UM PhD student Wins Runner-up in Student Paper Award at International Conference on Nuclear Science and Medical Imaging
澳大博士生榮獲國際核科學研討會與醫學影像會議學生論文獎亞軍
劉怡斌(中)與IEEE NSS/MIC 2025會議共同主席韓國首爾大學李在成教授(左)及
日本量子科學技術研究所山谷泰賀教授(右)合影留念
Yibin Liu (center) with Co-chairs of the 2025 IEEE NSS/MIC, i.e., Prof. Jae Sung Lee (left) from Seoul National University, Korea and
Prof. Taiga Yamaya (right) from the National Institutes for Quantum Science and Technology, Japan
劉怡斌(左)與指導教授莫昇萍於獲獎海報前
Yibin Liu (left) and supervisor Prof. Greta Mok in front of the award-winning poster
澳大劉怡斌博士生榮獲 2025 年 IEEE NPSS Christopher J. Thompson 學生論文獎亞軍
Yibin Liu received the Runner-up Award at the 2025 IEEE NPSS Christopher J. Thompson Student Awards
UM PhD student Wins Runner-up in Student Paper Award at International Conference on Nuclear Science and Medical Imaging
The Biomedical Imaging Laboratory (BIG) of the Faculty of Science and Technology (FST), University of Macau (UM), achieved outstanding results at the 2025 IEEE Nuclear Science Symposium and Medical Imaging Conference (IEEE NSS/MIC) held in Yokohama, Japan. The team had five papers accepted by the conference. Among them, PhD student Yibin Liu was selected as the Runner-up for the IEEE NPSS Christopher J. Thompson Student Paper Awards with his work, entitled “A Preliminary Study on a Two-Step Deep Learning Approach for Voxel-based Lu-177-PSMA Absorbed Dose Prediction from Pre-therapeutic F-18-PSMA PET.”
The award-winning research was led by Prof. Greta Mok from the Department of Electrical and Computer Engineering of FST, with clinical data provided by Inselspital, University Hospital of Bern, Switzerland. The novel Lu-177-PSMA targeted radioligand therapy has shown remarkable efficacy in patients with advanced prostate cancer, offering new hope for the treatment of metastatic disease. However, current clinical practice still predominantly adopts fixed-dose regimens, neglecting inter-patient pharmacokinetic variability, which limits the development of personalized therapy. Optimization of treatment efficacy and safety is desirable.
To address this clinical problem, the research team proposed an innovative Two-Step Deep Learning framework for voxel-based absorbed dose prediction, which leverages pre-therapeutic F-18-PSMA PET/CT images and biomarker information to accurately predict the voxel-wise absorbed dose distribution in Lu-177-PSMA therapy. This method significantly improves the prediction accuracy for both tumors and organs, shielding new lights towards personalized radionuclide therapy for prostate cancer.
The IEEE NSS/MIC Annual Conference is one of the most influential international conferences in the fields of nuclear science and medical imaging. With a 72-year history, it is organized by the IEEE Nuclear and Plasma Sciences Society (NPSS) and attracts over a thousand experts and scholars from around the world each year. The UM research team stood out from nearly 700 submissions in the Medical Imaging category, demonstrating its international competitiveness and scientific strength in radionuclide therapy dosimetry and artificial intelligence applications in nuclear medicine. This research was supported by the National Natural Science Foundation of China (Grant No. 82572278).
澳大博士生榮獲國際核科學研討會與醫學影像會議學生論文獎亞軍
澳門大學科技學院生物醫學影像實驗室團隊於日本橫濱舉行的2025年IEEE核科學研討會與醫學影像會議（IEEE NSS/MIC）年會上取得優異成績。團隊共有五篇論文獲大會接收，其中博士生劉怡斌憑藉其研究《A Preliminary Study on a Two-Step Deep Learning Approach for Voxel-based Lu-177-PSMA Absorbed Dose Prediction from Pre-therapeutic F-18-PSMA PET》，榮獲本屆會議的IEEE NPSS Christopher J. Thompson學生論文獎亞軍。
該獲獎研究由科技學院電機及電腦工程系莫昇萍教授指導，並獲瑞士伯爾尼大學附屬醫院（Inselspital）提供數據支持。新型Lu-177-PSMA 靶向放射性配體治療已在晚期前列腺癌中展示出顯著療效，被視為是治療轉移性癌症的新希望。但目前臨床普遍採用的固定劑量方案未能充分考慮個體間的藥代動力學差異，限制了個體化治療的發展，療效和安全性有待優化。針對這一挑戰，研究團隊創新提出基於雙步深度學習（Two-Step Deep Learning）的體素級吸收劑量預測框架，透過患者治療前的 F-18-PSMA PET/CT 影像和生物標誌物資訊，有效預測 Lu-177-PSMA 治療的體素級吸收劑量分佈。該方法顯著提升了腫瘤與臟器的吸收劑量預測精度，為推進前列腺癌的個體化核素治療的臨床實現提供了新的思路與方法。
IEEE NSS/MIC年會作為核科學與醫學影像領域最具影響力的國際學術盛會之一，迄今已有 72 年歷史，由IEEE核與等離子體科學學會（NPSS）主辦，每年吸引全球上千名專家學者參與。在醫學影像類別近700篇投稿論文中，澳大團隊能夠脫穎而出，充分展示了其在放射性核素治療劑量學與醫學影像人工智慧領域的國際競爭力與科研實力。該項研究由國家自然科學基金面上項目（82572278）資助。