Continuing the proud tradition that we started last year, we invite you to a half day celebrating Louise Petrén, the first woman to obtain a PhD in mathematics in Sweden. This year's Louise Petrén lecture will be given by Prof. Nina Gantert, who works in the area of probability theory, in particular, stochastic processes, large deviations and random media. She is an exceptional mathematician, an amazing supervisor, mentor and role model. We will also listen to a talk by Dr Ida Arvidsson och Jennie Karlsson. Ida and Jennie won the LU innovation award 2022 in the category employees. Ida is a postdoc and Jennie is a PhD student at the Centre for Mathematical Sciences. In their talk, Ida and Jennie will tell us more about their innovation and something about the mathematics behind it. Between the talks, we will serve fika. Both talks are accessible to students! Everybody is welcome!

Program

14:00 - 15:00 Nina Gantert: Flexibility or decisiveness: who wins?

Abstract: 

We discuss this question on the example of biased random walk on percolation clusters. We also give some results on biased random walk among random conductances. Finally we announce new results on biased random walks on dynamcial percolation.

No previous knowledge is assumed.

Based on joint work with Sebastian Andres, Noam Berger, Xiaoqin Guo, Matthias Meiners, Sebastian Müller, Jan Nagel, Perla Sousi and Dominik Schmid.

15:00 - 15:30 Coffee break

15:30 - 16:15 Ida Arvidsson and Jennie Karlsson: Detection of breast cancer with help of machine learning

Abstract: 

In this talk we are going to present 23°N, a project in collaboration between the Centre for Mathematical Sciences and Skåne University Hospital. The goal of the project is to make breast imaging available for all women.

In low- and middle-income countries the access to breast imaging is often poor, and a lot of women dies in breast cancer every year. We focus on breast imaging with ultrasound and have collected data from normal ultrasound and point-of-care ultrasound. Point-of-care ultrasound is a very small ultrasound device which can be paired with a smartphone.

In this talk we will describe how detection of breast cancer can be performed with help of convolutional neural networks. We will also talk about how CycleGAN can be used to handle the scarcity of data. The project has achieved promising results and won the LU innovation award 2022.

All are welcome!