Results of the 2023 ISBI challenge to reduce GABA-edited MRS acquisition time

• Published in: Magma (New York, N.Y.) N.Y.), 2024-07, Vol.37 (3), p.449-463
• Main Authors: Berto, Rodrigo Pommot, Bugler, Hanna, Dias, Gabriel, Oliveira, Mateus, Ueda, Lucas, Dertkigil, Sergio, Costa, Paula D. P., Rittner, Leticia, Merkofer, Julian P., van de Sande, Dennis M. J., Amirrajab, Sina, Drenthen, Gerhard S., Veta, Mitko, Jansen, Jacobus F. A., Breeuwer, Marcel, van Sloun, Ruud J. G., Qayyum, Abdul, Rodero, Cristobal, Niederer, Steven, Souza, Roberto, Harris, Ashley D.
• Format: Article
• Language: English
• Subjects: Algorithms; Basic Science - MR spectroscopy methods; Biomedical Engineering and Bioengineering; Brain - diagnostic imaging; Brain - metabolism; Computer Appl. in Life Sciences; Computer Simulation; Deep Learning; gamma-Aminobutyric Acid - metabolism; Health Informatics; Humans; Image Processing, Computer-Assisted - methods; Imaging; Machine Learning; Magnetic Resonance Spectroscopy - methods; Medicine; Medicine & Public Health; Neural Networks, Computer; Radiology; Research Article; Signal-To-Noise Ratio; Solid State Physics
• ISSN: 1352-8661; 1352-8661
• DOI: 10.1007/s10334-024-01156-9

Purpose Use a conference challenge format to compare machine learning-based gamma-aminobutyric acid (GABA)-edited magnetic resonance spectroscopy (MRS) reconstruction models using one-quarter of the transients typically acquired during a complete scan. Methods There were three tracks: Track 1: simulated data, Track 2: identical acquisition parameters with in vivo data, and Track 3: different acquisition parameters with in vivo data. The mean squared error, signal-to-noise ratio, linewidth, and a proposed shape score metric were used to quantify model performance. Challenge organizers provided open access to a baseline model, simulated noise-free data, guides for adding synthetic noise, and in vivo data. Results Three submissions were compared. A covariance matrix convolutional neural network model was most successful for Track 1. A vision transformer model operating on a spectrogram data representation was most successful for Tracks 2 and 3. Deep learning (DL) reconstructions with 80 transients achieved equivalent or better SNR, linewidth and fit error compared to conventional 320 transient reconstructions. However, some DL models optimized linewidth and SNR without actually improving overall spectral quality, indicating a need for more robust metrics. Conclusion DL-based reconstruction pipelines have the promise to reduce the number of transients required for GABA-edited MRS.