# AD-Prediction Convolutional Neural Networks for Alzheimer's Disease Prediction Using Brain MRI Image ## Abstract Alzheimers disease (AD) is characterized by severe memory loss and cognitive impairment. It associates with significant brain structure changes, which can be measured by magnetic resonance imaging (MRI) scan. The observable preclinical structure changes provides an opportunity for AD early detection using image classification tools, like convolutional neural network (CNN). However, currently most AD related studies were limited by sample size. Finding an efficient way to train image classifier on limited data is critical. In our project, we explored different transfer-learning methods based on CNN for AD prediction brain structure MRI image. We find that both pretrained 2D AlexNet with 2D-representation method and simple neural network with pretrained 3D autoencoder improved the prediction performance comparing to a deep CNN trained from scratch. The pretrained 2D AlexNet performed even better (**86%**) than the 3D CNN with autoencoder (**77%**). ## Method #### 1. Data In this project, we used public brain MRI data from **Alzheimers Disease Neuroimaging Initiative (ADNI)** Study. ADNI is an ongoing, multicenter cohort study, started from 2004. It focuses on understanding the diagnostic and predictive value of Alzheimers disease specific biomarkers. The ADNI study has three phases: ADNI1, ADNI-GO, and ADNI2. Both ADNI1 and ADNI2 recruited new AD patients and normal control as research participants. Our data included a total of 686 structure MRI scans from both ADNI1 and ADNI2 phases, with 310 AD cases and 376 normal controls. We randomly derived the total sample into training dataset (n = 519), validation dataset (n = 100), and testing dataset (n = 67). #### 2. Image preprocessing Image preprocessing were conducted using Statistical Parametric Mapping (SPM) software, version 12. The original MRI scans were first skull-stripped and segmented using segmentation algorithm based on 6-tissue probability mapping and then normalized to the International Consortium for Brain Mapping template of European brains using affine registration. Other configuration includes: bias, noise, and global intensity normalization. The standard preprocessing process output 3D image files with an uniform size of 121x145x121. Skull-stripping and normalization ensured the comparability between images by transforming the original brain image into a standard image space, so that same brain substructures can be aligned at same image coordinates for different participants. Diluted or enhanced intensity was used to compensate the structure changes. the In our project, we used both whole brain (including both grey matter and white matter) and grey matter only. #### 3. AlexNet and Transfer Learning Convolutional Neural Networks (CNN) are very similar to ordinary Neural Networks. A CNN consists of an input and an output layer, as well as multiple hidden layers. The hidden layers are either convolutional, pooling or fully connected. ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the amount of parameters in the network. #### 3.1. AlexNet The net contains eight layers with weights; the first five are convolutional and the remaining three are fully connected. The overall architecture is shown in Figure 1. The output of the last fully-connected layer is fed to a 1000-way softmax which produces a distribution over the 1000 class labels. AlexNet maximizes the multinomial logistic regression objective, which is equivalent to maximizing the average across training cases of the log-probability of the correct label under the prediction distribution. The kernels of the second, fourth, and fifth convolutional layers are connected only to those kernel maps in the previous layer which reside on the same GPU (as shown in Figure1). The kernels of the third convolutional layer are connected to all kernel maps in the second layer. The neurons in the fully connected layers are connected to all neurons in the previous layer. Response-normalization layers follow the first and second convolutional layers. Max-pooling layers follow both response-normalization layers as well as the fifth convolutional layer. The ReLU non-linearity is applied to the output of every convolutional and fully-connected layer.  The first convolutional layer filters the 224x224x3 input image with 96 kernels of size 11x11x3 with a stride of 4 pixels (this is the distance between the receptive field centers of neighboring neurons in a kernel map). The second convolutional layer takes as input the (response-normalized and pooled) output of the first convolutional layer and filters it with 256 kernels of size 5x5x48. The third, fourth, and fifth convolutional layers are connected to one another without any intervening pooling or normalization layers. The third convolutional layer has 384 kernels of size 3x3x256 connected to the (normalized, pooled) outputs of the second convolutional layer. The fourth convolutional layer has 384 kernels of size 3x3x192 , and the fifth convolutional layer has 256 kernels of size 3x3x192. The fully-connected layers have 4096 neurons each. #### 3.2. Transfer Learning Training an entire Convolutional Network from scratch (with random initialization) is impractical[14] because it is relatively rare to have a dataset of sufficient size. An alternative is to pretrain a Conv-Net on a very large dataset (e.g. ImageNet), and then use the ConvNet either as an initialization or a fixed feature extractor for the task of interest. Typically, there are three major transfer learning scenarios: **ConvNet as fixed feature extractor:** We can take a ConvNet pretrained on ImageNet, and remove the last fully-connected layer, then treat the rest structure as a fixed feature extractor for the target dataset. In AlexNet, this would be a 4096-D vector. Usually, we call these features as CNN codes. Once we get these features, we can train a linear classifier (e.g. linear SVM or Softmax classifier) for our target dataset. **Fine-tuning the ConvNet:** Another idea is not only replace the last fully-connected layer in the classifier, but to also fine-tune the parameters of the pretrained network. Due to overfitting concerns, we can only fine-tune some higher-level part of the network. This suggestion is motivated by the observation that earlier features in a ConvNet contains more generic features (e.g. edge detectors or color blob detectors) that can be useful for many kind of tasks. But the later layer of the network becomes progressively more specific to the details of the classes contained in the original dataset. **Pretrained models:** The released pretrained model is usually the final ConvNet checkpoint. So it is common to see people use the network for fine-tuning. #### 4. 3D Autoencoder and Convolutional Neural Network We take a two-stage approach where we first train a 3D sparse autoencoder to learn filters for convolution operations, and then build a convolutional neural network whose first layer uses the filters learned with the autoencoder.  #### 4.1. Sparse Autoencoder An autoencoder is a 3-layer neural network that is used to extract features from an input such as an image. Sparse representations can provide a simple interpretation of the input data in terms of a small number of \parts by extracting the structure hidden in the data. The autoencoder has an input layer, a hidden layer and an output layer, and the input and output layers have same number of units, while the hidden layer contains more units for a sparse and overcomplete representation. The encoder function maps input x to representation h, and the decoder function maps the representation h to the output x. In our problem, we extract 3D patches from scans as the input to the network. The decoder function aims to reconstruct the input form the hidden representation h. #### 4.2. 3D Convolutional Neural Network Training the 3D convolutional neural network(CNN) is the second stage. The CNN we use in this project has one convolutional layer, one pooling layer, two linear layers, and finally a log softmax layer. After training the sparse autoencoder, we take the weights and biases of the encoder from trained model, and use them a 3D filter of a 3D convolutional layer of the 1-layer convolutional neural network. Figure 2 shows the architecture of the network. #### 5. Tools In this project, we used Nibabel for MRI image processing and PyTorch Neural Networks implementation.

This is the repository for the paper published in Medical Physics: "Synthetic CT generation from MRI using 3D transformer-based denoising diffusion model".

An MRI-pathology model (MRI-based Predicted Transformer for Prostate cancer (MRI-PTPCa)) was proposed to discover correlations between mp-MRI and tumor regressiveness of PCa and was further deployed for diagnosing non-PCa, PCa, non-CSPCa, CSPCa, and grading of GGG

This project takes MRI or CT scan raw *.dicom files as an input then it transform it to a visible picture in Hounsfield scale, then it builds a 3d model based on the slices data, the next step in the project is to export it to *.stl to be 3D printed into a real size plastic organ for educational and surgical purposes.

No description available

This code provides the Matlab implementation that detects the brain tumor region and also classify the tumor as benign and malignant. This code is implementation for the - A. Mathew and P. Anto, "Tumor detection and classification of MRI brain image using wavelet transform and SVM", 2017 International Conference on Signal Processing and Communication (ICSPC), 2017. Available: 10.1109/cspc.2017.8305810.

Medical image fusion is the process of combining two different modality images into a single image. The resultant image can help the physicians to extract features that may not be easily identifiable in an individual modality images. This paper aims to demonstrate an efficient method for detection of brain tumor from CT and MRI images of the brain, by applying image fusion, segmentation, feature extraction and classification. Initially, the source images are decomposed into low-level sub-band and high level sub-band by Discrete Wavelet Transform (DWT). The fused low level sub-band and high level sub-band are reconstructed to form the final fused image using Inverse Discrete Wavelet Transform (IDWT). Parameter analysis is done on the fused image. The fused image is then segmented using Otsu’s thresholding operation and the texture features are extracted forms the Grey Level Co-occurrence Matrix (GLCM) technique. Finally, the extracted feature is provided to Adaptive Neural Network (ANN) classifier to identify and predict the nature of the tumor. Further this proposed method gives an accuracy of 93.5% for 12 samples of MRI and CT images each.

Transforms for nifti images to use for data augmentation training models in pytorch on 3D MRI data.

A collection of deep learning models (CycleGAN, Pix2Pix, UNet) for medical image-to-image translation, with a focus on transforming 2D and 3D Ultrasound images to MRI.

Functionality for transforming dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) series.

Style Transfer using Generative Adversarial Networks (GAN) Project description- To ensure a better diagnosis of patients, doctors may need to look at multiple MRI scans. What if only one type of MRI needs to be done and others can be auto-generated? - Different MRIs are required for different abnormalities. A single type of MRI may not be sufficient for the diagnosis of an abnormality. Additional MRIs might enhance the accuracy of diagnosis, leading to better treatment of the patient. but Access to different imaging techniques is difficult and expensive. Moreover, doctors may advise getting one type of MRI to be done at a time, which makes it a time-consuming process. with the help of an exciting tool in the deep learning domain, which is known as Generative Adversarial Networks or GANs. - Generative Adversarial Networks (GANs) have been used for generating deepfakes, new fashion styles and high-resolution pictures from the low-resolution ones. - GANs can be used in the field of medical science, for instance, to create a different type of MRI from an existing one. a particular variant of GANs, called CycleGAN, is used to translate the style of one MRI scan into another, such as T1-weighted to T2-weighted or vice versa.

The DICOM 3D Medical Image Modeling (DMIM) project enhances patient care by transforming traditional 2D MRI, PET, and CT scan slices into comprehensive 3D models, making it easier for doctors and technicians to visualize and analyze medical images. 🩻✨

No description available

No description available

No description available

Interactive MRI image reconstruction using k-space visualization and Fourier Transform. Upload DICOM images and watch live step-by-step reconstruction in your browser!

Rhenium OS is a state-of-the-art, production-grade AI operating system engineered to transform diagnostic medical imaging across multiple modalities (MRI, CT, Ultrasound, X-ray).

Advanced Neural Rendering for Medical Imaging (HackMIT 2024 Healthcare Grand Prize): MindScape is a machine learning platform that transforms MRI scans into interactive 3D neural radiance fields (NeRFs) with integrated cancer detection capabilities.

My engineering thesis project: recognition of brain tumor Magnetic Resonance Images (MRI) using Wavelet Transform, Gray-Level Co-occurance Matrix (GLCM) and quadratic discriminant analysis (QDA). Data source: figshare.com/articles/brain_tumor_dataset/1512427

Wavelet-Based Contourlet Coding (WBCT) is an advanced image compression technique that combines wavelet transforms and contourlet filters to efficiently capture both global and local image features, providing high-quality compression for images like MRI scans.

The main aim of project is to classify MRI brain images as abnormal or normal images and also highlight tumor area. In our proposed system we are using transform learning in which VGG16 is used as feature extractor and Support Vector Machine (SVM) algorithm is used for classification .we are highlighting tumor region using segmentation method.

This Msc. project constitute the main part of my thesis at Imperial College London (2022-2023). This project has been supervised by A. Luati.

This project classifies tumor types in MRI scans using Vision Transformer (ViT), Swin Transformer, and ConvNext models. It includes data preprocessing, model training, and evaluation with ROC curves, AUC scores, and confusion matrices, comparing the effectiveness of each model in accurately identifying tumor types.

No description available

This study performs a comparative evaluation of Alzheimer’s disease detection using MRI brain scans, analyzing the performance of CNNs and Vision Transformer architectures such as ViT, DeepViT, and CaiT. DeepViT achieved the highest accuracy of 90.2%, highlighting the effectiveness of transformer-based models in medical imaging tasks.

This repository provides R code to convert a Multi-Regional Input-Output (MRIO) table into an Interregional Input-Output (IRIO) table for five Italian sub-regions, enabling detailed analysis of interregional trade flows.

Detection of Brain Tumor based on Moments of Local Binary Patterns and Discrete Wavelet Transform from MRI

Transform single MRI sequences into comprehensive multi-sequence scans using advanced GANs with Squeeze-Attention U-Net architecture.

An efficient brain tumor detection method, which can detect tumor and locate it in the brain MRI images. The detection and classification of MRI brain tumors are implemented using different wavelet transforms and SVM. Accurate and automated classification of MRI brain images is extremely important for medical analysis and interpretation.

MRI with compressive sensing speeds up scans by reconstructing high quality images from fewer data points, using mathematical algorithms to exploit image sparsity. This is achieved by leveraging the sparsity of medical images in certain transform domains. This makes MRI faster and more efficient.