Nadezhda Koriakina

Postdoctoral position at Department of Materials Science and Engineering; Biomedical Engineering

E-mail:: nadezhda.koriakina@angstrom.uu.se
Visiting address:: Ångströmlaboratoriet, Regementsvägen 10
Postal address:: Box 35
751 03 UPPSALA

Download contact information for Nadezhda Koriakina at Department of Materials Science and Engineering; Biomedical Engineering

Short presentation

Nadezhda joined the EMBLA research group - Enabling Microtechnologies for Biomedical and Life science Applications - in November 2024. Her research is focused on the analysis of data coming from organs-on-chip devices.

Keywords

machine learning
data driven life science
precision medicine

Biography

2024-, PostDoc in EMBLA group, Uppsala University

2024, PhD in Computerized Image Processing, Uppsala University

2016-2018, Postgraduate Studies in Advanced Medical Imaging, KU Leuven, Belgium

2015, MSc in Micro and Nano Systems Technology, Buskerud and Vestfold University College, Norway

Research

The interdisciplinary research project Nadezhda is involved in is aimed at exploring how machine learning and the combination of different types of data can be utilised to improve decisions during ongoing experiments on organs-on-chip as well as its potential for identifying new biological information.

Publications

Recent publications

All publications

Articles in journal

Deep multiple instance learning versus conventional deep single instance learning for interpretable oral cancer detection

Koriakina, Nadezhda; Sladoje, Nataša; Basic, Vladimir; Lindblad, Joakim

Part of PLOS ONE, 2024
- DOI for Deep multiple instance learning versus conventional deep single instance learning for interpretable oral cancer detection
- Download full text (pdf) of Deep multiple instance learning versus conventional deep single instance learning for interpretable oral cancer detection

Comprehensive doctoral thesis

Deep Learning and Explainable Artificial Intelligence for Biomedical Applications: Methods for Cytology-based Cancer Detection

Koriakina, Nadezhda

2024
- Download full text (pdf) of Deep Learning and Explainable Artificial Intelligence for Biomedical Applications: Methods for Cytology-based Cancer Detection

Conference papers

Towards robust bubble detection in diverse microphysiological systems by machine learning

Koriakina, Nadezhda; Sladoje, Nataša; Tenje, Maria; Johansson, Sofia

2025
End-to-end Multiple Instance Learning with Gradient Accumulation

Andersson, Axel; Koriakina, Nadezhda; Sladoje, Nataša; Lindblad, Joakim

Part of 2022 IEEE International Conference on Big Data (Big Data), p. 2742-2746, 2022
- DOI for End-to-end Multiple Instance Learning with Gradient Accumulation
The Effect of Within-Bag Sampling on End-to-End Multiple Instance Learning

Koriakina, Nadezhda; Sladoje, Natasa; Lindblad, Joakim

Part of Proceedings of the 12th International Symposium on Image and Signal Processing and Analysis (ISPA), p. 183-188, 2021
- DOI for The Effect of Within-Bag Sampling on End-to-End Multiple Instance Learning
Uncovering hidden reasoning of convolutional neural networks in biomedical image classification by using attribution methods

Koriakina, Nadezhda; Sladoje, Natasa; Wetzer, Elisabeth; Lindblad, Joakim

Part of 4th NEUBIAS Conference, Bordeaux, France, 2020
Visualization of convolutional neural network class activations in automated oral cancer detection for interpretation of malignancy associated changes

Koriakina, Nadezhda; Sladoje, Natasa; Bengtsson, Ewert; Ramqvist, Eva Darai et al.

Part of 3rd NEUBIAS Conference, Luxembourg, 2-8 February 2019, 2019

Manuscripts (preprints)

Sensitive key instance detection: Helping a cytotechnologist to find a needle in a haystack

Koriakina, Nadezhda; Sladoje, Nataša; Basic, Vladimir; Lindblad, Joakim