Lassi Paavolainen's research focuses on uncovering patterns in multiplexed microscopy images of (cancer) tissues and cells. Dr. Paavolainen studies state-of-the-art deep learning methods to uncover these patterns using self-supervised and unsupervised learning strategies. These novel methods and developed models are used in cancer research to improve understanding of cancer tissue and tumor microenvironment architecture needed for developing prognostic biomarkers and precision treatments. The developed deep learning methods are also used for image-based profiling of cancer cells under perturbations where possibilities of generative AI and foundation models are of interest to Dr. Paavolainen.
Lassi Paavolainen defended his doctoral research in 2013 that was done in Varpu Marjomäki's research group at the University of Jyväskylä, Finland, studying microscopy image analysis methods and software, and continued in the group in 2014 as a postdoctoral researcher. In 2015, Dr. Paavolainen started his postdoctoral research at FIMM in Horvath and Kallioniemi research groups working on phenotypic profiling of cells imaged with high-content microscopy. In 2017, Dr. Paavolainen worked as the head of the FIMM High Content Imaging and Analysis unit before starting as an Academy of Finland Postdoctoral Fellow studying deep learning for phenotypic profiling of cancer cells in Sep 2017. Paavolainen founded the Bioimage Profiling research group in Sep 2021 with Academy of Finland funding. Currently, Dr. Paavolainen leads the research activities of the group as Academy of Finland Research Fellow, leads the development of Life Science Data Competence Center at the University of Helsinki, and acts as a board member in the CytoData Society and the Nordic Microscopy Society.
Gantugs Atarsaikhan is a postdoctoral researcher in the Bioimage Profiling group. His research focuses on developing novel methods for analyzing fluorescent microscopy images. In particular, he works with multiplexed fluorescent images of various cancer tissues to identify relevant cancer biomarkers using self-supervised deep learning techniques.
He earned both his M.Sc. and Ph.D. in Information Science from Kyushu University, Japan, completing his doctoral degree in 2022. His Ph.D. research centered on developing algorithms for automatic font and logo generation using deep learning, merging computational creativity with visual design to produce unique and visually engaging outputs.
Isabel is a postdoc working in multidisciplinary joint projects with the Paavolainen, Kallionemi/Pietiäinen team and Pitkänen research groups, along with having responsibilities within the FIMM-HCA core unit. Isabel completed her bachelor's degree as a Biologist in Venezuela and her master’s thesis at Åbo Academi (Turku) in the field of Biomedical Imaging. In her doctoral work, Isabel studied mouse development, using single-cell resolution imaging and image analysis methods to study how single cells contribute to the formation of whole embryonic tissues. At FIMM, Isabel is working on both the imaging and the establishment of advanced image analysis technologies, applied to translational research projects in the field of cancer biology and personalised medicine. In particular, her focus is on developing an AI-driven platform for bioimage phenotypic profiling of the cancer architecture and microenvironment for functional precision medicine, working jointly with solid tumor slides and cancer organoids. Among other things, Isabel is interested in science communication, and outside work she is a proficient yoga practitioner.
Mohammad is currently a Ph.D. student in the Bioimage Profiling research group, where he is working on large-scale representation learning of biological samples. His research focuses on training transcriptomics and imaging models to address complex and pressing challenges in biology, with the goal of developing tools that can be directly applied by biologists in their research. He is also developing methods to integrate these two modalities, thereby enhancing the models’ ability to interpret and understand intricate biological samples.
Before his Ph.D., Mohammad completed his Master's degree at Sharif University of Technology. During this time, he focused on robust representation learning and predicting drug responses using cell microscopy images.
Franziska Bentz received her M.Sc. in Biology from the Karlsruhe Institute of Technology, Germany. During her Master’s she started to develop computational methods to simplify or improve data analysis compensating for technical and biological errors introduced during the wet-lab procedure. To further enhance both her computational and laboratory skills in a biomedical environment, she got enrolled in the FIMM-EMBL-DSHealth PhD program in 2021, where she did all three of her rotations in the field of cancer research. She is now doing a joint PhD in Lassi Paavolainen's and Tero Aittokallio's research groups.