Thomas Haugland Johansen

Title of the thesis:

«Leveraging Computer Vision for Applications in Biomedicine and Geoscience»

Popular sientific abstract:

Skin cancer is the most common type of cancer and is usually classified as either non-melanoma or melanoma skin cancer. Melanoma skin cancer accounts for about half of all skin cancer deaths. The 5-year survival rate is 99% when the cancer is detected early but only 25% once it spreads to distant organs — early detection is vital to saving lives.

Foraminifera are microscopic single-celled organisms that exist in marine environments. Roughly 50 000 species are known to have existed, and 9 000 are still living today. They are important for reconstructing past climate conditions and for detecting pollution. The identification and counting of foraminifera are performed manually and require a lot of time and expertise.

In this dissertation, we leverage recent advances in computer vision, driven by breakthroughs in artificial intelligence and advanced image analysis, to make progress towards early detection of melanoma skin cancer and automating the identification and counting of foraminifera.

The thesis is published and available in Munin

.Supervisors:

• Professor Fred Godtliebsen, Department of Mathematics and Statistics, UiT (main supervisor)
• Associate Professor Kajsa Møllersen, Department of Community Medicine, UiT

Evlauation Committee:

• Associate Professor Edgar Lobaton, Electrical and Computer Engineering, NC State University, US (1. Opponent)
• Maryam Tayefi, Norwegian Center for E-health Research (2. Opponent)
• Professor Robert Jenssen, Department of Physics and Technology, UiT (internal member and leader of the commitee)

Leader of the public defence:

The leader of the public defence is Associate Professor Annette Bayer, Head of Department of Chemistry, Faculty of Science and Technology.

Opposition ex auditorio:

If you have any qiestions for the candidate during the public defence, please send an e-mail to annette.bayer@uit.no. She will announce the questions during the defence.

Trial lecture:

The trial lecture is held on Friday June 25th at 10.15 in the same auditorium. The title of the trial lecture is: «Deep Generative Models»

Streaming:

The defence and trial lecture will be streamed via Mediasite:

Audience:

UiT follows the national guidelines regarding infection control. 

Audience:

UiT follows the national guidelines regarding infection control. A maximum of 31 people are allowed in the auditorium during the defence, as long as everybody keeps a distance of 1 meter at all times.