Neuroscience informatics最新文献

Precise detection of Alzheimer's disease (AD), especially at the early stages, i.e., early mild cognitive impairment (EMCI) and MCI, allows the physicians to promptly intervene to prevent the progression to advanced stages. However, identification of such stages using non-invasive brain imaging techniques like DWI, remains one of the most challenging tasks due to the subtle and mild changes in the brain structures of the subjects. Findings from previous studies suggested that topological organization alterations occur in the DTI-derived structural connectomes in MCI patients. Therefore, for improving diagnosis performance, we presented a connectome-based deep learning architecture based on BrainNet Convolutional neural network (CNN) model. The proposed model automatically extracts hidden topological features from structural networks using specially-designed convolutional filters. Experiments on 360 subjects, including 120 subjects with EMCI, 120 subjects with MCI and, 120 normal controls (NCs), with both T1-weighted MRI and DWI scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI), provided the highest binary classification accuracies of 0.96, 0.98, and 0.95 for NC/EMCI, NC/MCI and EMCI/MCI respectively.

In addition, we also investigated the effect of different atlas sizes and fiber descriptors as edge weights on the discriminative ability of the classification performance. Experimental results indicate that our approach exhibited superior performance to previous methods and performed effectively without any prior complex feature engineering and regardless the variability of imaging acquisition protocols and medical scanners.

Finally, we observed that DTI-based graph representation of brain regions connections preserve important but hidden connectivity pattern information to discriminate between clinical profiles, and our proposed approach could be easily extended to other neurodegenerative and neuropsychiatric diseases.

Introduction

Pretreatment positron emission tomography (PET) with 2-deoxy-2-[¹⁸F]fluoro-D-glucose (FDG) and magnetic resonance spectroscopy (MRS) may identify biomarkers for predicting remission (absence of depression). Yet, no such image-based biomarkers have achieved clinical validity. The purpose of this study was to identify biomarkers of remission using machine learning (ML) with pretreatment FDG-PET/MRS neuroimaging, to reduce patient suffering and economic burden from ineffective trials.

Methods

This study used simultaneous PET/MRS neuroimaging from a double-blind, placebo-controlled, randomized antidepressant trial on 60 participants with major depressive disorder (MDD) before initiating treatment. After eight weeks of treatment, those with ≤7 on 17-item Hamilton Depression Rating Scale were designated a priori as remitters (free of depression, 37%). Metabolic rate of glucose uptake (metabolism) from 22 brain regions were acquired from PET. Concentrations (mM) of glutamine and glutamate and gamma-aminobutyric acid (GABA) in anterior cingulate cortex were quantified from MRS. The data were randomly split into 67% train and cross-validation ($n=40$), and 33% test ($n=20$) sets. The imaging features, along with age, sex, handedness, and treatment assignment (selective serotonin reuptake inhibitor or SSRI vs. placebo) were entered into the eXtreme Gradient Boosting (XGBoost) classifier for training.

Results

In test data, the model showed 62% sensitivity, 92% specificity, and 77% weighted accuracy. Pretreatment metabolism of left hippocampus from PET was the most predictive of remission.

Conclusions

The pretreatment neuroimaging takes around 60 minutes but has potential to prevent weeks of failed treatment trials. This study effectively addresses common issues for neuroimaging analysis, such as small sample size, high dimensionality, and class imbalance.

Skin cancer is one of the most prevalent and deadly types of cancer. Dermatologists diagnose this disease primarily visually. Multiclass skin cancer classification is challenging due to the fine-grained variability in the appearance of its various diagnostic categories. On the other hand, recent studies have demonstrated that convolutional neural networks outperform dermatologists in multiclass skin cancer classification. We developed a preprocessing image pipeline for this work. We removed hairs from the images, augmented the dataset, and resized the imageries to meet the requirements of each model. By performing transfer learning on pre-trained ImageNet weights and fine-tuning the Convolutional Neural Networks, we trained the EfficientNets B0-B7 on the HAM10000 dataset. We evaluated the performance of all EfficientNet variants on this imbalanced multiclass classification task using metrics such as Precision, Recall, Accuracy, F1 Score, and Confusion Matrices to determine the effect of transfer learning with fine-tuning. This article presents the classification scores for each class as Confusion Matrices for all eight models. Our best model, the EfficientNet B4, achieved an F1 Score of 87 percent and a Top-1 Accuracy of 87.91 percent. We evaluated EfficientNet classifiers using metrics that take the high-class imbalance into account. Our findings indicate that increased model complexity does not always imply improved classification performance. The best performance arose with intermediate complexity models, such as EfficientNet B4 and B5. The high classification scores resulted from many factors such as resolution scaling, data enhancement, noise removal, successful transfer learning of ImageNet weights, and fine-tuning [70], [71], [72]. Another discovery was that certain classes of skin cancer worked better at generalization than others using Confusion Matrices.

Background

Modern day's society is engaged in commitment-based and time-bound jobs. This invites tension and mental depression among many people who are not able to cope up with this type of working environment. Cases of mental depression are increasing day by day all over the world. Recently, the onset of the COVID-19 pandemic has added further fuel to the fire. In many countries, the ratio between patients with mental depression and psychiatrists or psychologists is remarkably poor. Under such a situation, the design, and development of an expert system by exploiting the hidden power of various deep learning (DL) and machine learning (ML) techniques can solve the problem up to a greater extent.

Methodology

Each deep learning and machine learning technique has got its advantages and disadvantages to handle different classification problems. In this article four neural network-based deep learning architectures namely MLP, CNN, RNN, RNN with LSTM, and two Supervised Machine Learning Techniques such as SVM and LR are implemented to investigate and compare their suitability to track the mental depression from EEG Data.

Result

Among Neural Network-Based Deep Learning techniques RNN model has achieved the highest accuracy with 97.50% in Training Set and 96.50% in the Testing set respectively. It has been followed with RNN with LSTM model when there were 40% data in the Testing Set. Whereas both the Supervised Machine Learning Models namely SVM and LR have outperformed with 100.00% accuracies in Training Phase and approximately 97.25% accuracies in Testing Phase respectively.

Conclusion

This investigation and comparison-oriented study establish the suitability of RNN, RNN with LSTM, SVM and LR model to track mental depression from EEG data. This type of comparative research using Machine Learning and Deep learning architectures must be framed out on this topic to finalize the design and development of an expert system for the automatic detection of depression from EEG data.

Parkinson's disease (PD) is a heterogeneous neurodegenerative disorder associated with several motor and non-motor dysfunctions. The wide variety of clinical features often leads to divergent symptom progressions. Most PD studies have attempted subgrouping based on clinical features to help understand the disease etiology and thereby contribute toward specific treatment. However, clinical symptoms have proven to be overlapping, arbitrary, and non-reliable in several cases, often biasing the deciphered subgroups. Moreover, the prodromal phase complicates diagnosis and subgrouping as it is characterized by limited clinical symptom expression. Hence, recent studies have used data-driven machine learning and deep learning methods to data-mine the heterogeneity and obtain subgroups. Structural Magnetic Resonance Imaging (sMRI) is a non-invasive approach for visualization and analysis of anatomical tissue properties of brain. It has enabled the detection of brain abnormalities and is a potential modality for subgrouping.

This review article starts with a comprehensive discussion of clinical symptoms-based and data-driven structural neuroimaging-based subgrouping approaches in PD. Secondly, we summarize the work done in brain connectivity studies using structural MRI for PD. We give an overview of mathematical definitions, connectivity metrics, brain connectivity software, and widespread network atlases. Finally, we discuss the inherent challenges and give practical suggestions on selecting methods that could be attempted for subgrouping and connectivity analysis using structural MRI data for future Parkinson's research.

Cerebral temperature is one of the key indicators of fever, trauma, and physical activity. It has been reported that the temperature of the healthy brain is up to 2 °C higher than the core body temperature. The main methods to monitor brain temperature include infrared spectroscopy, radiometry, and acoustic thermometry. While these methods are useful, they are not very effective when portability is desired, the temperature needs to be monitored for a longer period, or localized monitoring is required. This paper presents a short review of invasive and non-invasive brain temperature monitoring sensors and tools. We discuss the type of temperature sensors that can be integrated with probes. Furthermore, implantable and bioresorbable sensors are briefly mentioned. Biocompatibility and invasiveness of the sensors in terms of their functional materials, encapsulation, and size are highlighted.

The novel Coronavirus, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) spread all over the world, causing a dramatic shift in circumstances that resulted in a massive pandemic, affecting the world's well-being and stability. It is an RNA virus that can infect both humans as well as animals. Diagnosis of the virus as soon as possible could contain and avoid a serious COVID-19 outbreak. Current pharmaceutical techniques and diagnostic methods tests such as Reverse Transcription-Polymerase Chain Reaction (RT-PCR) and Serology tests are time-consuming, expensive, and require a well-equipped laboratory for analysis, making them restrictive and inaccessible to everyone. Deep Learning has grown in popularity in recent years, and it now plays a crucial role in Image Classification, which also involves Medical Imaging. Using chest CT scans, this study explores the problem statement automation of differentiating COVID-19 contaminated individuals from healthy individuals. Convolutional Neural Networks (CNNs) can be trained to detect patterns in computed tomography scans (CT scans). Hence, different CNN models were used in the current study to identify variations in chest CT scans, with accuracies ranging from 91% to 98%. The Multiclass Classification method is used to build these architectures. This study also proposes a new approach for classifying CT images that use two binary classifications combined to work together, achieving 98.38% accuracy. All of these architectures' performances are compared using different classification metrics.

Introduction

In modern days, checking the huge number of MRI (magnetic resonance imaging) images and finding a brain tumour manually by a human is a very tedious and inaccurate task. It can affect the proper medical treatment of the patient. Again, it can be a hugely time-consuming task as it involves a huge number of image datasets. There is a good similarity between normal tissue and brain tumour cells in appearance, so segmentation of tumour regions become a difficult task to do. So there is an essentiality for a highly accurate automatic tumour detection method.

Method

In this paper, we proposed an algorithm to segment brain tumours from 2D Magnetic Resonance brain Images (MRI) by a convolutional neural network which is followed by traditional classifiers and deep learning methods. We have taken various MRI images with diverse Tumour sizes, locations, shapes, and different image intensities to train the model well. Furthermore, we have applied SVM classifier and other activation algorithms (softmax, RMSProp, sigmoid, etc) to cross-check our work. We implement our proposed method using “TensorFlow” and “Keras” in “Python” as it is an efficient programming language to perform fast work.

Result

In our work, CNN gained an accuracy of 99.74%, which is better than the state of the result obtained so far.

Conclusion

Our CNN based model will help the doctors to detect brain tumours in MRI images accurately, so that the speed in treatment will increase a lot.

Purpose

Different imaging sequences (T1 etc.) depict different aspects of a brain tumor. As clinical MRI examinations of the brain might be terminated prematurely, not all sequences may be acquired, decreasing the performance of automated tumor segmentation. We attempt to optimize the order of sequences, to maximize information gain in case of incomplete examination.

Methods

For segmentation we used the winner algorithm of the Brain Tumor Segmentation challenge 2018, trained on the BraTS 2020 dataset, with the objective to segment necrotic core, peritumoral edema, and enhancing tumor. We compared the segmentation performance for all combinations of sequences, using the Dice score (DS) as the primary metric. We compare the results with those which would be obtained by attempting to follow the consensus recommendations for brain tumor imaging [T1, FLAIR, T2, T1CE].

Results

The average segmentation accuracy varies between 0.476 for T1 only and 0.751 for the full set of sequences. T1CE has a high information content, even regarding peritumoral edema and information of T2 and FLAIR were highly redundant. The optimal order of sequences appears to be [T1, T2, T1CE, FLAIR]. Comparing segmentation accuracy after each fully acquired sequence, the first sequence (T1) is the same for both, DS for [T1, T2] (proposed) is 6.2% higher than [T1, FLAIR] (aborted recommendations), and [T1, T2, T1CE] (proposed) is 34.8% higher than [T1, FLAIR, T2] (aborted recommendations).

Conclusion

For the purpose of optimal deep-learning-based segmentation purposes in potentially incomplete MRI examinations, the T1CE sequence should be acquired as early as possible.