Brain Images Research Articles

Brain morphometric changes in children born as small for gestational age without catch up growth.

Most infants born as small for gestational age (SGA) demonstrate catch up growth by 2-4 years, but some fail to do so. This failure is associated with several health risks, including neuropsychological development issues. However, data on the morphological characteristics of the brains of infants born as SGA without achieving catch up growth are lacking. This study aims to determine the structural aspects of the brains of children born as SGA without catch up growth. We conducted voxel- and surface-based morphometric analyses of 1.5-T T1-weighted brain images scanned from eight infants born as SGA who could not achieve catch up growth by 3 years and sixteen individuals with idiopathic short stature (ISS) to exclude body size effects. Growth hormone (GH) secretion stimulation tests were used to rule out GH deficiency in all SGA and ISS cases. The magnetic resonance imaging data were assessed using Levene's test for equality of variances and a two-tailed unpaired t-test for equality of means. The Benjamini-Hochberg procedure was used to apply discovery rate correction for multiple comparisons. Morphometric analyses of both t-statical map and surface-based analyses using general linear multiple analysis determined decreased left insula thickness and volume in SGA without catch up growth compared with ISS. The brain scans of patients with SGA who lack catch up growth indicated distinct morphological disparities when compared to those with ISS. The discernible features of brain morphology observed in patients born as SGA without catch up growth may improve understanding of the association of SGA without catch up growth with both intellectual and psychological outcomes.

LYNSU: automated 3D neuropil segmentation of fluorescent images for Drosophila brains.

The brain atlas, which provides information about the distribution of genes, proteins, neurons, or anatomical regions, plays a crucial role in contemporary neuroscience research. To analyze the spatial distribution of those substances based on images from different brain samples, we often need to warp and register individual brain images to a standard brain template. However, the process of warping and registration may lead to spatial errors, thereby severely reducing the accuracy of the analysis. To address this issue, we develop an automated method for segmenting neuropils in the Drosophila brain for fluorescence images from the FlyCircuit database. This technique allows future brain atlas studies to be conducted accurately at the individual level without warping and aligning to a standard brain template. Our method, LYNSU (Locating by YOLO and Segmenting by U-Net), consists of two stages. In the first stage, we use the YOLOv7 model to quickly locate neuropils and rapidly extract small-scale 3D images as input for the second stage model. This stage achieves a 99.4% accuracy rate in neuropil localization. In the second stage, we employ the 3D U-Net model to segment neuropils. LYNSU can achieve high accuracy in segmentation using a small training set consisting of images from merely 16 brains. We demonstrate LYNSU on six distinct neuropils or structures, achieving a high segmentation accuracy comparable to professional manual annotations with a 3D Intersection-over-Union (IoU) reaching up to 0.869. Our method takes only about 7 s to segment a neuropil while achieving a similar level of performance as the human annotators. To demonstrate a use case of LYNSU, we applied it to all female Drosophila brains from the FlyCircuit database to investigate the asymmetry of the mushroom bodies (MBs), the learning center of fruit flies. We used LYNSU to segment bilateral MBs and compare the volumes between left and right for each individual. Notably, of 8,703 valid brain samples, 10.14% showed bilateral volume differences that exceeded 10%. The study demonstrated the potential of the proposed method in high-throughput anatomical analysis and connectomics construction of the Drosophila brain.

Multi-Modal Diagnosis of Alzheimer's Disease using Interpretable Graph Convolutional Networks.

The interconnection between brain regions in neurological disease encodes vital information for the advancement of biomarkers and diagnostics. Although graph convolutional networks are widely applied for discovering brain connection patterns that point to disease conditions, the potential of connection patterns that arise from multiple imaging modalities has yet to be fully realized. In this paper, we propose a multi-modal sparse interpretable GCN framework (SGCN) for the detection of Alzheimer's disease (AD) and its prodromal stage, known as mild cognitive impairment (MCI). In our experimentation, SGCN learned the sparse regional importance probability to find signature regions of interest (ROIs), and the connective importance probability to reveal disease-specific brain network connections. We evaluated SGCN on the Alzheimer's Disease Neuroimaging Initiative database with multi-modal brain images and demonstrated that the ROI features learned by SGCN were effective for enhancing AD status identification. The identified abnormalities were significantly correlated with AD-related clinical symptoms. We further interpreted the identified brain dysfunctions at the level of large-scale neural systems and sex-related connectivity abnormalities in AD/MCI. The salient ROIs and the prominent brain connectivity abnormalities interpreted by SGCN are considerably important for developing novel biomarkers. These findings contribute to a better understanding of the network-based disorder via multi-modal diagnosis and offer the potential for precision diagnostics. The source code is available at https://github.com/Houliang-Zhou/SGCN.

Exploring approaches to tackle cross-domain challenges in brain medical image segmentation: a systematic review.

Brain medical image segmentation is a critical task in medical image processing, playing a significant role in the prediction and diagnosis of diseases such as stroke, Alzheimer's disease, and brain tumors. However, substantial distribution discrepancies among datasets from different sources arise due to the large inter-site discrepancy among different scanners, imaging protocols, and populations. This leads to cross-domain problems in practical applications. In recent years, numerous studies have been conducted to address the cross-domain problem in brain image segmentation. This review adheres to the standards of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) for data processing and analysis. We retrieved relevant papers from PubMed, Web of Science, and IEEE databases from January 2018 to December 2023, extracting information about the medical domain, imaging modalities, methods for addressing cross-domain issues, experimental designs, and datasets from the selected papers. Moreover, we compared the performance of methods in stroke lesion segmentation, white matter segmentation and brain tumor segmentation. A total of 71 studies were included and analyzed in this review. The methods for tackling the cross-domain problem include Transfer Learning, Normalization, Unsupervised Learning, Transformer models, and Convolutional Neural Networks (CNNs). On the ATLAS dataset, domain-adaptive methods showed an overall improvement of ~3 percent in stroke lesion segmentation tasks compared to non-adaptive methods. However, given the diversity of datasets and experimental methodologies in current studies based on the methods for white matter segmentation tasks in MICCAI 2017 and those for brain tumor segmentation tasks in BraTS, it is challenging to intuitively compare the strengths and weaknesses of these methods. Although various techniques have been applied to address the cross-domain problem in brain image segmentation, there is currently a lack of unified dataset collections and experimental standards. For instance, many studies are still based on n-fold cross-validation, while methods directly based on cross-validation across sites or datasets are relatively scarce. Furthermore, due to the diverse types of medical images in the field of brain segmentation, it is not straightforward to make simple and intuitive comparisons of performance. These challenges need to be addressed in future research.

Improved Global U-Net applied for multi-modal brain tumor fuzzy segmentation

In this paper, we extended our work from Global U-Net combined with fuzzy amalgamation of Inception Model and Improved Kernel Variation for MRI Brain Image Segmentation [1] which was meant for single modality MRI images only to a brain tumor fuzzy segmentation. Many CNNs gives state of art results for a particular type of images. However, they cannot achieve the same result for the images captured from different imaging techniques. We experimented the Global U-Net model for MRI images earlier and this time we intended to make it applicable for other type of images too using the concept of fuzzy segmentation. The major concern was to overcome the limitations of single modality system that is not all the kernels of U-Net are capable of generating clear feature vectors for different image modalities. The result generated was satisfactory and we would further extend it for colored images.

Accelerated multiple-quantum-filtered sodium magnetic resonance imaging using compressed sensing at 7 T

PurposeMultiple-quantum-filtered (MQF) sodium magnetic resonance imaging (MRI), such as enhanced single-quantum and triple-quantum-filtered imaging of 23Na (eSISTINA), enables images to be weighted towards restricted sodium, a promising biomarker in clinical practice, but often suffers from clinically infeasible acquisition times and low image quality. This study aims to mitigate the above limitation by implementing a novel eSISTINA sequence at 7 T with the application of compressed sensing (CS) to accelerate eSISTINA acquisitions without a noticeable loss of information. MethodsA novel eSISTINA sequence with a 3D spiral-based sampling scheme was implemented at 7 T for the application of CS. Fully sampled datasets were obtained from one phantom and ten healthy subjects, and were then retrospectively undersampled by various undersampling factors. CS undersampled reconstructions were compared to fully sampled and undersampled nonuniform fast Fourier transform (NUFFT) reconstructions. Reconstruction performance was evaluated based on structural similarity (SSIM), signal-to-noise ratio (SNR), weightings towards total and compartmental sodium, and in vivo quantitative estimates. ResultsCS-based phantom and in vivo images have less noise and better structural delineation while maintaining the weightings towards total, non-restricted (predominantly extracellular), and restricted (primarily intracellular) sodium. CS generally outperforms NUFFT with a higher SNR and a better SSIM, except for the SSIM in TQ brain images, which is likely due to substantial noise contamination. CS enables in vivo quantitative estimates with <15% errors at an undersampling factor of up to two. ConclusionsSuccessful implementation of an eSISTINA sequence with an incoherent sampling scheme at 7 T was demonstrated. CS can accelerate eSISTINA by up to twofold at 7 T with reduced noise levels compared to NUFFT, while maintaining major structural information, reasonable weightings towards total and compartmental sodium, and relatively reliable in vivo quantification. The associated reduction in acquisition time has the potential to facilitate the clinical applicability of MQF sodium MRI.

Synthesizing brain images using deep learning

Synthesizing brain images using deep learning

Deep Neural Network-based Methods for Brain Image De-noising: A Short Comparison

Deep Neural Network-based Methods for Brain Image De-noising: A Short Comparison

Brain image enhancement and segmentation using anatomically constrained neural networks

Brain image enhancement and segmentation using anatomically constrained neural networks

Generative AI Enables Synthesizing Cross-Modality Brain Image via Multi-Level-Latent Representation Learning

Generative AI Enables Synthesizing Cross-Modality Brain Image via Multi-Level-Latent Representation Learning

Brain cancer analysis using deep learning architecture on MRI brain image

Brain cancer analysis using deep learning architecture on MRI brain image

Automated Brain Tumor Detection and Classification in MRI Images: A Hybrid Image Processing Techniques

Due to the complex structure of brain images, accurately detecting and segmenting brain tumors with Magnetic Resonance Imaging (MRI) is a difficult process. This paper suggests an automated brain tumor identification and segmentation approach employing hybrid salient segmentation with K-Means clustering and hybrid CLEACH-median filter algorithm on MRI images. The proposed method enhances the contrast and detail of MRI images using a hybrid CLEACH-median filter algorithm, and segments the most important features of the images using a hybrid salient segmentation method with K-Means clustering. The proposed method includes a stages classification step to determine the stage of the brain tumor. The findings show that the suggested approach outperformed existing methods in terms of efficiency and accuracy for both detecting and segmenting brain tumors. The suggested technique can be a useful tool for automating the detection and segmentation of brain tumors, which will help radiologists and physicians make quicker and more accurate diagnosis.

Super-Resolution Imaging of Tau Proteins in Isolated and Immobilized Brain Synaptosomes.

Tau protein has important physiological functions at both presynaptic and postsynaptic terminals. Pathological tau species are also associated with synaptic dysfunctions in several neurodegenerative disorders, especially Alzheimer's disease. To understand tau distribution inside synaptic compartments, super-resolution imaging is required. Here, we describe a facile protocol to immobilize and image brain synaptosomes without aggregation artefacts, by substituting the standard fixative paraformaldehyde with ethylene glycol bis(succinimidyl succinate) (EGS). Super-resolution imaging of tau proteins is achieved through three-color direct stochastic optical reconstruction microscopy (dSTORM). Tau protein is found to colocalize with synaptic vesicles as well as postsynaptic densities.

EVALUATION OF U-Net AND ResNet ARCHITECTURES FOR BIOMEDICAL IMAGE SEGMENTATION

Medical professionals need methods that provide reliable information in diagnosing and monitoring neurological diseases. Among such methods, studies based on medical image analysis are essential among the active research topics in this field. Tumor segmentation is a popular area, especially with magnetic resonance imaging (MRI). Early diagnosis of tumours plays an essential role in the treatment process. This situation also increases the survival rate of the patients. Manually segmenting a tumour from MR images is a difficult and time-consuming task within the anatomical knowledge of medical professionals. This has necessitated the need for automatic segmentation methods. Convolutional neural networks (CNN), one of the deep learning methods that provide the most advanced results in the field of tumour segmentation, play an important role. This study, tumor segmentation was performed from brain and heart MR images using CNN-based U-Net and ResNet50 deep network architectures. In the segmentation process, their performance was tested using Dice, Sensitivity, PPV and Jaccard metrics. High performance levels were sequentially achieved using the U-Net network architecture on brain images, with success rates of approximately 98.47%, 98.1%, 98.85%, and 96.07%

Brain tumor classification by CNN

In this paper, we proposed a disease classification method for brain images using CNN. The research dataset collects images of normal, blastoma, meningioma, adenoma, and glioblastoma by brain tumor disease from journals such as NEJM and AuntMinnie and files them in Exam_Brain.Zip. As a result of CNN processing the images for each file folder in the Exam_Brain.zip file 10 times, the AUC was found to be 0.8825. As a result of the experiment, the brain disease classification accuracy of brain magnetic resonance imaging was found to be 88.25%. These results indicate that the results of this study can be used to classify other diseases once the data set is established, and can also be used to classify objects in other industries.This study has the following limitations. A challenge in automatically classifying diseases is the vastness and integrity of the data sets. If the data is not prepared, the accuracy of classification will be problematic. Additionally, in the medical field, a doctor's decision-making (diagnosis) is not limited to image data but is made based on comprehensive data, and there is a limitation in that medical decisions cannot be made by machines without the intervention of a doctor. Determining the decision threshold in AI decision-making problems is also an important limitation. This is because setting reference points for sensitivity and specificity is not the domain of AI experts, but the domain of experts in the field. Therefore, field medical staff must participate in the process of developing AI in medical settings. Future research tasks include developing applications that improve performance by developing the input stage of the program, and continuing research in connection with medical big data servers.

Detection and classification of brain tumor using hybrid deep learning models

Accurately classifying brain tumor types is critical for timely diagnosis and potentially saving lives. Magnetic Resonance Imaging (MRI) is a widely used non-invasive method for obtaining high-contrast grayscale brain images, primarily for tumor diagnosis. The application of Convolutional Neural Networks (CNNs) in deep learning has revolutionized diagnostic systems, leading to significant advancements in medical imaging interpretation. In this study, we employ a transfer learning-based fine-tuning approach using EfficientNets to classify brain tumors into three categories: glioma, meningioma, and pituitary tumors. We utilize the publicly accessible CE-MRI Figshare dataset to fine-tune five pre-trained models from the EfficientNets family, ranging from EfficientNetB0 to EfficientNetB4. Our approach involves a two-step process to refine the pre-trained EfficientNet model. First, we initialize the model with weights from the ImageNet dataset. Then, we add additional layers, including top layers and a fully connected layer, to enable tumor classification. We conduct various tests to assess the robustness of our fine-tuned EfficientNets in comparison to other pre-trained models. Additionally, we analyze the impact of data augmentation on the model's test accuracy. To gain insights into the model's decision-making, we employ Grad-CAM visualization to examine the attention maps generated by the most optimal model, effectively highlighting tumor locations within brain images. Our results reveal that using EfficientNetB2 as the underlying framework yields significant performance improvements. Specifically, the overall test accuracy, precision, recall, and F1-score were found to be 99.06%, 98.73%, 99.13%, and 98.79%, respectively.

Automated identification of protein expression intensity and classification of protein cellular locations in mouse brain regions from immunofluorescence images.

Knowledge of protein expression in mammalian brains at regional and cellular levels can facilitate understanding of protein functions and associated diseases. As the mouse brain is a typical mammalian brain considering cell type and structure, several studies have been conducted to analyze protein expression in mouse brains. However, labeling protein expression using biotechnology is costly and time-consuming. Therefore, automated models that can accurately recognize protein expression are needed. Here, we constructed machine learning models to automatically annotate the protein expression intensity and cellular location in different mouse brain regions from immunofluorescence images. The brain regions and sub-regions were segmented through learning image features using an autoencoder and then performing K-means clustering and registration to align with the anatomical references. The protein expression intensities for those segmented structures were computed on the basis of the statistics of the image pixels, and patch-based weakly supervised methods and multi-instance learning were used to classify the cellular locations. Results demonstrated that the models achieved high accuracy in the expression intensity estimation, and the F1 score of the cellular location prediction was 74.5%. This work established an automated pipeline for analyzing mouse brain images and provided a foundation for further study of protein expression and functions.

SMILE: Siamese Multi-scale Interactive-representation LEarning for Hierarchical Diffeomorphic Deformable image registration

Deformable medical image registration plays an important role in many clinical applications. It aims to find a dense deformation field to establish point-wise correspondences between a pair of fixed and moving images. Recently, unsupervised deep learning-based registration methods have drawn more and more attention because of fast inference at testing stage. Despite remarkable progress, existing deep learning-based methods suffer from several limitations including: (a) they often overlook the explicit modeling of feature correspondences due to limited receptive fields; (b) the performance on image pairs with large spatial displacements is still limited since the dense deformation field is regressed from features learned by local convolutions; and (c) desirable properties, including topology-preservation and the invertibility of transformation, are often ignored. To address above limitations, we propose a novel Convolutional Neural Network (CNN) consisting of a Siamese Multi-scale Interactive-representation LEarning (SMILE) encoder and a Hierarchical Diffeomorphic Deformation (HDD) decoder. Specifically, the SMILE encoder aims for effective feature representation learning and spatial correspondence establishing while the HDD decoder seeks to regress the dense deformation field in a coarse-to-fine manner. We additionally propose a novel Local Invertible Loss (LIL) to encourage topology-preservation and local invertibility of the regressed transformation while keeping high registration accuracy. Extensive experiments conducted on two publicly available brain image datasets demonstrate the superiority of our method over the state-of-the-art (SOTA) approaches. Specifically, on the Neurite-OASIS dataset, our method achieved an average DSC of 0.815 and an average ASSD of 0.633 mm.

An interactive image segmentation method for the anatomical structures of the main olfactory bulb with micro-level resolution.

The main olfactory bulb is the key element of the olfactory pathway of rodents. To precisely dissect the neural pathway in the main olfactory bulb (MOB), it is necessary to construct the three-dimensional morphologies of the anatomical structures within it with micro-level resolution. However, the construction remains challenging due to the complicated shape of the anatomical structures in the main olfactory bulb and the high resolution of micro-optical images. To address these issues, we propose an interactive volume image segmentation method with micro-level resolution in the horizontal and axial direction. Firstly, we obtain the initial location of the anatomical structures by manual annotation and design a patch-based neural network to learn the complex texture feature of the anatomical structures. Then we randomly sample some patches to predict by the trained network and perform an annotation reconstruction based on intensity calculation to get the final location results of the anatomical structures. Our experiments were conducted using Nissl-stained brain images acquired by the Micro-optical sectioning tomography (MOST) system. Our method achieved a mean dice similarity coefficient (DSC) of 81.8% and obtain the best segmentation performance. At the same time, the experiment shows the three-dimensional morphology reconstruction results of the anatomical structures in the main olfactory bulb are smooth and consistent with their natural shapes, which addresses the possibility of constructing three-dimensional morphologies of the anatomical structures in the whole brain.

Automated deep bottleneck residual 82-layered architecture with Bayesian optimization for the classification of brain and common maternal fetal ultrasound planes.

Despite a worldwide decline in maternal mortality over the past two decades, a significant gap persists between low- and high-income countries, with 94% of maternal mortality concentrated in low and middle-income nations. Ultrasound serves as a prevalent diagnostic tool in prenatal care for monitoring fetal growth and development. Nevertheless, acquiring standard fetal ultrasound planes with accurate anatomical structures proves challenging and time-intensive, even for skilled sonographers. Therefore, for determining common maternal fetuses from ultrasound images, an automated computer-aided diagnostic (CAD) system is required. A new residual bottleneck mechanism-based deep learning architecture has been proposed that includes 82 layers deep. The proposed architecture has added three residual blocks, each including two highway paths and one skip connection. In addition, a convolutional layer has been added of size 3 × 3 before each residual block. In the training process, several hyper parameters have been initialized using Bayesian optimization (BO) rather than manual initialization. Deep features are extracted from the average pooling layer and performed the classification. In the classification process, an increase occurred in the computational time; therefore, we proposed an improved search-based moth flame optimization algorithm for optimal feature selection. The data is then classified using neural network classifiers based on the selected features. The experimental phase involved the analysis of ultrasound images, specifically focusing on fetal brain and common maternal fetal images. The proposed method achieved 78.5% and 79.4% accuracy for brain fetal planes and common maternal fetal planes. Comparison with several pre-trained neural nets and state-of-the-art (SOTA) optimization algorithms shows improved accuracy.