人工智慧在醫學影像的分析與應用 (Spring, 2022)
We address the most pressing human needs by solving real-world medical problems with the latest AI and computational technologies.
佈告欄
近期訊息
[2022/04/28] 新增 4/27 課程參考資料:「醫療器材軟體分類與試驗計畫書」投影片 [XYZ01] 與 「試驗計畫書統計分析範例」[XYZ01]
[2022/04/26] 新增 4/20 課程影片,請見NTU Cool
[2022/04/26] 4/27 的課程維持線上的方式,上課連結於課程 Line 群組公佈
[2022/04/19] 考量目前疫情狀況,4/20 的課程將採取線上的方式進行,上課連結於課程 Line 群組公佈
[2022/04/14] 新增 Radiomics 參考資料,請見第九週進度
[2022/04/12] 新增 MONAI Segmentation [20:39]
[2022/04/07] 新增 Nvidia 提供的 MONAI Label [19:36], MONAI Core [01:24:14], MONAI Deploy [18:44] 與投影片
[2022/04/06] 新增 黃宗祺博士投影片 以及 Papers with Code 連結 (參見線上資源)
[2022/04/03] 新增人體解剖學影片、Segmentation、Imaging Analytics、Medicine background、Summer School 等參考資料
[2022/03/29] 新增 Writing Medical AI Papers 參考資料
[2022/03/05] 有興趣嘗試量子機器學習 (Quantum Machine Learning) 的同學,學期計畫可以利用 Pennylane 模擬量子電腦計算
[2022/03/03] 請於 3/7 星期一 23:59 前填寫「問卷_03_學期計畫初步意向_2022_03_02」,可以參考學期計畫選題指南
[2022/02/28] MICCAI Industry Zoom Webinar: "Modern Medical Image Segmentation, AutoML and Beyond" [404] by Dr. Dong Yang (Nvidia), 2022/3/9 10:00 pm
[2022/02/27] 新增「學術活動」欄位
[2022/02/25] 3D Slicer 與 Segmentation 簡介影片上線 [22:10]
[2022/02/24] 請填寫 問卷_02_2022_02_24 [XYZ02]
[2022/02/23] Python 簡介 [48:40] 上線。歡迎使用線上 QA,也記得填寫,「上課參與」次數統計表
[2022/02/16] 請於 2/17 11:59 pm 前填寫 Students self-introduction form 以及 問卷_01_2022_02_16
[2022/02/16] 教室更動為天文數學館 202 室 (lecture)、301 室 (lab)
[2022/02/06] Why AI in medical imaging? Maybe you want to take a look at the talks "Accelerating AI Innovation in Medical Imaging", "How AI Can Augment Radiology" and "為醫療照護和生命科學領域打造人工智慧運算平台" [404]
[2022/02/02] 修課注意事項:(1) 請在 2:17 pm 前入座,我們 2:20 pm 準時開始上課,5:10 pm 準時下課;(2) 上課請攜帶筆電
快速連結
每週進度
12. Clinical Trial (5/4)
Lecture
楊嵐燕博士 (長庚醫院臨床試驗中心統計組組長)
臨床試驗計畫撰寫工作坊
Lab / Homework
Project
Project teamwork
Reference
11. Clinical Trial (4/27)
Lecture
楊嵐燕博士 (長庚醫院臨床試驗中心統計組組長)
軟體醫材臨床評估與範例
Lab / Homework
10. Clinical Trial (4/20)
Lecture
楊嵐燕博士 (長庚醫院臨床試驗中心統計組組長)
臨床試驗基本原則
Lab / Homework
Project
Project teamwork
Reference
9. Radiomics (4/13)
Lab / Homework
Read the papers Lambin et al., 2017 and Gillies et al., 2016.
Work on the Hands-on session in 機器學習與影像組學 [404]
Browse PyRadiomics documentation
Project
Project teamwork
Reference
機器學習與影像組學 [404]
Lambin, Philippe, et al. "Radiomics: the bridge between medical imaging and personalized medicine." Nature Reviews Clinical Oncology 14.12 (2017): 749-762.
Gillies, Robert J., Paul E. Kinahan, and Hedvig Hricak. "Radiomics: images are more than pictures, they are data." Radiology 278.2 (2016): 563-577.
PyRadiomics: Van Griethuysen, Joost JM, et al. "Computational radiomics system to decode the radiographic phenotype." Cancer Research 77.21 (2017): e104-e107.
8. Tools and Workflows (4/6)
Lecture
Tools and Workflows with GPU Accelerated for Medical AI, (Nvidia Solution Architecture Team)
Lab
N/A
Project
Project teamwork
Proposal Working Sheet (version 1)
Homework
C13: Using segmentation to find suspected nodules
Reference
Medical image segmentation using deep learning: A survey (2022)
Writing AI Medical Papers (2021) [55:00]
Datasheets for dataset (2021)
Project
Term project topics and group members
Homework
C11: Training a classification model to detect suspected tumors
C12: Improving training with metrics and augmentation
Homework: Week 06 [Slide]
3. Showcases (3/2)
Lecture
Medical Image Data Analytics and Beyond (Part 2)
Medical AI Projects
Lab
Q&A for DLPyTorch C4-C5
2. A Simple Model for Data Driven Discovery (2/23)
Lecture
Lab
Q&A for DLPyTorch C1-C3
Reference
Google Colab 使用教學 [18:47]
Python 簡介 [30:07]
MeDA Framework
I. Project PDCA
Writing Paper
Checklists
MI-CLAIM: Norgeot, Beau, et al. "Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist." Nature medicine 26.9 (2020): 1320-1324.
CLAIM: Mongan, John, Linda Moy, and Charles E. Kahn Jr. "Checklist for artificial intelligence in medical imaging (CLAIM): a guide for authors and reviewers." Radiology: Artificial Intelligence 2.2 (2020): e200029.
Guidelines
England, Joseph R., and Phillip M. Cheng. "Artificial intelligence for medical image analysis: a guide for authors and reviewers." American journal of roentgenology 212.3 (2019): 513-519. [link]
Marques, Oge, and Christian Garbin. "Assessing Methods and Tools to Improve Reporting, Increase Transparency, and Reduce Failures in Machine Learning Applications in Health Care." Radiology: Artificial Intelligence (2022): e210127. [link].
II. Multimodality Data
Data
Guideline
Datasheets for Datasets: Gebru, Timnit, et al. "Datasheets for datasets." Communications of the ACM 64.12 (2021): 86-92.
TBA
TBA
III. Intelligent Analytics
Segmentation
Segmentation Models
Survey: Wang, Risheng, et al. "Medical image segmentation using deep learning: A survey." IET Image Processing (2022).
nnU-Net: Isensee, Fabian, et al. "nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation." Nature methods 18.2 (2021): 203-211.
UNETR: Hatamizadeh, Ali, et al. "UNETR: Transformers for 3D medical image segmentation." Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2022.
DiNTS: He, Yufan, et al. "DiNTS: Differentiable neural network topology search for 3d medical image segmentation." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021.
Labelling Tools
MONAI Label: Diaz-Pinto, Andres, et al. "MONAI Label: A framework for AI-assisted Interactive Labeling of 3D Medical Images." arXiv preprint arXiv:2203.12362 (2022).
Classification
Classification Models
TBA
Detection
Detection Models
nnDetection: Baumgartner, Michael, et al. "nnDetection: A self-configuring method for medical object detection." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2021.
Imaging Analytics
Imaging
醫學影像成像原理 (台灣大學・周呈霙教授) [404]
Medicine
Medicine Department
Intensive Care Unit
AI in Hemodynamics (台大醫院・葉育彰醫師) [35:02]
AI in Critical Care (台大醫院・葉育彰醫師) [48:53]
Cardiology
Overview of Cardiovascular System Pathophysiology and Monitoring Instruments (新竹台大醫院・廖敏村醫師) [01:03:13]
Cardiovascular | EKG Basics [52:28]
Medical Imaging
Fundamental of Body Imaging: Pancreas (台大醫院・陳柏廷醫師) [56:09]
Scrollable computed tomography images
Brain Case 1: No intravenous contrast.
Brain Case 2: contrast CT, axial plane only
Anatomy
Anatomy (Sam Webster)
Thorax organs - plastic anatomy [21:00]
Thorax transverse CT imaging [27:02]
Lungs (anatomy) [18:00]
Lung shapes (anatomy) [10:09]
Airway anatomy [18:58]
Diaphragm (anatomy) [24:26]
Heart
Heart (anatomy) [26:43]
Pericardium (anatomy) [21:43]
Heart electrics anatomy [15:37]
Anatomy (Sam Webster)
Cardiovascular
Breast
Breast anatomy [19:36]
Pancreas (anatomy) [25:41]
Hepatic portal vein (anatomy) [16:20]
Adrenal glands [19:21]
Bony abdomen landmarks (anatomy) [14:02]
Pelvis anatomy
IV. Medical Workflow
TBA
TBA
TBA
TBA
TBA
V. Regulation & Ethic
TBA
TBA
TBA
TBA
TBA
VI. Solution Landing
TBA
TBA
TBA
TBA
TBA
課程資訊
學期計畫:選題參考
PAN (PANCREASaver) 張大衛、陳彥嘉、葉彥辰
pancreatic cancer segmentation
pancreatic cancer detection and classification
GYN (Gynecological Oncology Screening) 張大衛
body part index
ovarian cancer detection
PCA (Precision Cancer Advisor) 蔡欣翰、張君豪
smart annotation
prognosis prediction
OXR (Omni X-Ray) 黃田宇、張瑋姍、林子敬、陳冠儒
catheter misplacing
pneumothorax detection
tuberculosis detection
MOS (Medical Object Segmentation) 王柏川
Organ/Tumor segmentation
Multi-view multi-label segmentation
CVS (Cardiovascular Suite) 吳耀緯、胡彥川、呂明修
ECG AF detection
ECG LVEF detection
coronary artery calcium score estimation
ICU (Intensive Care Unit Watcher) 郭唯崢、郭庭嘉、吳承彥
acute kidney injury prediction
acute respiratory distress syndrome prediction
sepsis prediction
DBT (Digital Breast Twin) 陳彥嘉
whole slide image analysis
MicroC (Microcirculation video analysis) 呂明修
Medical AI Analytics
X-ray multi-view registration
Data-Centric AI models or tools
Medical AI Technologies
Compressed deep learning model (based on Core ML)
Healthcare data extraction and data cleaning 呂明修
Federated learning 王柏川
Medical MLOps
Quantum machine learning via Pennylane
Medical AI Solutions
Medical AI system development and integration
IP
Business model
Healthcare/medical hackathon or online competition
Online course (e.g. RSNA Magician's Corner, MeDA School)
線上資源
Medical AI Datasets
Algorithms and Codes
時間地點
110學年度第二學期・星期三・第七八九節 (2:20-5:20 pm)
國立台灣大學・天文數學館 202 室 (lecture)、301 室 (lab)
授課教師
王偉仲教授
國立台灣大學・應用數學科學所・數學系・資料科學學位學程
助教
呂明修,r08946008@ntu.edu.tw [XYZ02]
胡彥川,oiyenchuan@ntu.edu.tw [XYZ02]
課程概述
人工智慧在醫學影像的分析與應用 (AI for Medical Image Analysis):「智慧醫療」的實現,不僅得以提升醫療品質,更能使醫療資源的使用更有效率,因而獲得各界引頸期盼,全球對智慧醫療產業的關注和投資,近年來也大幅提升。對於臺灣而言,智慧醫療無疑是重要的利基產業,行政院也將生技醫療視為國家六大戰略產業之一。近期臺灣於疫情期間的表現,備受國際肯定及讚譽,更彰顯出智慧醫療產業之國際戰略地位。本課程針對醫學影像,以醫學資料分析框架 (Medical Data Analytics Framework) 為整體架構,逐步介紹智慧醫療背景知識與解決方案。學習光譜多元且豐富,除了醫學知識、數學、統計、人工智慧、軟體工程等各式科學領域外,更涵蓋醫學倫理、政策法規、商業模式、技術落地等議題。透過宏觀視角,解析智慧醫療產業的發展脈動。智慧醫療的實踐,必須仰賴跨域及跨界的通力合作。課程期間將邀集各領域的專業講師及醫師,協助修課學生了解各種新興人工智慧技術,如何應用在臨床場域,以及衍伸的相關議題。本課程將藉由團隊合作的模式,進行專案實作,以激發跨領域學習與創新。課程始於界定重要臨床問題與任務,同時考量臨床醫事人員的工作流程與準則,而後設計完整且具實務價值之解決方案。在修習本門課程後,修課學生將更具備全球化思維及全面性視野。此外,也能強化醫學影像及數值資料的分析能力,以及智慧醫療解決方案的設計思維。未來在面對智慧醫療的議題時,能夠善用所學的智慧醫療解決方案開發生命週期,著手實作以發揮正向影響力。
課程目標
學習醫學影像分析與應用相關知識與理論
培養設計與實作醫學影像與智慧醫療解決方案能力
促進學生智慧與精準健康全面視野
先修課程
若具備基本 Python、微積分、線性代數、機率與統計、機器學習、深度學習等知識與能力,對學習本課程會有幫助。若無相關背景,可於開學前或學期中自學。
課程內容
縮寫說明
MeDA School: http://medaschool.ai
DL with PyTorch: "Deep Learning with PyTorch" by Stevens, Antiga, Viehmann. ISBN: 978-1617295263 (2020). 中文版:PyTorch深度學習攻略, ISBN:9789863126737 (2021)
Lab: 上課實作
HW: 回家作業
FC: Flipped Classroom 翻轉教室
評量方式
暫定內容,將視修課同學背景與興趣微調
小組計劃進度報告:10%
小組計劃期末成果 (軟體、海報、一分鐘影片) :個人貢獻:15%,整合成效:15%
個人期末書面報告:20%
上課參與 (提問、回答問題、發表意見、協助同學):10%
作業 (書面與口頭報告):30%
參考書目
教師提供講義、論文與參考資料
重要期刊
線上資料庫
書籍
"Deep Learning with PyTorch" by Eli Stevens, Luca Antiga, Thomas Viehmann. ISBN: 978-1617295263 (2020). 中文版:PyTorch深度學習攻略, ISBN:9789863126737 (2021).
學術活動
Summer School
International Summer School on Neuroimaging, NeuroScience, Neuroncology. Lipari, Italy. 2022/7/4-9 [XYZ02]
UCL Medical Image Computing Summer School, UCL, 2022/7/11-15 (virtual)
Summer school on deep learning for medical imaging, Montréal, 2022/7/4-9
Surgical Data Science Summer School, Strasbourg, France, 2022/7/18-22
Annual Multicell Virtual-Tissue Modeling Online Summer School and Hackathon, Indiana University, USA, 2022/7/31-8/7
Project
Project teamwork
Reference
「醫療器材軟體分類與試驗計畫書」投影片 [XYZ01]
試驗計畫書統計分析範例 [XYZ01]