輸入您購票時的 Email:

胡翔崴 - Advancing Medical Innovation with Multimodal Edge Computing Technologies - 2024 Taiwan AI Academy Conf

VIP / Speakers

« Back to List

Flash Talk Advancing Medical Innovation with Multimodal Edge Computing Technologies

Time / Place:

⏱️ 09/28 (Sat.) 14:00-14:30 at R3 - 1st Conference Room

Abstract:

This presentation will focus on the principles of multimodal language model technology and its applications in the healthcare field, demonstrating its extensive use through six different medical scenarios. Each scenario will be accompanied by a 30-second demo video, designed to give the audience a quick and immersive experience of these technologies' powerful capabilities in healthcare.

We will begin by showcasing a chronic disease management Q&A system for dementia knowledge. This system is presented in a website format and provides accurate, real-time medical information through a language model, helping users better understand disease-related information. Following this, we will introduce a Line Chatbot for remote diabetes care, which facilitates intelligent Q&A between patients and healthcare providers, offering immediate medical support to diabetes patients.

In the area of emotional companionship, we will share a TVBS interview video demonstrating how an emotional companion QA robot uses language models for question-answering, offering psychological support and enhancing patients' emotional well-being. Additionally, we will demonstrate a digital interaction display for standardized dental patient interactions, where AI automates assessments to enhance the interactivity and effectiveness of dental education.

For image processing applications, we will introduce an X-ray dental image report generation system that assists doctors in automatically generating dental imaging reports, significantly improving medical efficiency. Lastly, we will present a text-to-dental 3D image generation technology that converts text input into 3D images, applied in the development of VR imaging for surgical planning, showcasing cutting-edge technologies in dental surgery.

Through these demo videos, we will guide the audience to a deeper understanding of how multimodal language model technology plays a key role in various healthcare scenarios, highlighting its significance in the future development of medical technology.

😊 Share this page to friends:

Biography:

胡翔崴
  • 胡翔崴 Hsiang-Wei Hu
    Website: https://www.linkedin.com/in/hsiang-wei-hu-19b23b74/
  • Industrial Technology Research Institute / Information and Communications Research Laboratories(ICL), Computational Intelligent Technology Division, Engineer
    • 10 years of AI research and commercialization experience, specializing in time series forecasting, multimodal language models, 2D/3D image generation and segmentation, and knowledge graphs.
    • Vice Chair and Supervisor of the Continuing Education and Exchange Committee, Taiwan AI Academy.
    • Mentor at NCKU Phoenix Startup Platform.
    • Lecturer on Generative AI at ITRI College of Industry.
    • Recipient of the AI Award for Best Speaker from the Taiwan AI Academy.
    • Co-founder of the International Institute for Innovative Biomedical Technology.
    • Lecturer at New York University, Taipei Medical University, and Mackay Medical College.

    胡翔崴博士為 H.I.T. for Asia、翔安生醫共同創辦人及成大鳳凰新創導師,致力於 AI 技術研究 10 年,曾擔任多間新創技術顧問,擁有多項發明專利、發表超過 20 篇國際研討會與期刊論文。 現任職於工研院資通所,專注生成式 AI 技術研發,文字、語音、3D 影像、及化合物生成等多模態大型語言模型開發,主要應用於醫療及金融領域,目前已有十多項產品專案與多家機構合作,研發成果協助工研院資通所智慧運算組榮獲台灣人工智慧協會 2024 AI Award Best Solutions 特優賞,個人演講能力獲得 Best Speaker 優秀賞。 胡翔崴博士以 AI 技術落地應用共創產業價值為願景,所領導的 HIT 社群以共學模式每年培育 100 位生成式 AI 人才為目標,期許能為台灣各領域提升國際競爭力。

😊 Share this page to friends:

😊 Share this page to friends: