Ruengchaijatuporn, N. | bioRxiv Science

bioRxiv2023

Deep Learning Initialized Compressed Sensing (Deli-CS) in Volumetric Spatio-Temporal Subspace Reconstruction

IntroductionSpatio-temporal MRI methods enable whole-brain multi-parametric mapping at ultra-fast acquisition times through efficient k-space encoding, but can have very long reconstruction times, which limit their integration into clinical practice. Deep learning (DL) is a promising approach to accelerate reconstruction, but can be computationally intensive to train and deploy due to the large dimensionality of spatio-temporal MRI. DL methods also need large training data sets and can produce results that dont match the acquired data if data consistency is not enforced. The aim of this project is to reduce reconstruction time using DL whilst simultaneously limiting the risk of deep learning induced hallucinations, all with modest hardware requirements. MethodsDeep Learning Initialized Compressed Sensing (Deli-CS) is proposed to reduce the reconstruction time of iterative reconstructions by "kick-starting" the iterative reconstruction with a DL generated starting point. The proposed framework is applied to volumetric multi-axis spiral projection MRF that achieves whole-brain T1 and T2 mapping at 1-mm isotropic resolution for a 2-minute acquisition. First, the traditional reconstruction is optimized from over two hours to less than 40 minutes while using more than 90% less RAM and only 4.7 GB GPU memory, by using a memory-efficient GPU implementation. The Deli-CS framework is then implemented and evaluated against the above reconstruction. ResultsDeli-CS achieves comparable reconstruction quality with 50% fewer iterations bringing the full reconstruction time to 20 minutes. ConclusionDeli-CS reduces the reconstruction time of subspace reconstruction of volumetric spatio-temporal acquisitions by providing a warm start to the iterative reconstruction algorithm.

neuroscience↗

bioRxiv2021

An explainable self-attention deep neural network for detecting mild cognitive impairment using multi-inputbdigital drawing tasks

Mild cognitive impairment (MCI) is an early stage of age-inappropriate cognitive decline, which could develop into dementia - an untreatable neurodegenerative disorder. An early detection of MCI is a crucial step for timely prevention and intervention. To tackle this problem, recent studies have developed deep learning models to detect MCI and various types of dementia using data obtained from the classic clock-drawing test (CDT), a popular neuropsychological screening tool that can be easily and rapidly implemented for assessing cognitive impairments in an aging population. While these models succeed at distinguishing severe forms of dementia, it is still difficult to predict the early stage of the disease using the CDT data alone. Also, the state-of-the-art deep learning techniques still face the black-box challenges, making it questionable to implement them in the clinical setting. Here, we propose a novel deep learning modeling framework that incorporates data from multiple drawing tasks including the CDT, cube-copying, and trail-making tasks obtained from a digital platform. Using self-attention and soft-label methods, our model achieves much higher classification performance at detecting MCI compared to those of a well-established convolutional neural network model. Moreover, our model can highlight features of the MCI data that considerably deviate from those of the healthy aging population, offering accurate predictions for detecting MCI along with visual explanation that aids the interpretation of the deep learning model.

neuroscience↗