AI Models Close to your Chest: Robust Federated Learning Strategies for Multi-site CT
Authors:
Edward H. Lee,
Brendan Kelly,
Emre Altinmakas,
Hakan Dogan,
Maryam Mohammadzadeh,
Errol Colak,
Steve Fu,
Olivia Choudhury,
Ujjwal Ratan,
Felipe Kitamura,
Hernan Chaves,
Jimmy Zheng,
Mourad Said,
Eduardo Reis,
Jaekwang Lim,
Patricia Yokoo,
Courtney Mitchell,
Golnaz Houshmand,
Marzyeh Ghassemi,
Ronan Killeen,
Wendy Qiu,
Joel Hayden,
Farnaz Rafiee,
Chad Klochko,
Nicholas Bevins
, et al. (5 additional authors not shown)
Abstract:
While it is well known that population differences from genetics, sex, race, and environmental factors contribute to disease, AI studies in medicine have largely focused on locoregional patient cohorts with less diverse data sources. Such limitation stems from barriers to large-scale data share and ethical concerns over data privacy. Federated learning (FL) is one potential pathway for AI developm…
▽ More
While it is well known that population differences from genetics, sex, race, and environmental factors contribute to disease, AI studies in medicine have largely focused on locoregional patient cohorts with less diverse data sources. Such limitation stems from barriers to large-scale data share and ethical concerns over data privacy. Federated learning (FL) is one potential pathway for AI development that enables learning across hospitals without data share. In this study, we show the results of various FL strategies on one of the largest and most diverse COVID-19 chest CT datasets: 21 participating hospitals across five continents that comprise >10,000 patients with >1 million images. We also propose an FL strategy that leverages synthetically generated data to overcome class and size imbalances. We also describe the sources of data heterogeneity in the context of FL, and show how even among the correctly labeled populations, disparities can arise due to these biases.
△ Less
Submitted 13 April, 2023; v1 submitted 23 March, 2023;
originally announced March 2023.
Searching Neutrino-Nucleus Coherent Scattering with Mössbauer Spectroscopy
Authors:
C. Marques,
G S Dias,
H. S. Chaves,
S. B. Duarte
Abstract:
The Mössbauer technique is proposed as an alternative experimental procedure to be used in the detection of Coherent Elastic $ν$-Nucleus Scattering (CENNS). The $Z_{0}$-boson exchange is considered as a perturbation on the nuclear mean-field potential with a change in the valence neutron quantum states in the $^{57}Fe$ nucleus of the Mössbauer detector sample. This nuclei is a typical isotope used…
▽ More
The Mössbauer technique is proposed as an alternative experimental procedure to be used in the detection of Coherent Elastic $ν$-Nucleus Scattering (CENNS). The $Z_{0}$-boson exchange is considered as a perturbation on the nuclear mean-field potential with a change in the valence neutron quantum states in the $^{57}Fe$ nucleus of the Mössbauer detector sample. This nuclei is a typical isotope used in Mössbauer spectroscopy. Perturbed level of the valence neutron accommodates the transferred energy, modifying the location of the isometric peak of the Mössbauer electromagnetic resonance. We calculate the isometric shift correction due CENNS and show that this quantity is able to be detected with enough precision. Therefore, the difference between the Mössbauer isometric shift in the presence of a reactor-neutrino beam and without the neutrinos flux is pointed out as a figure of merit to manifest CENNS. In this work, we show that the CENNS correction of the Isomeric Shift is of $\approx 10^{-7}$eV, which is greater than the $10^{-10}$eV resolution of the technique.
△ Less
Submitted 24 May, 2018; v1 submitted 12 April, 2018;
originally announced April 2018.
