Zheng Rong JIA
It’s never too late to begin again.
MSc Student
Young Ambassador & Researcher
Research Focus
Clinical Prediction · Trustworthy Clinical AI · Reproducible Pipelines
Research Interests
Clinical Prediction
Deep learning on electronic health records for ICU risk stratification and mortality prediction
Trustworthy Clinical AI
Inference-time safeguards against physiological outliers, and attention-based interpretability a clinician can check
Reproducible Pipelines
Multicenter clinical databases, leakage-free cohort pipelines, and open code
Publications
DualTower-FT with an Adaptive Runtime Safeguard: A Deep Tabular Approach for ICU Stroke Mortality
The 23rd Pacific Rim International Conference on Artificial Intelligence · Guangzhou, November 2026
Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework
2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI) · Nanjing, May 2026
GitHub Activity
News
- Aug 19, 2026 Short paper accepted at PRICAI 2026 — DualTower-FT with an Adaptive Runtime Safeguard: A Deep Tabular Approach for ICU Stroke Mortality. Details.
- Jul 15, 2026 Reviewed four submissions for PRICAI 2026 and was invited to serve on its Program Committee.
- May 22–24, 2026 Received Best Industrial Paper Award & Best Presentation Award at CCAI 2026 — Nanjing.
- May 22–24, 2026 Attended CCAI 2026 in Nanjing — delivered an oral presentation of Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework. Slides and photos now available.
- May 2026 Released the DT-Transformer framework open-source for reproducible clinical AI research — training pipeline and data preprocessing pipeline.
- Feb 2, 2026 Paper accepted at CCAI 2026 — Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework.
- Aug 2025 Earned B.Sc. in Software Engineering from Macau University of Science and Technology.
About
When machine learning meets real clinical stakes, a model's failure is not an accuracy number but a missed diagnosis. Those are the problems I am drawn to. My work focuses on building deep learning systems for electronic health records that are not only accurate, but interpretable and honest about their limits. I believe the most durable research is reproducible, open, and built with the long game in mind.
Z. Jia and K.-C. Wong, "DualTower-FT with an Adaptive Runtime Safeguard: A Deep Tabular Approach for ICU Stroke Mortality," in Proc. 23rd Pacific Rim Int. Conf. Artif. Intell. (PRICAI 2026), Guangzhou, China, Nov. 2026, to be published.
@INPROCEEDINGS{jia2026dualtowerft,
author={Jia, Zhengrong and Wong, Kwong-Cheong},
booktitle={Proceedings of the 23rd Pacific Rim International Conference on Artificial Intelligence (PRICAI 2026)},
title={{DualTower-FT} with an Adaptive Runtime Safeguard: A Deep Tabular Approach for {ICU} Stroke Mortality},
series={Lecture Notes in Artificial Intelligence},
publisher={Springer},
year={2026},
note={Accepted, to be published}}
TY - CONF TI - DualTower-FT with an Adaptive Runtime Safeguard: A Deep Tabular Approach for ICU Stroke Mortality T2 - Proceedings of the 23rd Pacific Rim International Conference on Artificial Intelligence (PRICAI 2026) T3 - Lecture Notes in Artificial Intelligence AU - Jia, Zhengrong AU - Wong, Kwong-Cheong PB - Springer PY - 2026 N1 - Accepted, to be published ER -
Z. Jia and K.-C. Wong, "Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework," 2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI), Nanjing, China, 2026, pp. 76-81, doi: 10.1109/CCAI69603.2026.11641950.
@INPROCEEDINGS{11641950,
author={Jia, Zhengrong and Wong, Kwong-Cheong},
booktitle={2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI)},
title={Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework},
year={2026},
volume={},
number={},
pages={76-81},
doi={10.1109/CCAI69603.2026.11641950}}
TY - CONF TI - Deep Learning for Stroke Mortality Prediction in eICU: A Dual-Tower Transformer Framework T2 - 2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI) SP - 76 EP - 81 AU - Z. Jia AU - K. -C. Wong PY - 2026 DO - 10.1109/CCAI69603.2026.11641950 JO - 2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI) Y1 - 22-24 May 2026 ER -
Cite This Paper