Reliability of Quantification Estimates in MR Spectroscopy: CNNs vs Traditional Model Fitting

Rizzo, Rudy; Dziadosz, Martyna; Kyathanahally, Sreenath P.; Reyes, Mauricio; Kreis, Roland (2022). Reliability of Quantification Estimates in MR Spectroscopy: CNNs vs Traditional Model Fitting. Lecture notes in computer science, 13438, pp. 715-724. Springer 10.1007/978-3-031-16452-1_68

[img] Text
978-3-031-16452-1_68.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (3MB) | Request a copy

Magnetic Resonance Spectroscopy (MRS) and Spectroscopic Imaging (MRSI) are non-invasive techniques to map tissue contents of many metabolites in situ in humans. Quantification is traditionally done via model fitting (MF), and Cramer Rao Lower Bounds (CRLBs) are used as a measure of fitting uncertainties. Signal-to-noise is limited due to clinical time constraints and MF can be very time-consuming in MRSI with thousands of spectra. Deep Learning (DL) has introduced the possibility to speed up quantitation while reportedly preserving accuracy and precision. However, questions arise about how to access quantification uncertainties in the case of DL. In this work, an optimal-performance DL architecture that uses spectrograms as input and maps absolute concentrations of metabolites referenced to water content as output was taken to investigate this in detail. Distributions of predictions and Monte-Carlo dropout were used to investigate data and model-related uncertainties, exploiting ground truth knowledge in a synthetic setup mimicking realistic brain spectra with metabolic composition that uniformly varies from healthy to pathological cases. Bias and CRLBs from MF are then compared to DL-related uncertainties. It is confirmed that DL is a dataset-biased technique where accuracy and precision of predictions scale with metabolite SNR but hint towards bias and increased uncertainty at the edges of the explored parameter space (i.e., for very high and very low concentrations), even at infinite SNR (noiseless training and testing). Moreover, training with uniform datasets or if augmented with critical cases showed to be insufficient to prevent biases. This is dangerous in a clinical context that requires the algorithm to be unbiased also for concentrations far from the norm, which may well be the focus of the investigation since these correspond to pathology, the target of the diagnostic investigation

Item Type:

Journal Article (Original Article)


04 Faculty of Medicine > Faculty Institutions > sitem Center for Translational Medicine and Biomedical Entrepreneurship
04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Institute of Diagnostic, Interventional and Paediatric Radiology > DCR Magnetic Resonance Spectroscopy and Methodology (AMSM)
04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Institute of Diagnostic and Interventional Neuroradiology
10 Strategic Research Centers > ARTORG Center for Biomedical Engineering Research

Graduate School:

Graduate School for Cellular and Biomedical Sciences (GCB)

UniBE Contributor:

Rizzo, Rudy, Dziadosz, Martyna, Reyes, Mauricio, Kreis, Roland


600 Technology > 610 Medicine & health
500 Science > 570 Life sciences; biology








Roland Kreis

Date Deposited:

20 Dec 2022 14:39

Last Modified:

07 Jun 2023 13:27

Publisher DOI:


Web of Science ID:


Additional Information:

Anyone you share the following link with will be able to read this content:
Provided by the Springer Nature SharedIt content-sharing initiative

Medical Image Computing and Computer Assisted Intervention – MICCAI 2022. - ISBN 978-3-031-16452-1




Actions (login required)

Edit item Edit item
Provide Feedback