3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies

Khvostikov, Alexander; Aderghal, Karim; Benois‐Pineau, Jenny; Krylov, Andrey S.; Catheline, Gwénaëlle

doi:10.48550/arxiv.1801.05968

Cited by 25 publications

(30 citation statements)

References 30 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present paper we continue our previous work [15]. We give a substantial overview of recent trends in classification of different brain imaging modalities in the problem of computer-aided diagnostics of Alzheimer's Disease and its prodromal stage, i.e.…”

Section: Introductionmentioning

confidence: 71%

3D Inception-based CNN with sMRI and MD-DTI data fusion for Alzheimer's Disease diagnostics

Khvostikov¹,

Aderghal²,

Krylov³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

In the last decade, computer-aided early diagnostics of Alzheimers Disease (AD) and its prodromal form, Mild Cognitive Impairment (MCI), has been the subject of extensive research. Some recent studies have shown promising results in the AD and MCI determination using structural and functional Magnetic Resonance Imaging (sMRI, fMRI), Positron Emission Tomography (PET) and Diffusion Tensor Imaging (DTI) modalities. Furthermore, fusion of imaging modalities in a supervised machine learning framework has shown promising direction of research.In this paper we first review major trends in automatic classification methods such as feature extraction based methods as well as deep learning approaches in medical image analysis applied to the field of Alzheimer's Disease diagnostics. Then we propose our own design of a 3D Inception-based Convolutional Neural Network (CNN) for Alzheimer's Disease diagnostics. The network is designed with an emphasis on the interior resource utilization and uses sMRI and DTI modalities fusion on hippocampal ROI. The comparison with the conventional AlexNet-based network using data from the Alzheimers Disease Neuroimaging Initiative (ADNI) dataset (http://adni.loni.usc.edu) demonstrates significantly better performance of the proposed 3D Inception-based CNN.

show abstract

Section: Introductionmentioning

confidence: 71%

3D Inception-based CNN with sMRI and MD-DTI data fusion for Alzheimer's Disease diagnostics

Khvostikov¹,

Aderghal²,

Krylov³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Accuracy Precision Recall F-score Baseline [18] 0.544 ± 0.140 0.478 ± 0.177 0.506 ± 0.135 0.489 ± 0.152 3D CNN fusion [11] 0.587 ± 0.093 0.479 ± 0.114 0.533 ± 0.071 0.501 ± 0.141 TPCNN [26] 0.636 ± 0.000 ------Multi-channel ResNet 0.624 ± 0.084 0.556 ± 0.167 0.572 ± 0.091 0.559 ± 0.136 Modality-specific nets 0.621 ± 0.069 0.580 ± 0.177 0.576 ± 0.067 0.569 ± 0.110 Multi-modal shared net 0.611 ± 0.071 0.449 ± 0.071 0.566 ± 0.045 0.499 ± 0.056 M 2 Net 0.664 ± 0.061 0.574 ± 0.141 0.613 ± 0.075 0.589 ± 0.102 using a bilinear model and then obtains the final prediction result. Note that, for fair comparison, we use the supplemental features for all comparison methods.…”

Section: Methodsmentioning

confidence: 99%

“…• 3D CNN fusion method. We use a 3D CNN [11] for each modality (i.e., the extracted 3D patches) and then fuse them using a bilinear model (denoted as "3D CNN + fusion"). • TPCNN [26].…”

Section: Experimental Settingsmentioning

confidence: 99%

M2Net: Multi-modal Multi-channel Network for Overall Survival Time Prediction of Brain Tumor Patients

Zhou,

Fu,

Zhang

et al. 2020

Preprint

View full text Add to dashboard Cite

Early and accurate prediction of overall survival (OS) time can help to obtain better treatment planning for brain tumor patients. Although many OS time prediction methods have been developed and obtain promising results, there are still several issues. First, conventional prediction methods rely on radiomic features at the local lesion area of a magnetic resonance (MR) volume, which may not represent the full image or model complex tumor patterns. Second, different types of scanners (i.e., multi-modal data) are sensitive to different brain regions, which makes it challenging to effectively exploit the complementary information across multiple modalities and also preserve the modality-specific properties. Third, existing methods focus on prediction models, ignoring complex data-to-label relationships. To address the above issues, we propose an end-to-end OS time prediction model; namely, Multi-modal Multi-channel Network (M 2 Net). Specifically, we first project the 3D MR volume onto 2D images in different directions, which reduces computational costs, while preserving important information and enabling pre-trained models to be transferred from other tasks. Then, we use a modality-specific network to extract implicit and high-level features from different MR scans. A multi-modal shared network is built to fuse these features using a bilinear pooling model, exploiting their correlations to provide complementary information. Finally, we integrate the outputs from each modality-specific network and the multi-modal shared network to generate the final prediction result. Experimental results demonstrate the superiority of our M 2 Net model over other methods.

show abstract

“…After this step, all the PET scans resulted in volumes of 193 * 229 * 193. We developed a 3D Convolutional Neural Network inspired by the architecture [6] for Alzheimer's Disease diagnosis. The CNN architecture is depicted in Figure 1.…”

Section: Methodsmentioning

confidence: 99%

Understanding 3D CNN Behavior for Alzheimer's Disease Diagnosis from Brain PET Scan

Islam¹,

Zhang²

2019

Preprint

View full text Add to dashboard Cite

In recent days, Convolutional Neural Networks (CNN) have demonstrated impressive performance in medical image analysis. However, there is a lack of clear understanding of why and how the Convolutional Neural Network performs so well for image analysis tasks. How CNN analyzes an image and discriminates among samples of different classes are usually considered as non-transparent. As a result, it becomes difficult to apply CNN based approaches in clinical procedures and automated disease diagnosis systems. In this paper, we consider this issue and work on visualizing and understanding the decision of Convolutional Neural Network for Alzheimer's Disease (AD) Diagnosis. We develop a 3D deep convolutional neural network for AD diagnosis using brain PET scans and propose using five visualizations techniques -Sensitivity Analysis (Backpropagation), Guided Backpropagation, Occlusion, Brain Area Occlusion, and Layer-wise Relevance Propagation (LRP) to understand the decision of the CNN by highlighting the relevant areas in the PET data.

show abstract

3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies

Cited by 25 publications

References 30 publications

3D Inception-based CNN with sMRI and MD-DTI data fusion for Alzheimer's Disease diagnostics

3D Inception-based CNN with sMRI and MD-DTI data fusion for Alzheimer's Disease diagnostics

M2Net: Multi-modal Multi-channel Network for Overall Survival Time Prediction of Brain Tumor Patients

Understanding 3D CNN Behavior for Alzheimer's Disease Diagnosis from Brain PET Scan

Contact Info

Product

Resources

About