Advanced Study on ArtificialIntelligence andInformatics Ⅱ

Advanced Study on ArtificialIntelligence andInformatics Ⅱ

-

Department of Electrical Engineering
Department of Electronics and Computer Science

Graduate School of Engineering

1Year

HETDA7MCA1

2.00Credits

講義 (Lecture)

Fall semester 2026

(Fall semester)

Syoji Kobashi,Takayuki Wada,Manabu Nii,Kouki Nagamune

Graduate School of Engineering

Japanese

Japanese or English, depending on students’ preferences.

3／9

Himeji Engineering Campus, Building 6, Room 6313 (Prof. Kobashi’s office)
Thursdays, 12:10–13:00
Advance reservation via the UNIPA Q&A system is recommended.

Please contact faculty members via the UNIPA Q&A system.

ー

1◎／2◎／3〇

N/a

ー

[Lecture Objectives]
This course aims to cultivate advanced expertise and research skills required of graduate-level researchers through the close reading, discussion, and presentation of cutting-edge studies in medical image engineering and medical artificial intelligence. Emphasis is placed on medical image analysis, deep learning, clinical data utilization, medical–engineering collaboration, and pathways to social implementation, enabling students to appropriately position their own research topics from both academic and societal perspectives.

[Learning Outcomes]
Upon successful completion of this course, students will be able to:

1. Organize and explain major research trends in medical image engineering and medical artificial intelligence.
2. Critically analyze research papers and logically summarize their originality, effectiveness, and limitations.
3. Explain research design principles and evaluation methodologies using clinical data.
4. Clearly articulate the background, objectives, methods, and significance of their own research and present them orally.

Course Subtitle:
Advanced Research Seminar on Medical Imaging and Artificial Intelligence

Keywords:
Medical Image Engineering, Medical Artificial Intelligence, Deep Learning, CT and MRI Image Analysis, Explainable AI, Clinical Data Analysis, Medical–Engineering Collaboration, Software as a Medical Device (SaMD)

This course covers advanced topics in medical image engineering and medical artificial intelligence through paper reading, student presentations, and discussions. The course does not focus on a specific organ or disease, but addresses medical imaging research in a broad and comprehensive manner, from research design to social implementation.

Week 1 Guidance
The objectives, structure, and evaluation methods of the course are explained, together with an overview of medical imaging and medical AI research.

Week 2 Fundamentals and Trends in Medical Image Engineering
Basic concepts and recent trends in CT, MRI, and X-ray image analysis are reviewed.

Week 3 Medical AI and Deep Learning
Fundamental architectures and challenges of deep learning for medical image analysis are discussed.

Week 4 Research Design for Medical Imaging AI
Research design, including data collection, preprocessing, training, and evaluation metrics, is examined through representative studies.

Week 5 Explainable AI in Medical Applications
The importance of explainability in medical AI is discussed using visualization methods such as Grad-CAM.

Week 6 Case Studies in Medical Image Analysis
Representative studies on medical image analysis across various organs and modalities are reviewed.

Week 7 Quantitative Analysis and Clinical Applications of Medical Images
Quantitative evaluation of medical images and applications to diagnosis and decision support are discussed.

Week 8 Medical Data, Ethics, and Regulations
Ethical review, data protection, and researchers’ responsibilities in the use of clinical data are covered.

Week 9 Medical–Engineering Collaboration
Collaborative research frameworks involving universities, hospitals, and industry are examined.

Week 10 Social Implementation of Medical AI and SaMD
Concepts and evaluation strategies for medical AI as software as a medical device (SaMD) are introduced.

Week 11 Critical Reading of Research Papers (Student Presentations)
Students present selected papers related to medical imaging and medical AI and discuss their originality and limitations.

Week 12 Development and Refinement of Research Topics
Students organize and refine their own research topics through discussion.

Week 13 Dissemination of Research Outcomes
Strategies for publishing papers, conference presentations, and patents are discussed.

Week 14 Student Research Presentations
Students present their research plans or progress for comprehensive discussion.

Week 15 Summary and Review
The course content is reviewed, and essential perspectives for researchers in medical imaging and medical AI are summarized.

Remote (Fully Online)

This course will be conducted as a fully online remote class.

Limited permission for use

In this course, the use of generative AI is permitted as an effective tool during the process of research and exploration. However, students must verify the accuracy and validity of any AI-generated content by consulting primary sources, such as original research articles or official documents.

When using generative AI for reports or presentation materials, direct copying of AI-generated output is not permitted. Students are required to understand the content thoroughly, organize it independently, and write it in their own words, taking full responsibility for their submissions. If the submitted work contains incorrect statements due to hallucinations or includes information that is not supported by reliable primary sources, the evaluation will be significantly lowered.

No textbook is specified. Lecture materials will be distributed as needed.

Handbook of Medical Image Computing and Computer Assisted Intervention
Nicholas Ayache, Bennett A. Landman, et al. (Eds.)
Handbook of Medical Image Computing and Computer Assisted Intervention
Academic Press (Elsevier), 2020.
ISBN: 978-0-12-816176-0

Preparation:
Before each class, students are expected to review the assigned reference materials and lecture handouts to grasp the outline of the topic. When necessary, students should also read related research papers to prepare for the lecture.
Estimated time: 1–2 hours per class (total 15–30 hours)

Review:
After each class, students should review the lecture content and deepen their understanding using the distributed materials and reference books. This includes preparing reports, presentation materials, and organizing their research topics.
Estimated time: 1–2 hours per class (total 15–30 hours)

This course incorporates student-centered active learning. Students are required to read research papers related to medical image engineering and medical artificial intelligence, present summaries of selected papers, and participate in question-and-answer sessions. In addition, discussions and peer feedback on research topics are conducted to deepen understanding and refine research plans. Through these activities, students develop critical thinking, communication, and discussion skills.

Grades are determined based on a comprehensive evaluation of the following components.

1. Contribution to Presentations and Discussions (40%)
Students are evaluated on their presentations of research paper summaries and their participation in discussions, with emphasis on their level of understanding, the logical clarity and appropriateness of their explanations and comments, and their constructive contributions to discussions.

2. Reports and Assignments (40%)
Reports and assignments are evaluated based on the accuracy of the investigated content, clarity of logical structure, and depth of analysis. When generative AI is used, particular attention is paid to whether the content has been verified using primary sources and whether the work is written based on the student’s own understanding. If the submission contains incorrect statements due to hallucinations or other errors, the evaluation will be significantly lowered.

3. Engagement in Learning Activities (20%)
Students are evaluated on their preparation for classes, the quality of their questions and comments, and their involvement in peer feedback activities, reflecting proactive and sustained engagement in learning.

The results of reports and assignments will be disclosed through UNIPA or an equivalent system. Feedback will be provided during class or individually when necessary.

This course is a graduate-level special seminar and requires a proactive and independent learning attitude. Students are expected to have the ability to read and present academic papers written in English. As the course primarily consists of paper reading, presentations, and discussions, students are required to prepare in advance and participate actively.

This course does not fall under practical education.

This course will be conducted in accordance with the learning objectives and evaluation policies stated in the syllabus, and the content and pace of the course may be flexibly adjusted within this framework to accommodate students’ research topics and levels of understanding.