智慧醫療數據模式識別

The intent of this course aims to effectively address the pattern recognition problems of smart healthcare data by developing high performance machine learning models. In the 21th century, the medical data are generated by a variety of elaborated medical diagnostic machines, and these medical data are usually complicated with the uncertainty. Through addressing the pattern recognition problem of medical data, the pattern or hidden information can be positively discovered from medical raw data, and it can further bring a good diversity of contribution to smart healthcare, such as key symptom identification and clinical decision support. Through giving a brief concept of machine learning modeling, this also provides real works to accumulate the practical experience of addressing the pattern recognition problem of smart healthcare data. Upon completion of this course, students will have the rudimentary knowledge of machine learning and the corresponding skills to address the pattern recognition problem of smart healthcare data.

(A) Machine Learning (B) Discrete Mathematics (C) Programming

Pedagogy and other supplementary information (websites, TAs, handouts and/or databases): Part of lectures’ contents is organized through selecting the appropriate teaching materials, technical reports, journal papers and the relevant latest technological news. The corresponding exercises and projects are practically conducted by the medical data of real world.

1.Homework and Assignments: Biweekly Assignments 2. Exams and Quizzes: N/A 3. Evaluation and Grading Policy: Q & A in the class (20%), Assignments (40%) and Final Group Project (40%)

• Smart Healthcare
• Machine Learning

Pattern Recognition and Machine Learning (7th Edition) Author:Christopher M. Bishop Publisher:Springer Year of Publication:2011