Zhaolin Chen

Monash University (Australia) · Data science and AI

Deep learning (DL) models are effective in leveraging latent representations from MR data, emerging as state-of-the-art solutions for accelerated MRI reconstruction. However, challenges arise due to the inherent uncertainties associated with undersampling in k-space, coupled with the over- or under-parameterized and opaque nature of DL models. Addr...

To estimate proton density fat fraction (PDFF) from chemical shift encoded (CSE) MR images using a deep learning (DL)-based method that is precise and robust to different MR scanners and acquisition echo times (TEs). Variable echo times neural network (VET-Net) is a two-stage framework that first estimates nonlinear variables of the CSE-MR signal m...

Accurate hippocampus segmentation in brain MRI is critical for studying cognitive and memory functions and diagnosing neurodevelopmental disorders. While high-field MRIs provide detailed imaging, low-field MRIs are more accessible and cost-effective, which eliminates the need for sedation in children, though they often suffer from lower image quali...

Objective. The acceleration of magnetic resonance imaging (MRI) acquisition is crucial for both clinical and research applications, particularly in dynamic MRI. Existing compressed sensing (CS) methods, despite being effective for fast imaging, face limitations such as the need for incoherent sampling and residual noise, which restrict their practi...

Low-dose positron emission tomography (PET) image reconstruction methods have potential to significantly improve PET as an imaging modality. Deep learning provides a promising means of incorporating prior information into the image reconstruction problem to produce quantitatively accurate images from compromised signal. Deep learning-based methods...

Implicit Neural Representations (INRs) have recently advanced the field of deep learning due to their ability to learn continuous representations of signals without the need for large training datasets. Although INR methods have been studied for medical image super-resolution, their adaptability to localized priors in medical images has not been ex...

Magnetic Resonance Imaging (MRI) is instrumental in clinical diagnosis, offering diverse contrasts that provide comprehensive diagnostic information. However, acquiring multiple MRI contrasts is often constrained by high costs, long scanning durations, and patient discomfort. Current synthesis methods, typically focused on single-image contrasts, f...

This article proposes a practical and generalizable object detector, termed feature extraction-fusion-prediction network (FEFP-Net) for real-world application scenarios. The existing object detection methods have recently achieved excellent performance, however they still face three major challenges for real-world applications, i.e., feature simila...

Motion artifacts in Magnetic Resonance Imaging (MRI) are one of the frequently occurring artifacts due to patient movements during scanning. Motion is estimated to be present in approximately 30% of clinical MRI scans; however, motion has not been explicitly modeled within deep learning image reconstruction models. Deep learning (DL) algorithms hav...

Functional Positron Emission Tomography (fPET) has advanced as an effective tool for investigating dynamic processes in glucose metabolism and neurotransmitter action, offering potential insights into brain function, disease progression, and treatment development. Despite significant methodological advances, extracting stimulation-specific informat...

As intensities of MRI volumes are inconsistent across institutes, it is essential to extract universal features of multi-modal MRIs to precisely segment brain tumors. In this concept, we propose a volumetric vision transformer that follows two windowing strategies in attention for extracting fine features and local distributional smoothness (LDS) d...

Low-field portable magnetic resonance imaging (MRI) scanners are more accessible, cost-effective, sustainable with lower carbon emissions than superconducting high-field MRI scanners. However, the images produced have relatively poor image quality, lower signal-to-noise ratio, and limited spatial resolution. This study develops and investigates an...

Deep learning has led to tremendous progress in the field of medical artificial intelligence. However, training deep-learning models usually require large amounts of annotated data. Annotating large-scale datasets is prone to human biases and is often very laborious, especially for dense prediction tasks such as image segmentation. Inspired by semi...

Automatic defect detection plays an important role in industrial production. Deep learning-based defect detection methods have achieved promising results. However, there are still two challenges in the current defect detection methods: 1) high-precision detection of weak defects is limited and 2) it is difficult for current defect detection methods...

Level of Evidence 5 Technical Efficacy Stage 3

Purpose: We propose a novel contrastive learning latent space representation for MRI datasets with partially acquired scans. We show that this latent space can be utilized for accelerated MR image reconstruction. Theory and Methods: Our novel framework, referred to as COLADA (stands for Contrastive Learning for highly accelerated MR image reconstru...

Deep learning (DL) models are capable of successfully exploiting latent representations in MR data and have become state-of-the-art for accelerated MRI reconstruction. However, undersampling the measurements in k-space as well as the over- or under-parameterized and non-transparent nature of DL make these models exposed to uncertainty. Consequently...

Long acquisition times are a significant limitation in the clinical utility of diagnostic MRI. Whilst acquisition times can be reduced by under-sampling the measurements in k-space, there is a resultant loss of quality in the reconstructed image. Recently, artificial intelligence and specifically deep learning models have been utilized to transform...

Magnetic resonance imaging (MRI) provides excellent soft-tissue contrast for clinical diagnoses and research which underpin many recent breakthroughs in medicine and biology. The post-processing of reconstructed MR images is often automated for incorporation into MRI scanners by the manufacturers and increasingly plays a critical role in the final...

We propose a Transformer architecture for volumetric segmentation, a challenging task that requires keeping a complex balance in encoding local and global spatial cues, and preserving information along all axes of the volume. Encoder of the proposed design benefits from self-attention mechanism to simultaneously encode local and global cues, while...

As intensities of MRI volumes are inconsistent across institutes, it is essential to extract universal features of multi-modal MRIs to precisely segment brain tumors. In this concept, we propose a volumetric vision transformer that follows two windowing strategies in attention for extracting fine features and local distributional smoothness (LDS) d...

Objective Emerging evidences suggest that the trans-neural propagation of phosphorylated 43-kDa transactive response DNA-binding protein (pTDP-43) contributes to neurodegeneration in Amyotrophic Lateral Sclerosis (ALS). We investigated whether Network Diffusion Model (NDM), a biophysical model of spread of pathology via the brain connectome, could...

Global correlations are widely seen in human anatomical structures due to similarity across tissues and bones. These correlations are reflected in magnetic resonance imaging (MRI) scans as a result of close-range proton density and T1/T2 parameter. Furthermore, to achieve accelerated MRI, k-space data are undersampled which causes global aliasing a...

We propose Orthogonal-Nets consisting of a large number of ensembles of 2D encoder-decoder convolutional neural networks. The Orthogonal-Nets takes 2D slices of the image from axial, sagittal, and coronal views of the 3D brain volume and predicts the probability for the tumor segmentation region. The predicted probability distributions from all thr...

A deep learning method is proposed for brain tumor segmentation using a two-stage encoder-decoder convolutional neural network (CNN). To improve the generalization of the proposed network for federated evaluation, we propose a two-stage encoder-decoder CNN that performs coarse segmentation at stage-I and fine segmentation at stage-II. Stage-I consi...

This paper proposes an adversarial learning based training approach for brain tumor segmentation task. In this concept, the 3D segmentation network learns from dual reciprocal adversarial learning approaches. To enhance the generalization across the segmentation predictions and to make the segmentation network robust, we adhere to the Virtual Adver...

Parcellation of whole brain tractograms is a critical step to study brain white matter structures and connectivity patterns. The existing methods based on supervised classification of streamlines into predefined streamline bundle types are not designed to explore sub-bundle structures, and methods with manually designed features are expensive to co...

Deep learning (DL) models for accelerated image reconstruction involves retrospective undersampling of the fully sampled k-space data for training and validation. This strategy is not a true reflection of real-world data and in many instances, the input k-space data is corrupted with artifacts and errors, such as motion artifacts. In this work, we...

Background “Functional” [18F]-fluorodeoxyglucose positron emission tomography (FDG-fPET) is a new approach for measuring glucose uptake in the human brain. The goal of FDG-fPET is to maintain a constant plasma supply of radioactive FDG in order to track, with high temporal resolution, the dynamic uptake of glucose during neuronal activity that occu...

Image processing plays a crucial role in maximising diagnostic quality of positron emission tomography (PET) images. Recently, deep learning methods developed across many fields have shown tremendous potential when applied to medical image enhancement, resulting in a rich and rapidly advancing literature surrounding this subject. This review encaps...

Dysfunction of fronto-striato-thalamic (FST) circuits is thought to contribute to dopaminergic dysfunction and symptom onset in psychosis, but it remains unclear whether this dysfunction is driven by aberrant bottom-up subcortical signaling or impaired top-down cortical regulation. We used spectral dynamic causal modelling of resting-state function...

This paper proposes an adversarial learning based training approach for brain tumor segmentation task. In this concept, the 3D segmentation network learns from dual reciprocal adversarial learning approaches. To enhance the generalization across the segmentation predictions and to make the segmentation network robust, we adhere to the Virtual Adver...

This paper presents a Transformer architecture for volumetric medical image segmentation. Designing a computationally efficient Transformer architecture for volumetric segmentation is a challenging task. It requires keeping a complex balance in encoding local and global spatial cues, and preserving information along all axes of the volumetric data....

Accelerated MRI reconstruction attempts to find an optimal reconstruction function that takes highly undersampled k-space data as input and produces clean MR images which are suitable for medical diagnosis. Motivated by the Compressed Sensing theory, recent deep learning methods have been successful in exploiting the sparse latent representations i...

Understanding how the living human brain functions requires sophisticated in vivo neuroimaging technologies to characterise the complexity of neuroanatomy, neural function, and brain metabolism. Fluorodeoxyglucose positron emission tomography (FDG-PET) studies of human brain function have historically been limited in their capacity to measure dynam...

Parcellation of whole brain tractograms is a critical step to study brain white matter structures and connectivity patterns. The existing methods based on supervised classification of streamlines into predefined streamline bundle types are not designed to explore sub-bundle structures, and methods with manually designed features are expensive to co...

Segmentation of images is a long-standing challenge in medical AI. This is mainly due to the fact that training a neural network to perform image segmentation requires a significant number of pixel-level annotated data, which is often unavailable. To address this issue, we propose a semi-supervised image segmentation technique based on the concept...

Segmentation of images is a long-standing challenge in medical AI. This is mainly due to the fact that training a neural network to perform image segmentation requires a significant number of pixel-level annotated data, which is often unavailable. To address this issue, we propose a semi-supervised image segmentation technique based on the concept...

Background: Functional [18F]-fluorodeoxyglucose positron emission tomography (FDG-fPET) is a new approach for measuring glucose uptake in the human brain. The goal of FDG-fPET is to maintain a constant plasma supply of radioactive FDG in order to track, with high temporal resolution, the dynamic uptake of glucose during neuronal activity that occur...

A deep learning (DL) method for accelerated magnetic resonance (MR) imaging is presented that incorporates domain knowledge of parallel MR imaging to augment the DL networks for accurate and stable image reconstruction. The proposed DL method employs a novel loss function consisting of a combination of mean absolute error, structural similarity, an...

Radiation exposure in positron emission tomography (PET) imaging limits its usage in the studies of radiation-sensitive populations, e.g., pregnant women, children, and adults that require longitudinal imaging. Reducing the PET radiotracer dose or acquisition time reduces photon counts, which can deteriorate image quality. Recent deep-neural-networ...

The trans-neural propagation of phosphorylated 43-kDa transactive response DNA-binding protein (pTDP-43) contributes to neurodegeneration in Amyotrophic Lateral Sclerosis (ALS). We investigated whether Network Diffusion Model (NDM), a biophysical model of spread of pathology via the brain connectome, could capture the severity and progression of ne...

Radiation exposure in positron emission tomography (PET) imaging limits its usage in the studies of radiation-sensitive populations, e.g., pregnant women, children, and adults that require longitudinal imaging. Reducing the PET radiotracer dose or acquisition time reduces photon counts, which can deteriorate image quality. Recent deep-neural-networ...

Synopsis Data-driven deep learning (DL) image reconstruction from undersampled data has become a mainstream research area in MR image reconstruction. The generalization of the model on unseen data and out of sample data distribution is still a concern for the adoption of the DL reconstruction. In this work, we present a method of risk assessment in...

Synopsis There is a lack of pre-trained deep learning model weights on large scale medical image dataset, due to privacy concerns. Federated learning enables training deep networks while preserving privacy. This work explored co-training multi-task models on multiple heterogeneous datasets, and validated the usage of federated learning could serve...

Synopsis Clustering white matter streamlines is still a challenging task. The existing methods based on spatial coordinates rely on manually engineered features, and/or labeled dataset. This work introduced a novel method that solves the problem of streamline clustering without needing labeled data. This is achieved by training a deep LSTM-based au...

Synopsis Deep learning (DL) methods are superior to the conventional method of accelerated imaging such as parallel imaging and compressed sensing but the integration of DL methods into the MR scanners is still in its infancy. The integration of the DL methods into the MR scanner requires the design of new pulse sequences with a modified sampling p...

Synopsis Simultaneous positron emission tomography (PET) and magnetic resonance imaging (MRI) provide complementary structural and functional information. Recent developments in continuous infusion functional PET (fPET) have shown promising results to track dynamic changes in brain metabolism. Although fPET provides opportunities to investigate fun...

Synopsis We performed in-vivo measurements of the magnetic susceptibility in the motor cortex in individuals with Amyotrophic Lateral Sclerosis (ALS) at baseline and six-month follow-up, and healthy controls at baseline using Quantitative Susceptibility Mapping (QSM). The results show significant susceptibility difference between individuals with A...

Synopsis The deep learning techniques have been shown to reduce the motion artifact in simulated motion scenarios and a few volunteer scans, the validation of it during routine clinical scans remains an unanswered question. In this study, we focus on evaluating the quality of images from the DL motion correction approach on a cohort of 27 actual pa...

Dysfunction of fronto-striato-thalamic (FST) circuits is thought to contribute to dopaminergic dysfunction and symptom onset in psychosis, but it remains unclear whether this dysfunction is driven by aberrant bottom-up subcortical signaling or impaired top-down cortical regulation. Here, we used spectral dynamic causal modelling (DCM) of resting-st...

Functional positron emission tomography (fPET) imaging using continuous infusion of [18F]-fluorodeoxyglucose (FDG) is a novel neuroimaging technique to track dynamic glucose utilization in the brain. In comparison to conventional static or dynamic bolus PET, fPET maintains a sustained supply of glucose in the blood plasma which improves sensitivity...

Amyotrophic lateral sclerosis (ALS) results in progressive impairment of upper and lower motor neurons. Increasing evidence from both in vivo and ex vivo studies suggest that iron accumulation in the motor cortex is a neuropathological hallmark in ALS. An in vivo neuroimaging marker of iron dysregulation in ALS would be useful in disease diagnosis...

Simultaneous [18F]-fluorodeoxyglucose positron emission tomography functional magnetic resonance imaging (FDG-PET/fMRI) provides the capacity to image 2 sources of energetic dynamics in the brain-glucose metabolism and the hemodynamic response. fMRI connectivity has been enormously useful for characterizing interactions between distributed brain ne...

Simultaneous magnetic resonance and positron emission tomography provides an opportunity to measure brain haemodynamics and metabolism in a single scan session, and to identify brain activations from multimodal measurements in response to external stimulation. However, there are few analysis methods available for jointly analysing the simultaneousl...

PurposeEstimation of accurate attenuation maps for whole-body positron emission tomography (PET) imaging in simultaneous PET-MRI systems is a challenging problem as it affects the quantitative nature of the modality. In this study, we aimed to improve the accuracy of estimated attenuation maps from MRI Dixon contrast images by training an augmented...

Purpose To evaluate the clinical utility of the application of a deep learning motion correction technique on 3D MPRAGE magnetic resonance images acquired in routine clinical practice. Methods An encoder-decoder deep learning network inspired by InceptionResnet was trained on public datasets. The clinical utility of the trained network was evaluat...

Simultaneous [18 F]-fluorodeoxyglucose positron emission tomography and functional magnetic resonance imaging (FDG-PET/fMRI) provides the capability to image two sources of energetic dynamics in the brain – cerebral glucose uptake and the cerebrovascular haemodynamic response. Resting-state fMRI connectivity has been enormously useful for character...