reproducibility is the quiet strength (78/100 across four papers): pipelines, explainability maps, and honest reporting recur even where gains over baselines lack significance tests. The through-line is physics-informed deep learning — Pennes bioheat solvers, k-Wave ultrasound simulation, MRI-based data assimilation — applied to clinical imaging and biomechanics problems.
Collaboration pattern
The collaboration footprint is compact: 5 distinct coauthors in the last three years and a collaboration score of 36. Kyungho Yoon, Dongyun Lee, Junkyung Ju, Yeonjae Jung, and Nayoung Son recur across efforts. On all four Evolve-reviewed papers Shin appears as last author and never as corresponding author, consistent with supervising a small Yonsei-based group whose members take first-author roles rather than joining large consortia.
Research keywords
Highlights
PPIM: Pennes physics-informed Mamba for 3D bioheat simulation
Physics-guided synthetic high-frequency ultrasound for skin segmentation
Transformer reaches 90.07% on 12-class GRF gait disorder classification
OASA: observation-anchored forecasting of post-treatment glioma states
Timeline
- 2007OpenAlex logs the first 2 indexed works; output then goes quiet until 2018.
- 2018Activity resumes with 5 works (3 in 2019), starting a steady but modest stream.
- 2023Output jumps to 10 works, the highest level before 2026, opening a sustained active stretch.
- 202610 works so far this year, including all four Evolve-reviewed arXiv preprints: PPIM bioheat simulation, synthetic high-frequency ultrasound, glioma proxy forecasting, and GRF gait classification.
Papers in the Evolve catalog
Physics-Guided Synthetic High-Frequency Ultrasound Generation for Skin Layer Segmentation
3D Digital Twin Visualization of Multiclass GRF-Based Gait Disorder Classification
Observation-Anchored Selective Assimilation for Longitudinal Tumor-State Proxy Forecasting in Post-Treatment Glioma
PPIM: Pennes Physics-Informed Mamba for Heat-Source-Conditioned 3D Bioheat Simulation
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 figures come from OpenAlex only; citation-based output and impact dimensions lack sufficient data and are not scored.
- All four catalog papers are 2026 arXiv preprints; peer-review status and final venues are unknown.
- This profile is name-resolved from public bibliographic data (ORCID 0000-0002-0093-2866); no biographical or award information is included.
- Early-stage citation record: h-index 9 with a 2008–2017 gap in yearly counts limits trend interpretation.