RESUMO
Objective.Myocarditis poses a significant health risk, often precipitated by viral infections like coronavirus disease, and can lead to fatal cardiac complications. As a less invasive alternative to the standard diagnostic practice of endomyocardial biopsy, which is highly invasive and thus limited to severe cases, cardiac magnetic resonance (CMR) imaging offers a promising solution for detecting myocardial abnormalities.Approach.This study introduces a deep model called ELRL-MD that combines ensemble learning and reinforcement learning (RL) for effective myocarditis diagnosis from CMR images. The model begins with pre-training via the artificial bee colony (ABC) algorithm to enhance the starting point for learning. An array of convolutional neural networks (CNNs) then works in concert to extract and integrate features from CMR images for accurate diagnosis. Leveraging the Z-Alizadeh Sani myocarditis CMR dataset, the model employs RL to navigate the dataset's imbalance by conceptualizing diagnosis as a decision-making process.Main results.ELRL-DM demonstrates remarkable efficacy, surpassing other deep learning, conventional machine learning, and transfer learning models, achieving an F-measure of 88.2% and a geometric mean of 90.6%. Extensive experimentation helped pinpoint the optimal reward function settings and the perfect count of CNNs.Significance.The study addresses the primary technical challenge of inherent data imbalance in CMR imaging datasets and the risk of models converging on local optima due to suboptimal initial weight settings. Further analysis, leaving out ABC and RL components, confirmed their contributions to the model's overall performance, underscoring the effectiveness of addressing these critical technical challenges.
Assuntos
Aprendizado Profundo , Imageamento por Ressonância Magnética , Miocardite , Miocardite/diagnóstico por imagem , Humanos , Processamento de Imagem Assistida por Computador/métodos , Redes Neurais de ComputaçãoRESUMO
Aims: Brain diseases refer to intracranial tissue and organ inflammation, vascular diseases, tumors, degeneration, malformations, genetic diseases, immune diseases, nutritional and metabolic diseases, poisoning, trauma, parasitic diseases, etc. Taking Alzheimer's disease (AD) as an example, the number of patients dramatically increases in developed countries. By 2025, the number of elderly patients with AD aged 65 and over will reach 7.1 million, an increase of nearly 29% over the 5.5 million patients of the same age in 2018. Unless medical breakthroughs are made, AD patients may increase from 5.5 million to 13.8 million by 2050, almost three times the original. Researchers have focused on developing complex machine learning (ML) algorithms, i.e., convolutional neural networks (CNNs), containing millions of parameters. However, CNN models need many training samples. A small number of training samples in CNN models may lead to overfitting problems. With the continuous research of CNN, other networks have been proposed, such as randomized neural networks (RNNs). Schmidt neural network (SNN), random vector functional link (RVFL), and extreme learning machine (ELM) are three types of RNNs. Methods: We propose three novel models to classify brain diseases to cope with these problems. The proposed models are DenseNet-based SNN (DSNN), DenseNet-based RVFL (DRVFL), and DenseNet-based ELM (DELM). The backbone of the three proposed models is the pre-trained "customize" DenseNet. The modified DenseNet is fine-tuned on the empirical dataset. Finally, the last five layers of the fine-tuned DenseNet are substituted by SNN, ELM, and RVFL, respectively. Results: Overall, the DSNN gets the best performance among the three proposed models in classification performance. We evaluate the proposed DSNN by five-fold cross-validation. The accuracy, sensitivity, specificity, precision, and F1-score of the proposed DSNN on the test set are 98.46% ± 2.05%, 100.00% ± 0.00%, 85.00% ± 20.00%, 98.36% ± 2.17%, and 99.16% ± 1.11%, respectively. The proposed DSNN is compared with restricted DenseNet, spiking neural network, and other state-of-the-art methods. Finally, our model obtains the best results among all models. Conclusions: DSNN is an effective model for classifying brain diseases.
RESUMO
PURPOSE: To perform a review of the physical basis of DTI and DCE-MRI applied to Peripheral Nerves (PNs) evaluation with the aim of providing readers the main concepts and tools to acquire these types of sequences for PNs assessment. The potential added value of these advanced techniques for pre-and post-surgical PN assessment is also reviewed in diverse clinical scenarios. Finally, a brief introduction to the promising applications of Artificial Intelligence (AI) for PNs evaluation is presented. METHODS: We review the existing literature and analyze the latest evidence regarding DTI, DCE-MRI and AI for PNs assessment. This review is focused on a practical approach to these advanced sequences providing tips and tricks for implementing them into real clinical practice focused on imaging postprocessing and their current clinical applicability. A summary of the potential applications of AI algorithms for PNs assessment is also included. RESULTS: DTI, successfully used in central nervous system, can also be applied for PNs assessment. DCE-MRI can help evaluate PN's vascularization and integrity of Blood Nerve Barrier beyond the conventional gadolinium-enhanced MRI sequences approach. Both approaches have been tested for PN assessment including pre- and post-surgical evaluation of PNs and tumoral conditions. AI algorithms may help radiologists for PN detection, segmentation and characterization with promising initial results. CONCLUSION: DTI, DCE-MRI are feasible tools for the assessment of PN lesions. This manuscript emphasizes the technical adjustments necessary to acquire and post-process these images. AI algorithms can also be considered as an alternative and promising choice for PN evaluation with promising results.
Assuntos
Inteligência Artificial , Imageamento por Ressonância Magnética , Algoritmos , Humanos , Imageamento por Ressonância Magnética/métodos , Nervos Periféricos/diagnóstico por imagemRESUMO
As an important imaging modality, mammography is considered to be the global gold standard for early detection of breast cancer. Computer-Aided (CAD) systems have played a crucial role in facilitating quicker diagnostic procedures, which otherwise could take weeks if only radiologists were involved. In some of these CAD systems, breast pectoral segmentation is required for breast region partition from breast pectoral muscle for specific analysis tasks. Therefore, accurate and efficient breast pectoral muscle segmentation frameworks are in high demand. Here, we proposed a novel deep learning framework, which we code-named PeMNet, for breast pectoral muscle segmentation in mammography images. In the proposed PeMNet, we integrated a novel attention module called the Global Channel Attention Module (GCAM), which can effectively improve the segmentation performance of Deeplabv3+ using minimal parameter overheads. In GCAM, channel attention maps (CAMs) are first extracted by concatenating feature maps after paralleled global average pooling and global maximum pooling operation. CAMs are then refined and scaled up by multi-layer perceptron (MLP) for elementwise multiplication with CAMs in next feature level. By iteratively repeating this procedure, the global CAMs (GCAMs) are then formed and multiplied elementwise with final feature maps to lead to final segmentation. By doing so, CAMs in early stages of a deep convolution network can be effectively passed on to later stages of the network and therefore leads to better information usage. The experiments on a merged dataset derived from two datasets, INbreast and OPTIMAM, showed that PeMNet greatly outperformed state-of-the-art methods by achieving an IoU of 97.46%, global pixel accuracy of 99.48%, Dice similarity coefficient of 96.30%, and Jaccard of 93.33%, respectively.
RESUMO
Ductal carcinoma in situ (DCIS) is a pre-cancerous lesion in the ducts of the breast, and early diagnosis is crucial for optimal therapeutic intervention. Thermography imaging is a non-invasive imaging tool that can be utilized for detection of DCIS and although it has high accuracy (~ 88%), it is sensitivity can still be improved. Hence, we aimed to develop an automated artificial intelligence-based system for improved detection of DCIS in thermographs. This study proposed a novel artificial intelligence based system based on convolutional neural network (CNN) termed CNN-BDER on a multisource dataset containing 240 DCIS images and 240 healthy breast images. Based on CNN, batch normalization, dropout, exponential linear unit and rank-based weighted pooling were integrated, along with L-way data augmentation. Ten runs of tenfold cross validation were chosen to report the unbiased performances. Our proposed method achieved a sensitivity of 94.08 ± 1.22%, a specificity of 93.58 ± 1.49 and an accuracy of 93.83 ± 0.96. The proposed method gives superior performance than eight state-of-the-art approaches and manual diagnosis. The trained model could serve as a visual question answering system and improve diagnostic accuracy.
RESUMO
Predicting functional outcomes after surgery and early adjuvant treatment is difficult due to the complex, extended, interlocking brain networks that underpin cognition. The aim of this study was to test glioma functional interactions with the rest of the brain, thereby identifying the risk factors of cognitive recovery or deterioration. Seventeen patients with diffuse non-enhancing glioma (aged 22-56 years) were longitudinally MRI scanned and cognitively assessed before and after surgery and during a 12-month recovery period (55 MRI scans in total after exclusions). We initially found, and then replicated in an independent dataset, that the spatial correlation pattern between regional and global BOLD signals (also known as global signal topography) was associated with tumour occurrence. We then estimated the coupling between the BOLD signal from within the tumour and the signal extracted from different brain tissues. We observed that the normative global signal topography is reorganised in glioma patients during the recovery period. Moreover, we found that the BOLD signal within the tumour and lesioned brain was coupled with the global signal and that this coupling was associated with cognitive recovery. Nevertheless, patients did not show any apparent disruption of functional connectivity within canonical functional networks. Understanding how tumour infiltration and coupling are related to patients' recovery represents a major step forward in prognostic development.
RESUMO
Aims: Most blood diseases, such as chronic anemia, leukemia (commonly known as blood cancer), and hematopoietic dysfunction, are caused by environmental pollution, substandard decoration materials, radiation exposure, and long-term use certain drugs. Thus, it is imperative to classify the blood cell images. Most cell classification is based on the manual feature, machine learning classifier or the deep convolution network neural model. However, manual feature extraction is a very tedious process, and the results are usually unsatisfactory. On the other hand, the deep convolution neural network is usually composed of massive layers, and each layer has many parameters. Therefore, each deep convolution neural network needs a lot of time to get the results. Another problem is that medical data sets are relatively small, which may lead to overfitting problems. Methods: To address these problems, we propose seven models for the automatic classification of blood cells: BCARENet, BCR5RENet, BCMV2RENet, BCRRNet, BCRENet, BCRSNet, and BCNet. The BCNet model is the best model among the seven proposed models. The backbone model in our method is selected as the ResNet-18, which is pre-trained on the ImageNet set. To improve the performance of the proposed model, we replace the last four layers of the trained transferred ResNet-18 model with the three randomized neural networks (RNNs), which are RVFL, ELM, and SNN. The final outputs of our BCNet are generated by the ensemble of the predictions from the three randomized neural networks by the majority voting. We use four multi-classification indexes for the evaluation of our model. Results: The accuracy, average precision, average F1-score, and average recall are 96.78, 97.07, 96.78, and 96.77%, respectively. Conclusion: We offer the comparison of our model with state-of-the-art methods. The results of the proposed BCNet model are much better than other state-of-the-art methods.
RESUMO
BACKGROUND: Socioeconomic status (SES)-education, income level, and occupation-is associated with cardiovascular risk. OBJECTIVES: This study aimed to investigate the association between SES and subclinical atherosclerosis and the potential mechanisms involved. METHODS: SES, lifestyle habits (smoking, dietary patterns, physical activity, and hours of sleep), traditional risk factors, and subclinical atherosclerosis extent were prospectively assessed in 4,025 individuals aged 40 to 54 years without known cardiovascular disease enrolled in the PESA (Progression of Early Subclinical Atherosclerosis) study. After factors associated with atherosclerosis were identified, a multiple mediation model was created to quantify the effect of SES on subclinical atherosclerosis as explained by lifestyle behaviors. RESULTS: Although education level was significantly associated with the presence of atherosclerosis, no differences were found according to income level in this population. Participants with lower education presented with a higher risk of generalized atherosclerosis than those with higher education (odds ratio: 1.46; 95% confidence interval: 1.15 to 1.85; p = 0.002). Lifestyle behaviors associated with both education level and atherosclerosis extent were: smoking status, number of cigarettes/day, and dietary pattern, which explained 70.5% of the effect of SES on atherosclerosis. Of these, tobacco habit (smoking status 35% and number of cigarettes/day 32%) accounted for most of the explained differences between groups, whereas dietary pattern did not remain a significant mediator in the multiple mediation model. CONCLUSIONS: Despite the relative economic homogeneity of the cohort, lower education level is associated with increased subclinical atherosclerosis, mainly mediated by the higher and more frequent tobacco consumption. Smoking cessation programs are still needed, particularly in populations with lower education level.
Assuntos
Aterosclerose/epidemiologia , Modelos Teóricos , Adulto , Feminino , Humanos , Estilo de Vida , Masculino , Pessoa de Meia-Idade , Estudos Prospectivos , Medição de Risco , Fatores de Risco , Classe SocialRESUMO
This paper presents a computer-aided diagnosis technique for improving the accuracy of diagnosing the Alzheimer's type dementia. The proposed methodology is based on the calculation of the skewness for each m-by-m-by-m sliding block of the SPECT brain images. The center pixel in this m-by-m-by-m block is replaced by the skewness value to build a new 3-D brain image which is used for classification purposes. After that, voxels which present a Welch's t-statistic between classes, Normal and Alzheimer's images, higher (or lower) than a threshold are selected. The mean, standard deviation, skewness and kurtosis are calculated for these selected voxels and they are subjected as features to linear kernel based support vector machine classifier. The proposed methodology reaches accuracy higher than 99% in the classification task.