MDDI 演講稿 · 2024-09-30
陳杰豪高階政務部長在醫學 AI 中心啟動儀式上的開幕致辭
要點
- • C-AIM（Centre of AI in Medicine）由「全國醫療集團」（NHG）與南洋理工大學（NTU）聯合設立——把醫療與 AI 串起來。
- • AI 已經在改寫新加坡的醫療：AlphaFold 讓人類幾年內預測 6 億+ 蛋白結構（過去半世紀才解了幾十萬）；AI Singapore 的 JARVIS-DHL 用 AI 識別糖尿病、高血壓、高膽固醇風險患者；NUHS 自訓 RUSSELL-GPT 寫病程總結與轉診信，Endeavour AI 平臺預測床位。
- • 「信任」是新加坡的「最大 alpha 源」——醫療行業更不能丟。技術與倫理複雜度（診斷準確性、透明度、訓練集多樣性、敏感資料安全）必須並行處理。
- • C-AIM 的專案示例：PRIME-CXR 快速分診胸片、與 Resaro 合作做「評估 AI 臨床價值」的穩健框架。
- • C-AIM 同日與 NHG、奧林巴斯（Olympus）、耶魯醫學院（Yale School of Medicine）簽署 MOU。
完整譯文（繁體中文）
MDDI 英文原文譯文 · 翻譯日期: 2026-05-03
本文已從早期版本的網站遷移過來——格式可能有不一致之處。
數碼發展及新聞部高階政務部長 Tan Kiat How 在「醫學 AI 中心」（C-AIM）啟動儀式上的開幕致辭（2024 年 9 月 30 日）
新加坡李光前醫學院（LKC Medicine）治理委員會主席 Lim Chuan Poh 先生；
新加坡全國醫療集團（NHG）集團執行長 Joe Sim 教授；
LKC Medicine 院長、醫學 AI 中心聯合主任 Joseph Sung 教授；
南洋理工大學（NTU）計算與資料科學學院高階副院長、醫學 AI 中心聯合主任 Miao Chun Yan 教授；
各位女士、先生、嘉賓：
1. 我很高興出席今天「醫學 AI 中心」（Centre of AI in Medicine, C-AIM）的啟動儀式——它由全國醫療集團（NHG）與南洋理工大學（NTU）共同設立。
2. 該中心的成立——是新加坡「借力 AI 改善公民健康與福祉」旅程上的又一重要里程碑。
3. 新加坡一直擁抱「以技術克服約束」——同時對權衡保持清醒——並主動地縮小下行影響。這種務實姿態——塑造了我們對幾代通用目的技術（從計算機、到網際網路、再到其他數字創新）的使用。
4. 我們對 AI 也採取相同的姿態與方法——AI 近期取得了快速進展——尤其在生成式 AI 領域。這一信念——支撐著我們的《國家 AI 戰略 2.0》。
5. 我們對 AI 討論的近期轉變感到鼓舞——它已超越最初的炒作——走向更冷靜地聚焦「影響」與「投資回報」。對影響的聚焦——增加了我們「借力 AI 解決真實世界問題、為企業與社會帶來差別」的能力。對投資回報的強調——也鼓勵了更可行的商業模式。沒有對這些因素的認真聚焦——很難圍繞 AI 構建可持續且有活力的生態。
6. 因此——我很高興看到——醫療生態中的利益相關方——把心思放在「借力 AI 給我們的社群與患者帶來深遠影響」上。
7. AI 在醫療中的應用前景廣闊——從發現新療法到改善治療效果。比如——我們花了半世紀去理解幾十萬種蛋白的結構。藉助 AlphaFold 等 AI 工具——科學家們在短短幾年裡就預測了 6 億多種蛋白的結構與相互作用——這徹底改變了藥物發現。我們對這一領域的潛力——還只是觸及表層。
8. 在更下游——AI 也在應對「勞動力短缺與醫療成本上升」——這些挑戰在全球老齡化背景下持續累積。新加坡尤其切身感受到這些約束。預測性診斷等能力——幫助醫療專業人士克服這些挑戰、更好地服務患者需求。比如——AI Singapore 的「JARVIS-DHL」專案——用 AI 識別糖尿病、高血壓、高膽固醇風險患者——便於早期干預——延緩疾病進展與併發症。
9. AI 也在改造臨床工作流與患者旅程。新加坡國立大學醫學中心（NUHS）訓練了自家的大語言模型「RUSSELL-GPT」——能在幾秒內總結患者筆記、撰寫轉診信。其「Endeavour AI」平臺——還能預測醫院床位的可用性——最佳化容量與患者安置。這些努力賦能醫生——更快、更易做出資料驅動的決策——讓他們把更多時間放在患者照護上。
10. 當 AI 改造醫療時——「守護信任」至關重要。新加坡的品牌之所以有溢價——是因為我們在所做的一切上獲得的信任——無論國內還是國際。這正是黃循財總理稱為新加坡「最大 alpha 源」的——我們對「信任、誠信、可靠」的聲譽。
11. 同樣——「信任」也是醫療的核心——支撐著從藥物研發到臨床實務的醫療與健康照護的交付。即便我們借力 AI 實現醫療中的承諾——我們必須把這一追求與「守護信任」的努力同步進行。
12. 這意味著應對使用 AI 的「技術與倫理複雜度」——比如確保診斷準確性、透明度——並在多樣化資料集上訓練 AI 系統、確保跨人群的恰當代表性。我們也必須管理敏感醫療資料的安全與保護。
13. 我很高興——醫學 AI 中心正在推進下一波合作——把 AI 研究轉化為「負責任的臨床實務」。展位上展示的專案——突出了 C-AIM 在「構建可臨床落地的方案」上的努力。
14. 比如「PRIME-CXR」專案——能快速準確地為胸片做分診——把異常發現優先排隊——以加快臨床決策。PRIME-CXR 也與初創 Resaro 合作——開發一個穩健框架——評估「哪些 AI 方案帶來最高臨床價值」。
15. 這些方案之所以可能——是因為 C-AIM 中——AI 研究者、工程師與醫療專家形成了關鍵紐帶——並應用「實施科學」（Implementation Science）方法——以最大化 AI 在醫療中的影響。
16. 這種協作努力——以確保「倫理部署」的基本步驟為基礎——從「監督資料訪問」的委員會，到關於「知情 AI 實踐」的培訓與教育，再到與專家及公眾圍繞「風險感知與有效治理」的定期互動。
17. 展望未來——醫療不僅是國家級挑戰，更是全球性挑戰——需要跨學科、跨領域、跨行業、跨邊界地共享知識與專長。學界、產業、公共部門之間的協作——對促成醫療的「整體性轉型」越來越重要。
18. 因此——我很高興見證 C-AIM 與「全國醫療集團」（NHG）、奧林巴斯（Olympus）、耶魯大學醫學院（Yale School of Medicine）簽署的 MOU。
19. 我以感謝所有合作伙伴收尾——感謝在場的研究者、博士生、實務者、社群專家——感謝你們成為這一重要旅程的一部分——共同借力 AI 改造醫療、健康與公共服務的交付。
20. 我期待該中心未來令人激動的專案管線。
21. 非常感謝。
英文原文
MDDI 官網原始記錄 · 抓取日期: 2026-05-02
This article has been migrated from an earlier version of the site and may display formatting inconsistencies.
OPENING ADDRESS BY SENIOR MINISTER OF STATE, MINISTRY OF DIGITAL DEVELOPMENT AND INFORMATION, MR TAN KIAT HOW AT THE LAUNCH OF CENTRE AI IN MEDICINE (C-AIM) ON 30 SEP 2024
Mr Lim Chuan Poh, Chairman of the Governing Board at LKC Medicine;
Professor Joe Sim, Group CEO of the National Healthcare Group;
Professor Joseph Sung, Dean of LKC Medicine and Co-Director of the Centre of AI in Medicine;
Professor Miao Chun Yan, Senior Deputy Dean of the NTU College of Computing and Data Science, and Co-Director of the Centre of AI in Medicine;
Ladies and gentlemen, and distinguished guests.
1. It is my pleasure to join all of you today at the launch of the Centre of AI in Medicine, or ‘C- AIM’ in short. The Centre is jointly established by the National Healthcare Group and Nanyang Technological University.
2. The establishment of this Centre is another significant milestone in Singapore's journey to harness AI to improve the health and well-being of our citizens.
3. Singapore has always embraced the use of technologies to overcome our constraints, while remaining clear-eyed about the trade-offs, and actively working to minimise downsides. This pragmatic attitude has shaped our use of generations of general-purpose technologies, from computers, to the internet, to other digital innovations.
4. We adopt the same attitude and approach to AI, which has seen rapid advancements in recent times – especially in the field of generative AI. This belief underpins our National AI Strategy 2.0.
5. We are heartened by the recent shifts in discussions around AI, which have moved beyond the initial hype, towards a more sober focus on impact and returns on investment. The growing focus on impact has increased our ability to harness AI to solve real-world problems and make a difference to businesses and society. The emphasis on returns on investment has also encouraged more viable business models. Without a serious focus on these factors, it is difficult to build a sustainable and vibrant ecosystem around AI.
6. Hence, I am glad to see how stakeholders in the healthcare ecosystem are applying their minds to harness AI, to deliver profound impact to our community and patients.
7. There is significant promise in the use of AI in healthcare – from discovering new therapeutics to improving treatment efficacy. For example, it took us half a century to understand the structures of a few hundred thousand proteins. With AI tools like AlphaFold, scientists have predicted the structures and interactions of over 600 million proteins in just a few years. This has revolutionised drug discovery. And we are just scratching at the surface of the potential of applying AI in this area.
8. Further downstream, AI is also tackling workforce shortages and rising healthcare costs – challenges that continue to mount with an ageing population across the world. Singapore feels these constraints especially acutely. Capabilities like predictive diagnostics help healthcare professionals overcome these challenges and better serve patient needs. For example, AI Singapore’s JARVIS-DHL project uses AI to identify patients at risk of diabetes, hypertension, and high cholesterol for early intervention, thereby slowing disease progression and complications.
9. AI is transforming clinical workflows and the patient journey, too. NUHS has trained its own large language model, RUSSELL-GPT, to summarise patient notes and write referral letters in just seconds. Its Endeavour AI platform also forecasts bed availability in hospitals, better optimising capacity and patient placement. Such efforts empower doctors to make data-driven decisions with speed and ease – allowing them to spend more time on patient care.
10. As AI transforms healthcare, safeguarding trust is of crucial importance. Singapore’s brand has a premium because of the trust we command in whatever we do – both domestically and internationally. This is after all what Prime Minister Lawrence Wong calls Singapore’s ‘greatest source of alpha’ – our reputation for trust, integrity, and reliability.
11. Likewise, trust is also at the heart of healthcare, underpinning the delivery of health and medical care from drug research and development to clinical practices. Even as we harness AI to realise its promise in healthcare, we must twin this pursuit with efforts to safeguard trust.
12. This entails tackling the technical and ethical complexities with using AI, such as ensuring diagnostic accuracy, transparency, and training AI systems on diverse datasets to ensure proper representation across populations. We must also manage the security and protection of sensitive healthcare data.
13. I am glad that the Centre of AI in Medicine is embarking on the next bound of partnerships to translate AI research into responsible clinical practice. Projects showcased at the booths highlight C-AIM’s efforts in building clinically implementable solutions.
14. The PRIME-CXR project for example rapidly and accurately triages chest x-rays to prioritise abnormal findings for faster clinical decisions. PRIME-CXR is also collaborating with start-up, Resaro, to develop a robust framework that evaluates which AI solutions deliver the highest clinical value.
15. These solutions are possible because of the critical nexus of AI researchers, engineers, and healthcare experts in C-AIM, applying Implementation Science methods to maximise the impact of AI in healthcare.
16. This collaborative effort is underpinned by fundamental steps to ensure ethical deployment– from committees that oversee data access, to training and education on informed AI practices, to regular engagements with experts and public on risk perceptions and effective governance.
17. Looking ahead, healthcare will remain not only a national challenge, but a global one – requiring the sharing of knowledge and expertise not just across disciplines and domains, but across sectors and borders too. Collaborations between academia, industry and the public sector will be increasingly important to enable a holistic transformation of healthcare.
18. I am therefore delighted to witness the signing of MOUs between C-AIM and the National Healthcare Group, Olympus, and the University of Yale School of Medicine.
19. I would like to end by thanking all of our partners, all of you here, researchers, PhD students, practitioners and community experts for being part of this important journey as we use AI to transform healthcare, wellness and delivery of such services to the public.
20. I look forward to the Centre’s exciting pipeline of projects moving forward.
21. Thank you so much.