RESUMEN
The study proposes a novel machine learning (ML) paradigm for cardiovascular disease (CVD) detection in individuals at medium to high cardiovascular risk using data from a Greek cohort of 542 individuals with rheumatoid arthritis, or diabetes mellitus, and/or arterial hypertension, using conventional or office-based, laboratory-based blood biomarkers and carotid/femoral ultrasound image-based phenotypes. Two kinds of data (CVD risk factors and presence of CVD-defined as stroke, or myocardial infarction, or coronary artery syndrome, or peripheral artery disease, or coronary heart disease) as ground truth, were collected at two-time points: (i) at visit 1 and (ii) at visit 2 after 3 years. The CVD risk factors were divided into three clusters (conventional or office-based, laboratory-based blood biomarkers, carotid ultrasound image-based phenotypes) to study their effect on the ML classifiers. Three kinds of ML classifiers (Random Forest, Support Vector Machine, and Linear Discriminant Analysis) were applied in a two-fold cross-validation framework using the data augmented by synthetic minority over-sampling technique (SMOTE) strategy. The performance of the ML classifiers was recorded. In this cohort with overall 46 CVD risk factors (covariates) implemented in an online cardiovascular framework, that requires calculation time less than 1 s per patient, a mean accuracy and area-under-the-curve (AUC) of 98.40% and 0.98 (p < 0.0001) for CVD presence detection at visit 1, and 98.39% and 0.98 (p < 0.0001) at visit 2, respectively. The performance of the cardiovascular framework was significantly better than the classical CVD risk score. The ML paradigm proved to be powerful for CVD prediction in individuals at medium to high cardiovascular risk.
Asunto(s)
Artritis Reumatoide/complicaciones , Enfermedades Cardiovasculares/diagnóstico , Aprendizaje Automático , Placa Aterosclerótica/diagnóstico por imagen , Arterias Carótidas/diagnóstico por imagen , Estudios Transversales , Femenino , Arteria Femoral/diagnóstico por imagen , Factores de Riesgo de Enfermedad Cardiaca , Humanos , Masculino , Proyectos Piloto , Reproducibilidad de los ResultadosRESUMEN
Computer Tomography (CT) is currently being adapted for visualization of COVID-19 lung damage. Manual classification and characterization of COVID-19 may be biased depending on the expert's opinion. Artificial Intelligence has recently penetrated COVID-19, especially deep learning paradigms. There are nine kinds of classification systems in this study, namely one deep learning-based CNN, five kinds of transfer learning (TL) systems namely VGG16, DenseNet121, DenseNet169, DenseNet201 and MobileNet, three kinds of machine-learning (ML) systems, namely artificial neural network (ANN), decision tree (DT), and random forest (RF) that have been designed for classification of COVID-19 segmented CT lung against Controls. Three kinds of characterization systems were developed namely (a) Block imaging for COVID-19 severity index (CSI); (b) Bispectrum analysis; and (c) Block Entropy. A cohort of Italian patients with 30 controls (990 slices) and 30 COVID-19 patients (705 slices) was used to test the performance of three types of classifiers. Using K10 protocol (90% training and 10% testing), the best accuracy and AUC was for DCNN and RF pairs were 99.41 ± 5.12%, 0.991 (p < 0.0001), and 99.41 ± 0.62%, 0.988 (p < 0.0001), respectively, followed by other ML and TL classifiers. We show that diagnostics odds ratio (DOR) was higher for DL compared to ML, and both, Bispecturm and Block Entropy shows higher values for COVID-19 patients. CSI shows an association with Ground Glass Opacities (0.9146, p < 0.0001). Our hypothesis holds true that deep learning shows superior performance compared to machine learning models. Block imaging is a powerful novel approach for pinpointing COVID-19 severity and is clinically validated.
Asunto(s)
Inteligencia Artificial/normas , COVID-19/diagnóstico por imagen , Pulmón/diagnóstico por imagen , Tomografía Computarizada por Rayos X/métodos , Adulto , Anciano , Anciano de 80 o más Años , Aprendizaje Profundo/normas , Femenino , Humanos , Italia , Masculino , Persona de Mediana Edad , Reproducibilidad de los Resultados , SARS-CoV-2 , Índice de Severidad de la EnfermedadRESUMEN
Nowadays, federated learning is one of the most prominent choices for making decisions. A significant benefit of federated learning is that, unlike deep learning, it is not necessary to share data samples with the model owner. The weight of the global model in traditional federated learning is created by averaging the weights of all clients or sites. In the proposed work, a novel method has been discussed to generate an optimized base model without hampering its performance, which is based on a genetic algorithm. Chromosome representation, crossover, and mutation-all the intermediate operations of the genetic algorithm have been illustrated with useful examples. After applying the genetic algorithm, there is a significant improvement in inference time and a huge reduction in storage space. Therefore, the model can be easily deployed on resource-constrained devices. For the experimental work, sports data has been used in balanced and unbalanced scenarios with various numbers of clients in a federated learning environment. In addition, we have used four famous deep learning architectures, such as AlexNet, VGG19, ResNet50, and EfficientNetB3, as the base model. We have achieved 92.34% accuracy with 9 clients in the balanced data set by using EfficientNetB3 as the base model using a GA-based approach. Moreover, after applying the genetic algorithm to optimize EfficientNetB3, there is an improvement in inference time and storage space by 20% and 2.35%, respectively.
Asunto(s)
Algoritmos , Deportes , Humanos , Aprendizaje Profundo , Redes Neurales de la ComputaciónRESUMEN
Researchers have adapted the conventional deep learning classification networks to generate Fully Conventional Networks (FCN) for carrying out accurate semantic segmentation. However, such models are expensive both in terms of storage and inference time and not readily employable on edge devices. In this paper, a compressed version of VGG16-based Fully Convolution Network (FCN) has been developed using Particle Swarm Optimization. It has been shown that the developed model can offer tremendous saving in storage space and also faster inference time, and can be implemented on edge devices. The efficacy of the proposed approach has been tested using potato late blight leaf images from publicly available PlantVillage dataset, street scene image dataset and lungs X-Ray dataset and it has been shown that it approaches the accuracies offered by standard FCN even after 851× compression.
RESUMEN
BACKGROUND: The Neonatal mortality rate in the United States is 3.8 deaths per 1000 live births, which is comparably higher than other nations. PURPOSE: The aim of the proposed study is to design and develop Artificial Intelligence (AI) models (NeoAI 1.0, Global Biomedical Technologies, Inc., Roseville, CA, USA) on risk variables extracted from the National Center for Health Statistics (NCHS) data from 2014 to 2017 duration, consisting of birth-death infant files to predict neonatal and infant deaths. METHODOLOGY: The NCHS data consisted of 15.8 million live birth records, including 91,773 infant deaths, out of which 61,222 were neonatal (life <28 days) and the rest were non-deaths. We designed and developed two different kinds of systems, labelled as neonatal and infant death systems. The data preparation consisted of balancing the two classes using the Adaptive Synthetic oversampling technique (ADASYN) paradigm. The best features were extracted using mutual information followed by 5-fold cross-validation using four different models, namely AdaBoost, XGBoost, Random Forest, and Logistic Regression based on balanced and unbalanced paradigms. RESULTS: XGBoost gave the best results for the neonatal system with AUC of 0.97 and 0.99 (p < 0.0001), while for the infant system, the scores were 0.91 and 0.99, both systems, without/with ADASYN integration, respectively. Further, there was a 60% increase in F1-score and sensitivity with ADASYN integration. The most important risk factors for classifier models along with feature extraction were maternal age and maternal race by Hispanic classification. Further, gestational age, labour aid and newborn condition were also part of the top five risk factors for these models. CONCLUSIONS: NoeAI showed two independent powerful machine learning (ML) systems and selected the best risk predictors combined with classification models for neonatal and infant deaths. The response time of the online platform was less than a second.
Asunto(s)
Inteligencia Artificial , Mortalidad Infantil , Edad Gestacional , Humanos , Lactante , Muerte del Lactante , Recién Nacido , Aprendizaje Automático , Estados UnidosRESUMEN
Vision-based fire detection systems have been significantly improved by deep models; however, higher numbers of false alarms and a slow inference speed still hinder their practical applicability in real-world scenarios. For a balanced trade-off between computational cost and accuracy, we introduce dual fire attention network (DFAN) to achieve effective yet efficient fire detection. The first attention mechanism highlights the most important channels from the features of an existing backbone model, yielding significantly emphasized feature maps. Then, a modified spatial attention mechanism is employed to capture spatial details and enhance the discrimination potential of fire and non-fire objects. We further optimize the DFAN for real-world applications by discarding a significant number of extra parameters using a meta-heuristic approach, which yields around 50% higher FPS values. Finally, we contribute a medium-scale challenging fire classification dataset by considering extremely diverse, highly similar fire/non-fire images and imbalanced classes, among many other complexities. The proposed dataset advances the traditional fire detection datasets by considering multiple classes to answer the following question: what is on fire? We perform experiments on four widely used fire detection datasets, and the DFAN provides the best results compared to 21 state-of-the-art methods. Consequently, our research provides a baseline for fire detection over edge devices with higher accuracy and better FPS values, and the proposed dataset extension provides indoor fire classes and a greater number of outdoor fire classes; these contributions can be used in significant future research. Our codes and dataset will be publicly available at https://github.com/tanveer-hussain/DFAN.
RESUMEN
Alzheimer's disease is incurable at the moment. If it can be appropriately diagnosed, the correct treatment can postpone the patient's illness. To aid in the diagnosis of Alzheimer's disease and to minimize the time and expense associated with manual diagnosis, a machine learning technique is employed, and a transfer learning method based on 3D MRI data is proposed. Machine learning algorithms can dramatically reduce the time and effort required for human treatment of Alzheimer's disease. This approach extracts bottleneck features from the M-Net migration network and then adds a top layer to supervised training to further decrease the dimensionality and delete portions. As a consequence, the transfer network presented in this study has several advantages in terms of computational efficiency and training time savings when used as a machine learning approach for AD-assisted diagnosis. Finally, the properties of all subject slices are combined and trained in the classification layer, completing the categorization of Alzheimer's disease symptoms and standard control. The results show that this strategy has a 1.5 percentage point better classification accuracy than the one that relies exclusively on VGG16 to extract bottleneck features. This strategy could cut the time it takes for the network to learn and improve its ability to classify things. The experiment shows that the method works by using data from OASIS. A typical transfer learning network's classification accuracy is about 8% better with this method than with a typical network, and it takes about 1/60 of the time with this method.
Asunto(s)
Enfermedad de Alzheimer , Enfermedad de Alzheimer/diagnóstico por imagen , Computadores , Diagnóstico por Computador/métodos , Humanos , Interpretación de Imagen Asistida por Computador/métodos , Imagen por Resonancia Magnética/métodosRESUMEN
Aiming at the inadequacy of the group decision-making method with the current attribute value as interval language information, an interval binary semantic decision-making method is proposed, which considers the decision maker's psychological behavior. The scope of this research is that this paper is based on localized amplification method. The localized amplification method used in this research may amplify physiological movement after removing unwanted noise, allowing the movement trend to be seen with the naked eye, improving the CNN network's mental identification accuracy. These two algorithms analyze the input picture from various perspectives, allowing the CNN network to extract more information and enhance identification accuracy. A new distance formula with interval binary semantics closer to decision-makers thinking habits is defined; time degree is introduced. An optimization model is established to solve the time series weights by considering the comprehensive consistency of expert evaluation. Based on prospect theory, a prospect deviation value is constructed and minimized weight optimization model, using the interactive multiple attribute decision community making (TODIM) method based on the new distance measure to calculate the total overall dominance of the schemes to rank the schemes. Taking the selection and evaluation of supply chain collaboration partners as an example, the effectiveness and rationality of the proposed method are verified.
Asunto(s)
Medios de Comunicación Sociales , Algoritmos , Toma de Decisiones , Humanos , LenguajeRESUMEN
Magnetoencephalography (MEG) is now widely used in clinical examinations and medical research in many fields. Resting-state magnetoencephalography-based brain network analysis can be used to study the physiological or pathological mechanisms of the brain. Furthermore, magnetoencephalography analysis has a significant reference value for the diagnosis of epilepsy. The scope of the proposed research is that this research demonstrates how to locate spikes in the phase locking functional brain connectivity network of the Desikan-Killiany brain region division using a neural network approach. It also improves detection accuracy and reduces missed and false detection rates. The automatic classification of epilepsy encephalomagnetic signals can make timely judgments on the patient's condition, which is of tremendous clinical significance. The existing literature's research on the automatic type of epilepsy EEG signals is relatively sufficient, but the research on epilepsy EEG signals is relatively weak. A full-band machine learning automatic discrimination method of epilepsy brain magnetic spikes based on the brain functional connection network is proposed. The four classifiers are comprehensively compared. The classifier with the best effect is selected, and the discrimination accuracy can reach 93.8%. Therefore, this method has a good application prospect in automatically identifying and labeling epileptic spikes in magnetoencephalography.
Asunto(s)
Electroencefalografía , Epilepsia , Encéfalo/diagnóstico por imagen , Electroencefalografía/métodos , Epilepsia/diagnóstico por imagen , Humanos , Fenómenos Magnéticos , Magnetoencefalografía/métodosRESUMEN
Human activity recognition (HAR) has multifaceted applications due to its worldly usage of acquisition devices such as smartphones, video cameras, and its ability to capture human activity data. While electronic devices and their applications are steadily growing, the advances in Artificial intelligence (AI) have revolutionized the ability to extract deep hidden information for accurate detection and its interpretation. This yields a better understanding of rapidly growing acquisition devices, AI, and applications, the three pillars of HAR under one roof. There are many review articles published on the general characteristics of HAR, a few have compared all the HAR devices at the same time, and few have explored the impact of evolving AI architecture. In our proposed review, a detailed narration on the three pillars of HAR is presented covering the period from 2011 to 2021. Further, the review presents the recommendations for an improved HAR design, its reliability, and stability. Five major findings were: (1) HAR constitutes three major pillars such as devices, AI and applications; (2) HAR has dominated the healthcare industry; (3) Hybrid AI models are in their infancy stage and needs considerable work for providing the stable and reliable design. Further, these trained models need solid prediction, high accuracy, generalization, and finally, meeting the objectives of the applications without bias; (4) little work was observed in abnormality detection during actions; and (5) almost no work has been done in forecasting actions. We conclude that: (a) HAR industry will evolve in terms of the three pillars of electronic devices, applications and the type of AI. (b) AI will provide a powerful impetus to the HAR industry in future. Supplementary Information: The online version contains supplementary material available at 10.1007/s10462-021-10116-x.
RESUMEN
BACKGROUND: Early and automated detection of carotid plaques prevents strokes, which are the second leading cause of death worldwide according to the World Health Organization. Artificial intelligence (AI) offers automated solutions for plaque tissue characterization. Recently, solo deep learning (SDL) models have been used, but they do not take advantage of the tandem connectivity offered by AI's hybrid nature. Therefore, this study explores the use of hybrid deep learning (HDL) models in a multicenter framework, making this study the first of its kind. METHODS: We hypothesize that HDL techniques perform better than SDL and transfer learning (TL) techniques. We propose two kinds of HDL frameworks: (i) the fusion of two SDLs (Inception with ResNet) or (ii) 10 other kinds of tandem models that fuse SDL with ML. The system Atheromatic™ 2.0HDL (AtheroPoint, CA, USA) was designed on an augmentation framework and three kinds of loss functions (cross-entropy, hinge, and mean-square-error) during training to determine the best optimization paradigm. These 11 combined HDL models were then benchmarked against one SDL model and five types of TL models; thus, this study considers a total of 17 AI models. RESULTS: Among the 17 AI models, the best performing HDL system was that comprising CNN and decision tree (DT), as its accuracy and area-under-the-curve were 99.78 ± 1.05% and 0.99 (p<0.0001), respectively. These values are 6.4% and 3.2% better than those recorded for the SDL and TL models, respectively. We validated the performance of the HDL models with diagnostics odds ratio (DOR) and Cohen and Kappa statistics; here, HDL outperformed DL and TL by 23% and 7%, respectively. The online system ran in <2 s. CONCLUSION: HDL is a fast, reliable, and effective tool for characterizing the carotid plaque for early stroke risk stratification.
Asunto(s)
Aprendizaje Profundo , Placa Aterosclerótica , Accidente Cerebrovascular , Inteligencia Artificial , Arterias Carótidas , Humanos , Placa Aterosclerótica/diagnóstico por imagenRESUMEN
Breast cancer is the second leading cause of death among women, behind only heart disease. However, despite the high incidence and mortality rates associated with breast cancer, it is still unclear as to what is responsible for its development in the first place. The prevention of breast cancer is not possible with any of the current available methods. Patients who are diagnosed and treated for breast cancer at an early stage have a better chance of having a successful treatment and recovery. In the field of breast cancer detection, digital mammography is widely acknowledged to be a highly effective method of detecting the disease early on. We may be able to improve early detection of breast cancer with the use of image processing techniques, thereby boosting our chances of survival and treatment success. This article discusses a breast cancer image processing and machine learning framework that was developed. The input data set for this framework is a sequence of mammography images, which are used as input data. The CLAHE approach is then utilized to improve the overall quality of the photographs by means of image processing. It is called contrast restricted adaptive histogram equalization (CLAHE), and it is an improvement on the original histogram equalization technique. This aids in the removal of noise from photographs while simultaneously improving picture quality. The segmentation of images is the next step in the framework's development. An image is divided into distinct portions at this point because the pixels are labeled at this step. This assists in the identification of objects and the delineation of boundaries. To categorize these preprocessed images, techniques such as fuzzy SVM, Bayesian classifier, and random forest are employed, among others.
Asunto(s)
Neoplasias de la Mama , Algoritmos , Teorema de Bayes , Mama/diagnóstico por imagen , Neoplasias de la Mama/diagnóstico por imagen , Femenino , Humanos , Procesamiento de Imagen Asistido por Computador , Aprendizaje Automático , Masculino , Mamografía/métodosRESUMEN
OBJECTIVE: Cardiovascular disease (CVD) is a major healthcare challenge and therefore early risk assessment is vital. Previous assessment techniques use either "conventional CVD risk calculators (CCVRC)" or machine learning (ML) paradigms. These techniques are ad-hoc, unreliable, not fully automated, and have variabilities. We, therefore, introduce AtheroEdge-MCDLAI (AE3.0DL) windows-based platform using multiclass Deep Learning (DL) system. METHODS: Data was collected on 500 patients having both carotid ultrasound and corresponding coronary angiography scores (CAS), measured as stenosis in coronary arteries and considered as the gold standard. A total of 39 covariates were used, clubbed into three clusters, namely (i) Office-based: age, gender, body mass index, smoker, hypertension, systolic blood pressure, and diastolic blood pressure; (ii) Laboratory-based: Hyperlipidemia, hemoglobin A1c, and estimated glomerular filtration rate; and (iii) Carotid ultrasound image phenotypes: maximum plaque height, total plaque area, and intra-plaque neovascularization. Baseline characteristics for four classes (target labels) having significant (p < 0.0001) values were calculated using Chi-square and ANOVA. For handling the cohort's imbalance in the risk classes, AE3.0DL used the synthetic minority over-sampling technique (SMOTE). AE3.0DL used Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) DL models and the performance (accuracy and area-under-the-curve) was computed using 10-fold cross-validation (90% training, 10% testing) frameworks. AE3.0DL was validated and benchmarked. RESULTS: The AE3.0DL using RNN and LSTM showed an accuracy and AUC (p < 0.0001) pairs as (95.00% and 0.98), and (95.34% and 0.99), respectively, and showed an improvement of 32.93% and 9.94% against CCVRC and ML, respectively. AE3.0DL runs in <1 s. CONCLUSION: DL algorithms are a powerful paradigm for coronary artery disease (CAD) risk prediction and CVD risk stratification.
Asunto(s)
Enfermedades Cardiovasculares , Enfermedades de las Arterias Carótidas , Enfermedad de la Arteria Coronaria , Aprendizaje Profundo , Placa Aterosclerótica , Humanos , Enfermedad de la Arteria Coronaria/diagnóstico por imagen , Ultrasonografía de las Arterias Carótidas , Inteligencia Artificial , Arterias Carótidas/diagnóstico por imagen , Ultrasonografía/métodos , Factores de Riesgo , Placa Aterosclerótica/diagnóstico por imagen , Medición de Riesgo/métodosRESUMEN
BACKGROUND: COVLIAS 1.0: an automated lung segmentation was designed for COVID-19 diagnosis. It has issues related to storage space and speed. This study shows that COVLIAS 2.0 uses pruned AI (PAI) networks for improving both storage and speed, wiliest high performance on lung segmentation and lesion localization. METHOD: ology: The proposed study uses multicenter â¼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted. RESULTS: Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions. CONCLUSIONS: Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.
Asunto(s)
COVID-19 , Aprendizaje Profundo , COVID-19/diagnóstico por imagen , Prueba de COVID-19 , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Pulmón/diagnóstico por imagen , Redes Neurales de la Computación , Reproducibilidad de los Resultados , Tomografía Computarizada por Rayos X/métodosRESUMEN
Diabetes is one of the main causes of the rising cases of blindness in adults. This microvascular complication of diabetes is termed diabetic retinopathy (DR) and is associated with an expanding risk of cardiovascular events in diabetes patients. DR, in its various forms, is seen to be a powerful indicator of atherosclerosis. Further, the macrovascular complication of diabetes leads to coronary artery disease (CAD). Thus, the timely identification of cardiovascular disease (CVD) complications in DR patients is of utmost importance. Since CAD risk assessment is expensive for low-income countries, it is important to look for surrogate biomarkers for risk stratification of CVD in DR patients. Due to the common genetic makeup between the coronary and carotid arteries, low-cost, high-resolution imaging such as carotid B-mode ultrasound (US) can be used for arterial tissue characterization and risk stratification in DR patients. The advent of artificial intelligence (AI) techniques has facilitated the handling of large cohorts in a big data framework to identify atherosclerotic plaque features in arterial ultrasound. This enables timely CVD risk assessment and risk stratification of patients with DR. Thus, this review focuses on understanding the pathophysiology of DR, retinal and CAD imaging, the role of surrogate markers for CVD, and finally, the CVD risk stratification of DR patients. The review shows a step-by-step cyclic activity of how diabetes and atherosclerotic disease cause DR, leading to the worsening of CVD. We propose a solution to how AI can help in the identification of CVD risk. Lastly, we analyze the role of DR/CVD in the COVID-19 framework.
RESUMEN
Background and Purpose: Only 1-2% of the internal carotid artery asymptomatic plaques are unstable as a result of >80% stenosis. Thus, unnecessary efforts can be saved if these plaques can be characterized and classified into symptomatic and asymptomatic using non-invasive B-mode ultrasound. Earlier plaque tissue characterization (PTC) methods were machine learning (ML)-based, which used hand-crafted features that yielded lower accuracy and unreliability. The proposed study shows the role of transfer learning (TL)-based deep learning models for PTC. Methods: As pertained weights were used in the supercomputer framework, we hypothesize that transfer learning (TL) provides improved performance compared with deep learning. We applied 11 kinds of artificial intelligence (AI) models, 10 of them were augmented and optimized using TL approaches-a class of Atheromatic™ 2.0 TL (AtheroPoint™, Roseville, CA, USA) that consisted of (i-ii) Visual Geometric Group-16, 19 (VGG16, 19); (iii) Inception V3 (IV3); (iv-v) DenseNet121, 169; (vi) XceptionNet; (vii) ResNet50; (viii) MobileNet; (ix) AlexNet; (x) SqueezeNet; and one DL-based (xi) SuriNet-derived from UNet. We benchmark 11 AI models against our earlier deep convolutional neural network (DCNN) model. Results: The best performing TL was MobileNet, with accuracy and area-under-the-curve (AUC) pairs of 96.10 ± 3% and 0.961 (p < 0.0001), respectively. In DL, DCNN was comparable to SuriNet, with an accuracy of 95.66% and 92.7 ± 5.66%, and an AUC of 0.956 (p < 0.0001) and 0.927 (p < 0.0001), respectively. We validated the performance of the AI architectures with established biomarkers such as greyscale median (GSM), fractal dimension (FD), higher-order spectra (HOS), and visual heatmaps. We benchmarked against previously developed Atheromatic™ 1.0 ML and showed an improvement of 12.9%. Conclusions: TL is a powerful AI tool for PTC into symptomatic and asymptomatic plaques.
RESUMEN
SARS-CoV-2 has infected over â¼165 million people worldwide causing Acute Respiratory Distress Syndrome (ARDS) and has killed â¼3.4 million people. Artificial Intelligence (AI) has shown to benefit in the biomedical image such as X-ray/Computed Tomography in diagnosis of ARDS, but there are limited AI-based systematic reviews (aiSR). The purpose of this study is to understand the Risk-of-Bias (RoB) in a non-randomized AI trial for handling ARDS using novel AtheroPoint-AI-Bias (AP(ai)Bias). Our hypothesis for acceptance of a study to be in low RoB must have a mean score of 80% in a study. Using the PRISMA model, 42 best AI studies were analyzed to understand the RoB. Using the AP(ai)Bias paradigm, the top 19 studies were then chosen using the raw-cutoff of 1.9. This was obtained using the intersection of the cumulative plot of "mean score vs. study" and score distribution. Finally, these studies were benchmarked against ROBINS-I and PROBAST paradigm. Our observation showed that AP(ai)Bias, ROBINS-I, and PROBAST had only 32%, 16%, and 26% studies, respectively in low-moderate RoB (cutoff>2.5), however none of them met the RoB hypothesis. Further, the aiSR analysis recommends six primary and six secondary recommendations for the non-randomized AI for ARDS. The primary recommendations for improvement in AI-based ARDS design inclusive of (i) comorbidity, (ii) inter-and intra-observer variability studies, (iii) large data size, (iv) clinical validation, (v) granularity of COVID-19 risk, and (vi) cross-modality scientific validation. The AI is an important component for diagnosis of ARDS and the recommendations must be followed to lower the RoB.
Asunto(s)
COVID-19 , Síndrome de Dificultad Respiratoria , Inteligencia Artificial , Humanos , Pulmón , Síndrome de Dificultad Respiratoria/diagnóstico por imagen , SARS-CoV-2RESUMEN
Visual or manual characterization and classification of atherosclerotic plaque lesions are tedious, error-prone, and time-consuming. The purpose of this study is to develop and design an automated carotid plaque characterization and classification system into binary classes, namely symptomatic and asymptomatic types via the deep learning (DL) framework implemented on a supercomputer. We hypothesize that on ultrasound images, symptomatic carotid plaques have (a) a low grayscale median because of a histologically large lipid core and relatively little collagen and calcium, and (b) a higher chaotic (heterogeneous) grayscale distribution due to the composition. The methodology consisted of building a DL model of Artificial Intelligence (called Atheromatic 2.0, AtheroPoint, CA, USA) that used a classic convolution neural network consisting of 13 layers and implemented on a supercomputer. The DL model used a cross-validation protocol for estimating the classification accuracy (ACC) and area-under-the-curve (AUC). A sample of 346 carotid ultrasound-based delineated plaques were used (196 symptomatic and 150 asymptomatic, mean age 69.9 ± 7.8 years, with 39% females). This was augmented using geometric transformation yielding 2312 plaques (1191 symptomatic and 1120 asymptomatic plaques). K10 (90% training and 10% testing) cross-validation DL protocol was implemented and showed an (i) accuracy and (ii) AUC without and with augmentation of 86.17%, 0.86 (p-value < 0.0001), and 89.7%, 0.91 (p-value < 0.0001), respectively. The DL characterization system consisted of validation of the two hypotheses: (a) mean feature strength (MFS) and (b) Mandelbrot's fractal dimension (FD) for measuring chaotic behavior. We demonstrated that both MFS and FD were higher in symptomatic plaques compared to asymptomatic plaques by 64.15 ± 0.73% (p-value < 0.0001) and 6 ± 0.13% (p-value < 0.0001), respectively. The benchmarking results show that DL with augmentation (ACC: 89.7%, AUC: 0.91 (p-value < 0.0001)) is superior to previously published machine learning (ACC: 83.7%) by 6.0%. The Atheromatic runs the test patient in < 2 s. Deep learning can be a useful tool for carotid ultrasound-based characterization and classification of symptomatic and asymptomatic plaques.
Asunto(s)
Enfermedades Cardiovasculares , Estenosis Carotídea , Aprendizaje Profundo , Placa Aterosclerótica , Accidente Cerebrovascular , Anciano , Inteligencia Artificial , Arterias Carótidas/diagnóstico por imagen , Arteria Carótida Interna/diagnóstico por imagen , Estenosis Carotídea/diagnóstico por imagen , Femenino , Humanos , Masculino , Valor Predictivo de las Pruebas , Medición de Riesgo , UltrasonografíaRESUMEN
Wilson's disease (WD) is caused by copper accumulation in the brain and liver, and if not treated early, can lead to severe disability and death. WD has shown white matter hyperintensity (WMH) in the brain magnetic resonance scans (MRI) scans, but the diagnosis is challenging due to (i) subtle intensity changes and (ii) weak training MRI when using artificial intelligence (AI). Design and validate seven types of high-performing AI-based computer-aided design (CADx) systems consisting of 3D optimized classification, and characterization of WD against controls. We propose a "conventional deep convolution neural network" (cDCNN) and an "improved DCNN" (iDCNN) where rectified linear unit (ReLU) activation function was modified ensuring "differentiable at zero." Three-dimensional optimization was achieved by recording accuracy while changing the CNN layers and augmentation by several folds. WD was characterized using (i) CNN-based feature map strength and (ii) Bispectrum strengths of pixels having higher probabilities of WD. We further computed the (a) area under the curve (AUC), (b) diagnostic odds ratio (DOR), (c) reliability, and (d) stability and (e) benchmarking. Optimal results were achieved using 9 layers of CNN, with 4-fold augmentation. iDCNN yields superior performance compared to cDCNN with accuracy and AUC of 98.28 ± 1.55, 0.99 (p < 0.0001), and 97.19 ± 2.53%, 0.984 (p < 0.0001), respectively. DOR of iDCNN outperformed cDCNN fourfold. iDCNN also outperformed (a) transfer learning-based "Inception V3" paradigm by 11.92% and (b) four types of "conventional machine learning-based systems": k-NN, decision tree, support vector machine, and random forest by 55.13%, 28.36%, 15.35%, and 14.11%, respectively. The AI-based systems can potentially be useful in the early WD diagnosis. Graphical Abstract.
Asunto(s)
Inteligencia Artificial , Degeneración Hepatolenticular , Encéfalo/diagnóstico por imagen , Degeneración Hepatolenticular/diagnóstico por imagen , Humanos , Imagen por Resonancia Magnética , Reproducibilidad de los ResultadosRESUMEN
BACKGROUND: COVID-19 pandemic has currently no vaccines. Thus, the only feasible solution for prevention relies on the detection of COVID-19-positive cases through quick and accurate testing. Since artificial intelligence (AI) offers the powerful mechanism to automatically extract the tissue features and characterise the disease, we therefore hypothesise that AI-based strategies can provide quick detection and classification, especially for radiological computed tomography (CT) lung scans. METHODOLOGY: Six models, two traditional machine learning (ML)-based (k-NN and RF), two transfer learning (TL)-based (VGG19 and InceptionV3), and the last two were our custom-designed deep learning (DL) models (CNN and iCNN), were developed for classification between COVID pneumonia (CoP) and non-COVID pneumonia (NCoP). K10 cross-validation (90% training: 10% testing) protocol on an Italian cohort of 100 CoP and 30 NCoP patients was used for performance evaluation and bispectrum analysis for CT lung characterisation. RESULTS: Using K10 protocol, our results showed the accuracy in the order of DL > TL > ML, ranging the six accuracies for k-NN, RF, VGG19, IV3, CNN, iCNN as 74.58 ± 2.44%, 96.84 ± 2.6, 94.84 ± 2.85%, 99.53 ± 0.75%, 99.53 ± 1.05%, and 99.69 ± 0.66%, respectively. The corresponding AUCs were 0.74, 0.94, 0.96, 0.99, 0.99, and 0.99 (p-values < 0.0001), respectively. Our Bispectrum-based characterisation system suggested CoP can be separated against NCoP using AI models. COVID risk severity stratification also showed a high correlation of 0.7270 (p < 0.0001) with clinical scores such as ground-glass opacities (GGO), further validating our AI models. CONCLUSIONS: We prove our hypothesis by demonstrating that all the six AI models successfully classified CoP against NCoP due to the strong presence of contrasting features such as ground-glass opacities (GGO), consolidations, and pleural effusion in CoP patients. Further, our online system takes < 2 s for inference.