2023

1. Repurposing Traditional U-Net Predictions for Sparse SAM Prompting in Medical Image Segmentation.

   ZM Colbert, D Arrington, M Foote, and 5 more authors

   Biomedical Physics & Engineering Express, 2023
2. Dynamic Privacy Allocation for Locally Differentially Private Federated Learning with Composite Objectives

   Jiaojiao Zhang, Dominik Fay, and Mikael Johansson

   CoRR, 2023

   PDF
3. Privacy Amplification via Importance Sampling

   Dominik Fay, Sebastian Mair, and Jens Sjölund

   CoRR, 2023

   PDF
4. Towards Scalable Machine Learning with Privacy Protection

   Dominik Fay

   KTH Royal Institute of Technology, 2023

   PDF
5. Adaptive Hyperparameter Selection for Differentially Private Gradient Descent

   Dominik Fay, Sindri Magnússon, Jens Sjölund, and 1 more author

   Transactions on Machine Learning Research, 2023

   PDF

2022

1. Radiotherapy plan parameters with privacy guarantees

   Dominik Fay, and Jens Olof Sjolund

   Aug 2022

   US Patent App. 17/595,488
2. Private Learning Via Knowledge Transfer with High-Dimensional Targets

   Dominik Fay, Jens Sjölund, and Tobias J. Oechtering

   In IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022, Virtual and Singapore, 23-27 May 2022, Aug 2022

2019

1. Privacy of Real-Time Pricing in Smart Grid

   Mahrokh Ghoddousi Boroujeni, Dominik Fay, Christos Dimitrakakis, and 1 more author

   In 58th IEEE Conference on Decision and Control, CDC 2019, Nice, France, December 11-13, 2019, Aug 2019
2. Membership Privacy in Neural Networks for Medical Image Segmentation

   Dominik Fay

   Aug 2019
3. METASPACE: A community-populated knowledge base of spatial metabolomes in health and disease

   Theodore Alexandrov, Katja Ovchinnikova, Andrew Palmer, and 54 more authors

   bioRxiv, Aug 2019

   Abs PDF

   Metabolites, lipids, and other small molecules are key constituents of tissues supporting cellular programs in health and disease. Here, we present METASPACE, a community-populated knowledge base of spatial metabolomes from imaging mass spectrometry data. METASPACE is enabled by a high-performance engine for metabolite annotation in a confidence-controlled way that makes results comparable between experiments and laboratories. By sharing their results publicly, engine users continuously populate a knowledge base of annotated spatial metabolomes in tissues currently including over 3000 datasets from human cancer cohorts, whole-body sections of animal models, and various organs. The spatial metabolomes can be visualized, explored and shared using a web app as well as accessed programmatically for large-scale analysis. By using novel computational methods inspired by natural language processing, we illustrate that METASPACE provides molecular coverage beyond the capacity of any individual laboratory and opens avenues towards comprehensive metabolite atlases on the levels of tissues and organs.

2018

1. Curatr: a web application for creating, curating and sharing a mass spectral library

   Andrew Palmer, Prasad Phapale, Dominik Fay, and 1 more author

   Bioinform., Aug 2018
2. Implementing Adaptive Separable Convolution for Video Frame Interpolation

   Mart Kartasev, Carlo Rapisarda, and Dominik Fay

   CoRR, Aug 2018

   PDF

2017

1. FDR-controlled metabolite annotation for high-resolution imaging mass spectrometry

   Andrew Palmer, Prasad Phapale, Ilya Chernyavsky, and 8 more authors

   Nature methods, Aug 2017