Search

An essential aspect of medical research is the prediction for a health outcome and the scientific identification of important factors. As a result, numerous methods were developed for model selections in recent years. In the era of big data, machine learning has been broadly adopted for data analysis. In particular, the Support Vector Machine (SVM) has an excellent performance in classifications and predictions with the high-dimensional data. In this research, a novel model selection strategy is carried out, named as the Stepwise Support Vector Machine (StepSVM). The new strategy is based on the SVM to conduct a modified stepwise selection, where the tuning parameter could be determined by 10-fold cross-validation that minimizes the mean squared error. Two popular methods, the conventional stepwise logistic regression model and the SVM Recursive Feature Elimination (SVM-RFE), were compared to the StepSVM. The Stability and accuracy of the three strategies were evaluated by simulation studies with a complex hierarchical structure. Up to five variables were selected to predict the dichotomous cancer remission of a lung cancer patient. Regarding the stepwise logistic regression, the mean of the C-statistic was 69.19%. The overall accuracy of the SVM-RFE was estimated at 70.62%. In contrast, the StepSVM provided the highest prediction accuracy of 80.57%. Although the StepSVM is more time consuming, it is more consistent and outperforms the other two methods. [ABSTRACT FROM AUTHOR]

Switching different gait according to different movements is an important direction in the study of exoskeleton robot. Identifying the movement intention of the wearer to control the gait planning of the exoskeleton robot can effectively improve the man-machine interaction experience after the exoskeleton. This paper uses a support vector machine (SVM) to realize wearer's motion posture recognition by collecting sEMG signals on the human surface. The moving gait of the exoskeleton is planned according to the recognition results, and the decoding intention signal controls gait switching. Meanwhile, the stability of the planned gait during the movement was analyzed. Experimental results show that the sEMG signal decoding human motion intentional, and control exoskeleton robot gait switching has good accuracy and real-time performance. It helps patients to complete rehabilitation training more safely and quickly. [ABSTRACT FROM AUTHOR]

As a characteristic edible fungus with a high nutritional value and medicinal effect, the Bachu mushroom has a broad market. To distinguish among Bachu mushrooms with high value and other fungi effectively and accurately, as well as to explore a universal identification method, this study proposed a method to identify Bachu mushrooms by Fourier Transform Infrared Spectroscopy (FT-IR) combined with machine learning. In this experiment, two kinds of common edible mushrooms, Lentinus edodes and club fungi, were selected and classified with Bachu mushrooms. Due to the different distribution of nutrients in the caps and stalks, the caps and stalks were studied in this experiment. By comparing the average normalized infrared spectra of the caps and stalks of the three types of fungi, we found differences in their infrared spectra, indicating that the latter can be used to classify and identify the three types of fungi. We also used machine learning to process the spectral data. The overall steps of data processing are as follows: use partial least squares (PLS) to extract spectral features, select the appropriate characteristic number, use different classification algorithms for classification, and finally determine the best algorithm according to the classification results. Among them, the basis of selecting the characteristic number was the cumulative variance interpretation rate. To improve the reliability of the experimental results, this study also used the classification results to verify the feasibility. The classification algorithms used in this study were the support vector machine (SVM), backpropagation neural network (BPNN) and k-nearest neighbors (KNN) algorithm. The results showed that the three algorithms achieved good results in the multivariate classification of the caps and stalks data. Moreover, the cumulative variance explanation rate could be used to select the characteristic number. Finally, by comparing the classification results of the three algorithms, the classification effect of KNN was found to be the best. Additionally, the classification results were as follows: according to the caps data classification, the accuracy was 99.06%; according to the stalks data classification, the accuracy was 99.82%. This study showed that infrared spectroscopy combined with a machine learning algorithm has the potential to be applied to identify Bachu mushrooms and the cumulative variance explanation rate can be used to select the characteristic number. This method can also be used to identify other types of edible fungi and has a broad application prospect. [ABSTRACT FROM AUTHOR]

In this research, an efficient scheme to identify leaf types is proposed. In that scheme, the leaf boundary points are fitted in a continuous contour using Radial Basis Function Neural Networks (RBFNN) to calculate the centroid of the leaf shape. Afterwards, the distances between predetermined points and the centroid were computed and normalized. In addition, the time complexity of the features' extraction algorithm was calculated. The merit of this scheme is objects' independence to translation, rotation and scaling. Moreover, different classification techniques were evaluated against the leaf shape features. Those techniques included two of the most commonly used classification methods; RBFNN and SVM that were evaluated and compared with other researches that used complex features extraction algorithms with much higher dimensionality. Furthermore, a third classification method with an optimization technique for the SVM using Salp Swarm Algorithm (SSA) was utilized showing a significant improvement over RBFNN and SVM. [ABSTRACT FROM AUTHOR]

Breast cancer is the most common cancer among women and it is one of the main causes of death for women worldwide. To attain an optimum medical treatment for breast cancer, an early breast cancer detection is crucial. This paper proposes a multi- stage feature selection method that extracts statistically significant features for breast cancer size detection using proposed data normalization techniques. Ultra-wideband (UWB) signals, controlled using microcontroller are transmitted via an antenna from one end of the breast phantom and are received on the other end. These ultra-wideband analogue signals are represented in both time and frequency domain. The preprocessed digital data is passed to the proposed multi- stage feature selection algorithm. This algorithm has four selection stages. It comprises of data normalization methods, feature extraction, data dimensional reduction and feature fusion. The output data is fused together to form the proposed datasets, namely, 8-HybridFeature, 9-HybridFeature and 10-HybridFeature datasets. The classification performance of these datasets is tested using the Support Vector Machine, Probabilistic Neural Network and Naïve Bayes classifiers for breast cancer size classification. The research findings indicate that the 8-HybridFeature dataset performs better in comparison to the other two datasets. For the 8-HybridFeature dataset, the Naïve Bayes classifier (91.98%) outperformed the Support Vector Machine (90.44%) and Probabilistic Neural Network (80.05%) classifiers in terms of classification accuracy. The finalized method is tested and visualized in the MATLAB based 2D and 3D environment. [ABSTRACT FROM AUTHOR]

Background: Neoadjuvant chemotherapy (NAC) is the standard of care for patients with locally advanced breast cancer (LABC). The study was conducted to investigate the utility of quantitative ultrasound (QUS) carried out during NAC to predict the final tumour response in a multi-institutional setting. Methods: Fifty-nine patients with LABC were enrolled from three institutions in North America (Sunnybrook Health Sciences Centre (Toronto, Canada), MD Anderson Cancer Centre (Texas, USA), and Princess Margaret Cancer Centre (Toronto, Canada)). QUS data were collected before starting NAC and subsequently at weeks 1 and 4 during chemotherapy. Spectral tumour parametric maps were generated, and textural features determined using grey-level co-occurrence matrices. Patients were divided into two groups based on their pathological outcomes following surgery: responders and non-responders. Machine learning algorithms using Fisher's linear discriminant (FLD), K-nearest neighbour (K-NN), and support vector machine (SVM-RBF) were used to generate response classification models. Results: Thirty-six patients were classified as responders and twenty-three as non-responders. Among all the models, SVM-RBF had the highest accuracy of 81% at both weeks 1 and week 4 with area under curve (AUC) values of 0.87 each. The inclusion of week 1 and 4 features led to an improvement of the classifier models, with the accuracy and AUC from baseline features only being 76% and 0.68, respectively. Conclusion: QUS data obtained during NAC reflect the ongoing treatment-related changes during chemotherapy and can lead to better classifier performances in predicting the ultimate pathologic response to treatment compared to baseline features alone. [ABSTRACT FROM AUTHOR]

The Alzheimer's Disease Neuroimaging (ADNI) database is an expansive undertaking by government, academia, and industry to pool resources and data on subjects at various stage of symptomatic severity due to Alzheimer's disease. As expected, magnetic resonance imaging is a major component of the project. Full brain images are obtained at every 6-month visit. A range of cognitive tests studying executive function and memory are employed less frequently. Two blood draws (baseline, 6 months) provide samples to measure concentrations of approximately 145 plasma biomarkers. In addition, other diagnostic measurements are performed including PET imaging, cerebral spinal fluid measurements of amyloid-beta and tau peptides, as well as genetic tests, demographics, and vital signs. ADNI data is available upon review of an application. There have been numerous reports of how various processes evolve during AD progression, including alterations in metabolic and neuroendocrine activity, cell survival, and cognitive behavior. Lacking an analytic model at the onset, we leveraged recent advances in machine learning, which allow us to deal with large, non-linear systems with many variables. Of particular note was examining how well binary predictions of future disease states could be learned from simple, non-invasive measurements like those dependent on blood samples. Such measurements make relatively little demands on the time and effort of medical staff or patient. We report findings with recall/precision/area under the receiver operator curve after application of CART, Random Forest, Gradient Boosting, and Support Vector Machines, Our results show (i) Random Forests and Gradient Boosting work very well with such data, (ii) Prediction quality when applied to relatively easily obtained measurements (Cognitive scores, Genetic Risk and plasma biomarkers) achieve results that are competitive with magnetic resonance techniques. This is by no means an exhaustive study, but instead an exploration of the plausibility of defining a series of relatively inexpensive, broad population based tests. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the interaction impacts of body mass index (BMI) on the other important risk factors for venous thromboembolism (VTE), using deep venous thrombosis (DVT) patient data from the International Warfarin Pharmacogenetics Consortium (IWPC). We apply eight machine learning techniques, including naive Bayes classifier (NB), support vector machine (SVM), elastic net regression (ENET), logistic regression (LR), lasso regression (LAR), multivariate adaptive regression splines (MARS), boosted regression tree (BRT) and random forest model (RF). The RF method is selected as the best model for classification. Out of 33 features considered in this study, we identify 12 variables as relatively important risk factors for VTE. Finally, we examine the interaction impacts of BMI on these important VTE risk factors. We conclude that the impacts of risk factors on VTE incidence are varying across different BMI groups, and the variations are different for different risk factors. Therefore the interaction impacts of BMI on the other risk factors have to be taken into account in order to better understand the incidence of VTE. [ABSTRACT FROM AUTHOR]

Objectives: We aimed to build a machine learning predictive model to predict the risk of prolonged mechanical ventilation (PMV) for patients with Traumatic Brain Injury (TBI). Methods: This study included TBI patients who were hospitalized in a level 1 trauma center between January 2014 and February 2019. Data were analyzed for all adult patients who received mechanical ventilation following TBI with abbreviated injury severity (AIS) score for the head region of ≥ 3. This study designed three sets of machine learning models: set A defined PMV to be greater than 7 days, set B (PMV > 10 days) and set C (PMV >14 days) to determine the optimal model for deployment. Patients' demographics, injury characteristics and CT findings were used as predictors. Logistic regression (LR), Artificial neural networks (ANN) Support vector machines (SVM), Random Forest (RF) and C.5 Decision Tree (C.5 DT) were used to predict the PMV. Results: The number of eligible patients that were included in the study were 674, 643 and 622 patients in sets A, B and C respectively. In set A, LR achieved the optimal performance with accuracy 0.75 and Area under the curve (AUC) 0.83. SVM achieved the optimal performance among other models in sets B with accuracy/AUC of 0.79/0.84 respectively. ANNs achieved the optimal performance in set C with accuracy/AUC of 0.76/0.72 respectively. Machine learning models in set B demonstrated more stable performance with higher prediction success and discrimination power. Conclusion: This study not only provides evidence that machine learning methods outperform the traditional multivariate analytical methods, but also provides a perspective to reach a consensual definition of PMV. [ABSTRACT FROM AUTHOR]

Objective: Current urinary tract infection (UTI) diagnostic strategies that rely on leukocyte esterase have limited accuracy. We performed an aptamer-based proteomics pilot study to identify urine protein levels that could differentiate a culture proven UTI from culture negative samples, regardless of pyuria status. Methods: We analyzed urine from 16 children with UTIs, 8 children with culture negative pyuria and 8 children with negative urine culture and no pyuria. The urine levels of 1,310 proteins were quantified using the Somascan™ platform and normalized to urine creatinine. Machine learning with support vector machine (SVM)-based feature selection was performed to determine the combination of urine biomarkers that optimized diagnostic accuracy. Results: Eight candidate urine protein biomarkers met filtering criteria. B-cell lymphoma protein, C-X-C motif chemokine 6, C-X-C motif chemokine 13, cathepsin S, heat shock 70kDA protein 1A, mitogen activated protein kinase, protein E7 HPV18 and transgelin. AUCs ranged from 0.91 to 0.95. The best prediction was achieved by the SVMs with radial basis function kernel. Conclusions: Biomarkers panel can be identified by the emerging technologies of aptamer-based proteomics and machine learning that offer the potential to increase UTI diagnostic accuracy, thereby limiting unneeded antibiotics. [ABSTRACT FROM AUTHOR]

Segmentation of a natural scene into objects and background is a fundamental but challenging task for recognizing objects. Investigating intermediate-level visual cortical areas with a focus on local information is a crucial step towards understanding the formation of the cortical representations of figure and ground. We examined the activity of a population of macaque V4 neurons during the presentation of natural image patches and their respective variations. The natural image patches were optimized to exclude the influence of global context but included various characteristics of local stimulus. Around one fourth of the patch-responsive V4 neurons exhibited significant modulation of firing activity that was dependent on the positional relation between the figural region of the stimulus and the classical receptive field of the neuron. However, the individual neurons showed low consistency in figure-ground modulation across a variety of image patches (55–62%), indicating that individual neurons were capable of correctly signaling figure and ground only for a limited number of stimuli. We examined whether integration of the activity of multiple neurons enabled higher consistency across a variety of natural patches by training a support vector machine to classify figure and ground of the stimuli from the population firing activity. The integration of the activity of a few tens of neurons yielded discrimination accuracy much greater than that of single neurons (up to 85%), suggesting a crucial role of population coding for figure-ground discrimination in natural images. [ABSTRACT FROM AUTHOR]

Tracking the origin of propagating wave signals in an environment with complex reflective surfaces is, in its full generality, a nearly intractable problem which has engendered multiple domain-specific literatures. We posit that, if the environment and sensor geometries are fixed, machine learning algorithms can "learn" the acoustical geometry of the environment and accurately track signal origin. In this paper, we propose the first machine-learning-based approach to identifying the source locations of semi-stationary, tonal, dolphin-whistle-like sounds in a highly reverberant space, specifically a half-cylindrical dolphin pool. Our algorithm works by supplying a learning network with an overabundance of location "clues", which are then selected under supervised training for their ability to discriminate source location in this particular environment. More specifically, we deliver estimated time-difference-of-arrivals (TDOA's) and normalized cross-correlation values computed from pairs of hydrophone signals to a random forest model for high-feature-volume classification and feature selection, and subsequently deliver the selected features into linear discriminant analysis, linear and quadratic Support Vector Machine (SVM), and Gaussian process models. Based on data from 14 sound source locations and 16 hydrophones, our classification models yielded perfect accuracy at predicting novel sound source locations. Our regression models yielded better accuracy than the established Steered-Response Power (SRP) method when all training data were used, and comparable accuracy along the pool surface when deprived of training data at testing sites; our methods additionally boast improved computation time and the potential for superior localization accuracy in all dimensions with more training data. Because of the generality of our method we argue it may be useful in a much wider variety of contexts. [ABSTRACT FROM AUTHOR]

Background and purpose: Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke. Methods: We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154). Results: Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it. Conclusions: The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice. [ABSTRACT FROM AUTHOR]

Objective: The present study aims to explore the role of smoking factors in the risk of lung cancer and screen the feature risk pathways of smoking-induced lung cancer. Methods: The expression profiles of the patient data from GEO database were standardized, and differentially expressed genes (DEGs) were analyzed by limma algorithm. Samples and genes were analyzed by Unsupervised hierarchical clustering method, while GO and KEGG enrichment analyses were performed on DEGs. The data of the protein-protein interaction (PPI) network were downloaded from the BioGrid and HPRD databases, and the DEGs were mapped into the PPI network to identify the interaction relationship. The enriched significant pathways were used to calculate the anomaly score and RFE method was used to optimize the feature sets. The model was trained using the support vector machine (SVM) and the predicted results were plotted into ROC curves. The AUC value was calculated to evaluate the predictive performance of the SVM model. Results: A total of 1923 DEGs were obtained, of which 826 were down-regulated and 1097 were up-regulated. Unsupervised hierarchical clustering analysis showed that the diagnosis accuracy of lung cancer smokers was 74%, and that of non-lung cancer smokers was 75%. Five optimal feature pathway sets were obtained by screening, the clinical diagnostic ability of which was detected by SVM model with the accuracy improved to 84%. The diagnostic accuracy was 90% after combining clinical information. Conclusion: We verified that five signaling pathways combined with clinical information could be used as a feature risk pathway for identifying lung cancer smokers and non-lung cancer smokers and increased the diagnostic accuracy. [ABSTRACT FROM AUTHOR]

Background: The pattern of myocardial fibrosis differs significantly between different cardiomyopathies. Fibrosis in hypertrophic cardiomyopathy (HCM) is characteristically as patchy and regional but in dilated cardiomyopathy (DCM) as diffuse and global. We sought to investigate if texture analyses on myocardial native T1 mapping can differentiate between fibrosis patterns in patients with HCM and DCM. Methods: We prospectively acquired native myocardial T1 mapping images for 321 subjects (55±15 years, 70% male): 65 control, 116 HCM, and 140 DCM patients. To quantify different fibrosis patterns, four sets of texture descriptors were used to extract 152 texture features from native T1 maps. Seven features were sequentially selected to identify HCM- and DCM-specific patterns in 70% of data (training dataset). Pattern reproducibility and generalizability were tested on the rest of data (testing dataset) using support vector machines (SVM) and regression models. Results: Pattern-derived texture features were capable to identify subjects in HCM, DCM, and controls cohorts with 202/237(85.2%) accuracy of all subjects in the training dataset using 10-fold cross-validation on SVM (AUC = 0.93, 0.93, and 0.93 for controls, HCM and DCM, respectively), while pattern-independent global native T1 mapping was poorly capable to identify those subjects with 121/237(51.1%) accuracy (AUC = 0.78, 0.51, and 0.74) (P<0.001 for all). The pattern-derived features were reproducible with excellent intra- and inter-observer reliability and generalizable on the testing dataset with 75/84(89.3%) accuracy. Conclusion: Texture analysis of myocardial native T1 mapping can characterize fibrosis patterns in HCM and DCM patients and provides additional information beyond average native T1 values. [ABSTRACT FROM AUTHOR]

Background: Although dialysis patients are at a high risk of death, it is difficult for medical practitioners to simultaneously evaluate many inter-related risk factors. In this study, we evaluated the characteristics of hemodialysis patients using machine learning model, and its usefulness for screening hemodialysis patients at a high risk of one-year death using the nation-wide database of the Japanese Society for Dialysis Therapy. Materials and methods: The patients were separated into two datasets (n = 39,930, 39,930, respectively). We categorized hemodialysis patients in Japan into new clusters generated by the K-means clustering method using the development dataset. The association between a cluster and the risk of death was evaluated using multivariate Cox proportional hazards models. Then, we developed an ensemble model composed of the clusters and support vector machine models in the model development phase, and compared the accuracy of the prediction of mortality between the machine learning models in the model validation phase. Results: Average age of the subjects was 65.7±12.2 years; 32.7% had diabetes mellitus. The five clusters clearly distinguished the groups on the basis of their characteristics: Cluster 1, young male, and chronic glomerulonephritis; Cluster 2, female, and chronic glomerulonephritis; Cluster 3, diabetes mellitus; Cluster 4, elderly and nephrosclerosis; Cluster 5, elderly and protein energy wasting. These clusters were associated with the risk of death; Cluster 5 compared with Cluster 1, hazard ratio 8.86 (95% CI 7.68, 10.21). The accuracy of the ensemble model for the prediction of 1-year death was 0.948 and higher than those of logistic regression model (0.938), support vector machine model (0.937), and deep learning model (0.936). Conclusions: The clusters clearly categorized patient on their characteristics, and reflected their prognosis. Our real-world-data-based machine learning system is applicable to identifying high-risk hemodialysis patients in clinical settings, and has a strong potential to guide treatments and improve their prognosis. [ABSTRACT FROM AUTHOR]

The factors that determine Serum Thyrotropin (TSH) levels have been examined through different methods, using different covariates. However, the use of machine learning methods has so far not been studied in population databases like NHANES (National Health and Nutritional Examination Survey) to predict TSH. In this study, we performed a comparative analysis of different machine learning methods like Linear regression, Random forest, Support vector machine, multilayer perceptron and stacking regression to predict TSH and classify individuals with normal, low and high TSH levels. We considered Free T4, Anti-TPO antibodies, T3, Body Mass Index (BMI), Age and Ethnicity as the predictor variables. A total of 9818 subjects were included in this comparative analysis. We used coefficient of determination (r2) value to compare the results for predicting the TSH and show that the Random Forest, Gradient Boosting and Stacking Regression perform equally well in predicting TSH and achieve the highest r2 value = 0.13, with mean absolute error of 0.78. Moreover, we found that Anti-TPO is the most important feature in predicting TSH followed by Age, BMI, T3 and Free-T4 for the regression analysis. While classifying TSH into normal, high or low levels, our comparative analysis also shows that Random forest performs the best in the classification study, performed with individuals with normal, high and low levels of TSH. We found the following Areas Under Curve (AUC); for low TSH, AUC = 0.61, normal TSH, AUC = 0.61 and elevated TSH AUC = 0.69. Additionally, we found that Anti-TPO was the most important feature in classifying TSH. In this study, we suggest that artificial intelligence and machine learning methods might offer an insight into the complex hypothalamic-pituitary -thyroid axis and may be an invaluable tool that guides us in making appropriate therapeutic decisions (thyroid hormone dosing) for the individual patient. [ABSTRACT FROM AUTHOR]

Machine learning models to predict energy expenditure (EE) from accelerometer data have traditionally been trained on data from laboratory-based activity trials. However, accuracy is typically attenuated when implemented in free-living scenarios. Currently, no studies involving preschool children have evaluated the accuracy of EE prediction models trained on laboratory (LAB) under free-living conditions. Purpose: To evaluate the accuracy of LAB EE prediction models in preschool children completing a free-living active play session. Performance was benchmarked against EE prediction models trained on free living (FL) data. Methods: 25 children (mean age = 4.1±1.0 y) completed a 20-minute active play session while wearing a portable indirect calorimeter and ActiGraph GT3X+ accelerometers on their right hip and non-dominant wrist. EE was predicted using LAB models which included Random Forest (RF) and Support Vector Machine (SVM) models for the wrist, and RF and Artificial Neural Network (ANN) models for the hip. Two variations of the LAB models were evaluated; 1) an "off the shelf" model without additional training; 2) models retrained on free-living data, replicating the methodology used in the original calibration study (retrained LAB). Prediction errors were evaluated in a hold-out sample of 10 children. Results: Root mean square error (RMSE) for the FL and retrained LAB models ranged from 0.63–0.67 kcals/min. In the hold out sample, RMSE's for the hip LAB (0.62–0.71), retrained LAB (0.58–0.62) and FL models (0.61–0.65) were similar. For the wrist placement, FL SVM had a significantly higher RMSE (0.73 ± 0.29 kcals/min) than the retrained LAB SVM (0.63 ± 0.30 kcals/min) and LAB SVM (0.64 ± 0.18 kcals/min). The LAB (0.64 ± 0.28), retrained LAB (0.64 ± 0.25), and FL (0.62 ± 0.26) RF exhibited comparable accuracy. Conclusion: Machine learning EE prediction models trained on LAB and FL data had similar accuracy under free-living conditions. [ABSTRACT FROM AUTHOR]

Objective: Hip fractures are among the most frequently occurring fragility fractures in older adults, associated with a loss of quality of life, high mortality, and high use of healthcare resources. The aim was to apply the superlearner method to predict osteoporotic hip fractures using administrative claims data and to compare its performance to established methods. Methods: We devided claims data of 288,086 individuals aged 65 years and older without care level into a training (80%) and a validation set (20%). Subsequently, we trained a superlearner algorithm that considered both regression and machine learning algorithms (e.g., support vector machines, RUSBoost) on a large set of clinical risk factors. Mean squared error and measures of discrimination and calibration were employed to assess prediction performance. Results: All algorithms used in the analysis showed similar performance with an AUC ranging from 0.66 to 0.72 in the training and 0.65 to 0.70 in the validation set. Superlearner showed good discrimination in the training set but poorer discrimination and calibration in the validation set. Conclusions: The superlearner achieved similar predictive performance compared to the individual algorithms included. Nevertheless, in the presence of non-linearity and complex interactions, this method might be a flexible alternative to be considered for risk prediction in large datasets. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate ways to improve the generalizability of a deep learning algorithm for identifying glaucomatous optic neuropathy (GON) using a limited number of fundus photographs, as well as the key features being used for classification. Methods: A total of 944 fundus images from Taipei Veterans General Hospital (TVGH) were retrospectively collected. Clinical and demographic characteristics, including structural and functional measurements of the images with GON, were recorded. Transfer learning based on VGGNet was used to construct a convolutional neural network (CNN) to identify GON. To avoid missing cases with advanced GON, an ensemble model was adopted in which a support vector machine classifier would make final classification based on cup-to-disc ratio if the CNN classifier had low-confidence score. The CNN classifier was first established using TVGH dataset, and then fine-tuned by combining the training images of TVGH and Drishti-GS datasets. Class activation map (CAM) was used to identify key features used for CNN classification. Performance of each classifier was determined through area under receiver operating characteristic curve (AUC) and compared with the ensemble model by diagnostic accuracy. Results: In 187 TVGH test images, the accuracy, sensitivity, and specificity of the CNN classifier were 95.0%, 95.7%, and 94.2%, respectively, and the AUC was 0.992 compared to the 92.8% accuracy rate of the ensemble model. For the Drishti-GS test images, the accuracy of the CNN, the fine-tuned CNN and ensemble model was 33.3%, 80.3%, and 80.3%, respectively. The CNN classifier did not misclassify images with moderate to severe diseases. Class-discriminative regions revealed by CAM co-localized with known characteristics of GON. Conclusions: The ensemble model or a fine-tuned CNN classifier may be potential designs to build a generalizable deep learning model for glaucoma detection when large image databases are not available. [ABSTRACT FROM AUTHOR]

Spontaneous prediction of malignant ventricular arrhythmia (MVA) is useful to avoid delay in rescue operations. Recently, researchers have developed several algorithms to predict MVA using various features derived from electrocardiogram (ECG). However, there are several unresolved issues regarding MVA prediction such as the effect of number of ECG features on a prediction remaining unclear, possibility that an alert for occurring MVA may arrive very late and uncertainty in the performance of the algorithm predicting MVA minutes before onset. To overcome the aforementioned problems, this research conducts an in-depth study on the number and types of ECG features that are implemented in a decision tree classifier. In addition, this research also investigates an algorithm's execution time before the occurrence of MVA to minimize delays in warnings for MVA. Lastly, this research aims to study both the sensitivity and specificity of an algorithm to reveal the performance of MVA prediction algorithms from time to time. To strengthen the results of analysis, several classifiers such as support vector machine and naive Bayes are also examined for the purpose of comparison study. There are three phases required to achieve the objectives. The first phase is literature review on existing relevant studies. The second phase deals with design and development of four modules for predicting MVA. Rigorous experiments are performed in the feature selection and classification modules. The results show that eight ECG features with decision tree classifier achieved good prediction performance in terms of execution time and sensitivity. In addition, the results show that the highest percentage for sensitivity and specificity is 95% and 90% respectively, in the fourth 5-minute interval (15.1 minutes–20 minutes) that preceded the onset of an arrhythmia event. Such results imply that the fourth 5-minute interval would be the best time to perform prediction. [ABSTRACT FROM AUTHOR]

1. Remotely tracking distinct behaviours of animals using acceleration data and machine learning has been carried out successfully in several species in captive settings. In order to study the ecology of animals in natural habitats, such behaviour classification models need to be transferred to wild individuals. However, at present, the development of those models usually requires direct observation of the target animals. 2. The goal of this study was to infer the behaviour of wild, free-roaming animals from acceleration data by training behaviour classification models on captive individuals, without the necessity to observe their wild conspecifics. We further sought to develop methods to validate the credibility of the resulting behaviour extrapolations. 3. We trained two machine learning algorithms proposed by the literature, Random Forest (RF) and Support Vector Machine (SVM), on data from captive red foxes (Vulpes vulpes) and later applied them to data from wild foxes. We also tested a new advance for behaviour classification, by applying a moving window to an Artificial Neural Network (ANN). Finally, we investigated four strategies to validate our classification output. 4. While all three machine learning algorithms performed well under training conditions (Kappa values: RF (0.82), SVM (0.78), ANN (0.85)), the established methods, RF and SVM, failed in classifying distinct behaviours when transferred from captive to wild foxes. Behaviour classification with the ANN and a moving window, in contrast, inferred distinct behaviours and showed consistent results for most individuals. 5. Our approach is a substantial improvement over the methods previously proposed in the literature as it generated plausible results for wild fox behaviour. We were able to infer the behaviour of wild animals that have never been observed in the wild and to further illustrate the credibility of the output. This framework is not restricted to foxes but can be applied to infer the behaviour of many other species and thus empowers new advances in behavioural ecology. [ABSTRACT FROM AUTHOR]

Drug induced liver injury (DILI) is one of the key safety concerns in drug development. To assess the likelihood of drug candidates with potential adverse reactions of liver, we propose a compound attributes-based approach to predicting hepatobiliary disorders that are routinely reported to US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS). Specifically, we developed a support vector machine (SVM) model with recursive feature extraction, based on physicochemical and structural properties of compounds as model input. Cross validation demonstrates that the predictive model has a robust performance with averaged 70% of both sensitivity and specificity over 500 trials. An independent validation was performed on public benchmark drugs and the results suggest potential utility of our model for identifying safety alerts. This in silico approach, upon further validation, would ultimately be implemented, together with other in vitro safety assays, for screening compounds early in drug development. [ABSTRACT FROM AUTHOR]

Zanthoxyli Pericarpium (ZP) are the dried ripe peel of Zanthoxylum schinifolium Sieb. et Zucc (ZC) or Zanthoxylum bungeanum Maxim (ZB). It has wide range of uses both medicine and food, and favorable market value. The diverse specifications of components of ZP is exceptional, and the common aims of adulteration for economic profit is conducted. In this work, a novel method for the identification different species of ZP is proposed using convolutional neural networks (CNNs). The data used for the experiment is 5 classes obtained from camera and mobile phones. Firstly, the data considering 2 categories are trained to detect the labels by YOLO. Then, the multiple deep learning including VGG, ResNet, Inception v4, and DenseNet are introduced to identify the different species of ZP (HZB, DZB, OZB, ZA and JZC). In order to assess the performance of CNNs, compared with two traditional identification models including Support Vector Machines (SVM) and Back Propagation (BP). The experimental results demonstrate that the CNN model have a better performance to identify different species of ZP and the highest identification accuracy is 99.35%. The present study is proved to be a useful strategy for the discrimination of different traditional Chinese medicines (TCMs). [ABSTRACT FROM AUTHOR]

Background: Predicting which children will go on to develop mental health symptoms as adolescents is critical for early intervention and preventing future, severe negative outcomes. Although many aspects of a child's life, personality, and symptoms have been flagged as indicators, there is currently no model created to screen the general population for the risk of developing mental health problems. Additionally, the advent of machine learning techniques represents an exciting way to potentially improve upon the standard prediction modelling technique, logistic regression. Therefore, we aimed to I.) develop a model that can predict mental health problems in mid-adolescence II.) investigate if machine learning techniques (random forest, support vector machines, neural network, and XGBoost) will outperform logistic regression. Methods: In 7,638 twins from the Child and Adolescent Twin Study in Sweden we used 474 predictors derived from parental report and register data. The outcome, mental health problems, was determined by the Strengths and Difficulties Questionnaire. Model performance was determined by the area under the receiver operating characteristic curve (AUC). Results: Although model performance varied somewhat, the confidence interval overlapped for each model indicating non-significant superiority for the random forest model (AUC = 0.739, 95% CI 0.708–0.769), followed closely by support vector machines (AUC = 0.735, 95% CI 0.707–0.764). Conclusion: Ultimately, our top performing model would not be suitable for clinical use, however it lays important groundwork for future models seeking to predict general mental health outcomes. Future studies should make use of parent-rated assessments when possible. Additionally, it may not be necessary for similar studies to forgo logistic regression in favor of other more complex methods. [ABSTRACT FROM AUTHOR]

Multiple Sclerosis (MS) progresses at an unpredictable rate, but predictions on the disease course in each patient would be extremely useful to tailor therapy to the individual needs. We explore different machine learning (ML) approaches to predict whether a patient will shift from the initial Relapsing-Remitting (RR) to the Secondary Progressive (SP) form of the disease, using only "real world" data available in clinical routine. The clinical records of 1624 outpatients (207 in the SP phase) attending the MS service of Sant'Andrea hospital, Rome, Italy, were used. Predictions at 180, 360 or 720 days from the last visit were obtained considering either the data of the last available visit (Visit-Oriented setting), comparing four classical ML methods (Random Forest, Support Vector Machine, K-Nearest Neighbours and AdaBoost) or the whole clinical history of each patient (History-Oriented setting), using a Recurrent Neural Network model, specifically designed for historical data. Missing values were handled by removing either all clinical records presenting at least one missing parameter (Feature-saving approach) or the 3 clinical parameters which contained missing values (Record-saving approach). The performances of the classifiers were rated using common indicators, such as Recall (or Sensitivity) and Precision (or Positive predictive value). In the visit-oriented setting, the Record-saving approach yielded Recall values from 70% to 100%, but low Precision (5% to 10%), which however increased to 50% when considering only predictions for which the model returned a probability above a given "confidence threshold". For the History-oriented setting, both indicators increased as prediction time lengthened, reaching values of 67% (Recall) and 42% (Precision) at 720 days. We show how "real world" data can be effectively used to forecast the evolution of MS, leading to high Recall values and propose innovative approaches to improve Precision towards clinically useful values. [ABSTRACT FROM AUTHOR]

The assessment of examination questions is crucial in educational institutes since examination is one of the most common methods to evaluate students' achievement in specific course. Therefore, there is a crucial need to construct a balanced and high-quality exam, which satisfies different cognitive levels. Thus, many lecturers rely on Bloom's taxonomy cognitive domain, which is a popular framework developed for the purpose of assessing students' intellectual abilities and skills. Several works have been proposed to automatically handle the classification of questions in accordance with Bloom's taxonomy. Most of these works classify questions according to specific domain. As a result, there is a lack of technique of classifying questions that belong to the multi-domain areas. The aim of this paper is to present a classification model to classify exam questions based on Bloom's taxonomy that belong to several areas. This study proposes a method for classifying questions automatically, by extracting two features, TFPOS-IDF and word2vec. The purpose of the first feature was to calculate the term frequency-inverse document frequency based on part of speech, in order to assign a suitable weight for essential words in the question. The second feature, pre-trained word2vec, was used to boost the classification process. Then, the combination of these features was fed into three different classifiers; K-Nearest Neighbour, Logistic Regression, and Support Vector Machine, in order to classify the questions. The experiments used two datasets. The first dataset contained 141 questions, while the other dataset contained 600 questions. The classification result for the first dataset achieved an average of 71.1%, 82.3% and 83.7% weighted F1-measure respectively. The classification result for the second dataset achieved an average of 85.4%, 89.4% and 89.7% weighted F1-measure respectively. The finding from this study showed that the proposed method is significant in classifying questions from multiple domains based on Bloom's taxonomy. [ABSTRACT FROM AUTHOR]

Gradients in the elemental composition of a potato leaf tissue (i.e. its ionome) can be linked to crop potential. Because the ionome is a function of genetics and environmental conditions, practitioners aim at fine-tuning fertilization to obtain an optimal ionome based on the needs of potato cultivars. Our objective was to assess the validity of cultivar grouping and predict potato tuber yields using foliar ionomes. The dataset comprised 3382 observations in Québec (Canada) from 1970 to 2017. The first mature leaves from top were sampled at the beginning of flowering for total N, P, K, Ca, and Mg analysis. We preprocessed nutrient concentrations (ionomes) by centering each nutrient to the geometric mean of all nutrients and to a filling value, a transformation known as row-centered log ratios (clr). A density-based clustering algorithm (dbscan) on these preprocessed ionomes failed to delineate groups of high-yield cultivars. We also used the preprocessed ionomes to assess their effects on tuber yield classes (high- and low-yields) on a cultivar basis using k-nearest neighbors, random forest and support vector machines classification algorithms. Our machine learning models returned an average accuracy of 70%, a fair diagnostic potential to detect in-season nutrient imbalance of potato cultivars using clr variables considering potential confounding factors. Optimal ionomic regions of new cultivars could be assigned to the one of the closest documented cultivar. [ABSTRACT FROM AUTHOR]

China experiences frequent landslides, and therefore there is a need for landslide susceptibility maps (LSMs) to effectively analyze and predict regional landslides. However, the traditional methods of producing an LSM are unable to account for different spatial scales, resulting in spatial imbalances. In this study, Zigui-Badong in the Three Gorges Reservoir Area was used as a case study, and data was obtained from remote sensing images, digital elevation model, geological and topographic maps, and landslide surveys. A geographic weighted regression (GWR) was applied to segment the study area into different spatial scales, with three basic principles followed when the GWR model was applied for this propose. As a result, 58 environmental factors were extracted, and 18 factors were selected as LSM factors. Three of the most important factors (channel network basic level, elevation, and distance to river) were used as segmentation factors to segment the study area into 18 prediction regions. The particle swarm optimization (PSO) algorithm was used to optimize the parameters of a support vector machine (SVM) model for each prediction region. All of the prediction regions were merged to construct a GWR-PSO-SVM coupled model and finally, an LSM of the study area was produced. To verify the effectiveness of the proposed method, the outcomes of the GWR-PSO-SVM coupled model and the PSO-SVM coupled model were compared using three evaluation methods: specific category accuracy analysis, overall prediction accuracy analysis, and area under the curve analysis. The results for the GWR-PSO-SVM coupled model for these three evaluation methods were 85.75%, 87.86%, and 0.965, respectively, while the results for the traditional PSO-SVM coupled model were 68.35%, 84.44%, and 0.944, respectively. The method proposed in this study based on a spatial scale segmentation therefore acquired good results. [ABSTRACT FROM AUTHOR]

A citation is deemed as a potential parameter to determine linkage between research articles. The parameter has extensively been employed to form multifarious academic aspects like calculating the impact factor of journals, h-Index of researchers, allocate different research grants, find the latest research trends, etc. The current state-of-the-art contends that all citations are not of equal importance. Based on this argument, the current trend in citation classification community categorizes citations into important and non-important reasons. The community has proposed different approaches to extract important citations such as citation count, context-based, metadata, and textual based approaches. The contemporary state-of-the-art in citation classification community ignores significantly potential features that can play a vital role in citation classification. This research presents a novel approach for binary citation classification by exploiting section-wise in-text citation frequencies, similarity score, and overall citation count-based features. The study also introduces machine learning algorithms based novel approach for assigning appropriate weights to the logical sections of research papers. The weights are allocated to the citations with respect to their sections. To perform the classification, we used three classification techniques, Support Vector Machine, Kernel Linear Regression, and Random Forest. The experiment was performed on two annotated benchmark datasets that contain 465 and 311 citation pairs of research articles respectively. The results revealed that the proposed approach attained an improved value of precision (i.e., 0.84 vs 0.72) from contemporary state-of-the-art approach. [ABSTRACT FROM AUTHOR]

Simulation-based training is increasingly being used for assessment and training of psychomotor skills involved in medicine. The application of artificial intelligence and machine learning technologies has provided new methodologies to utilize large amounts of data for educational purposes. A significant criticism of the use of artificial intelligence in education has been a lack of transparency in the algorithms' decision-making processes. This study aims to 1) introduce a new framework using explainable artificial intelligence for simulation-based training in surgery, and 2) validate the framework by creating the Virtual Operative Assistant, an automated educational feedback platform. Twenty-eight skilled participants (14 staff neurosurgeons, 4 fellows, 10 PGY 4–6 residents) and 22 novice participants (10 PGY 1–3 residents, 12 medical students) took part in this study. Participants performed a virtual reality subpial brain tumor resection task on the NeuroVR simulator using a simulated ultrasonic aspirator and bipolar. Metrics of performance were developed, and leave-one-out cross validation was employed to train and validate a support vector machine in Matlab. The classifier was combined with a unique educational system to build the Virtual Operative Assistant which provides users with automated feedback on their metric performance with regards to expert proficiency performance benchmarks. The Virtual Operative Assistant successfully classified skilled and novice participants using 4 metrics with an accuracy, specificity and sensitivity of 92, 82 and 100%, respectively. A 2-step feedback system was developed to provide participants with an immediate visual representation of their standing related to expert proficiency performance benchmarks. The educational system outlined establishes a basis for the potential role of integrating artificial intelligence and virtual reality simulation into surgical educational teaching. The potential of linking expertise classification, objective feedback based on proficiency benchmarks, and instructor input creates a novel educational tool by integrating these three components into a formative educational paradigm. [ABSTRACT FROM AUTHOR]

Ionizing radiation induces amongst other the most critical type of DNA damage: double-strand breaks (DSBs). Efficient repair of such damage is crucial for cell survival and genomic stability. The analysis of DSB associated foci assays is often performed manually or with automatic systems. Manual evaluation is time consuming and subjective, while most automatic approaches are prone to changes in experimental conditions or to image artefacts. Here, we examined multiple machine learning models, namely a multi-layer perceptron classifier (MLP), linear support vector machine classifier (SVM), complement naive bayes classifier (cNB) and random forest classifier (RF), to correctly classify γH2AX foci in manually labeled images containing multiple types of artefacts. All models yielded reasonable agreements to the manual rating on the training images (Matthews correlation coefficient >0.4). Afterwards, the best performing models were applied on images obtained under different experimental conditions. Thereby, the MLP model produced the best results with an F1 Score >0.9. As a consequence, we have demonstrated that the used approach is sufficient to mimic manual counting and is robust against image artefacts and changes in experimental conditions. [ABSTRACT FROM AUTHOR]

Yield gaps of maize (Zea mays L.) in the smallholder farms of eastern India are outcomes of a complex interplay of climatic variations, soil fertility gradients, socio-economic factors, and differential management intensities. Several machine learning approaches were used in this study to investigate the relative influences of multiple biophysical, socio-economic, and crop management features in determining maize yield variability using several machine learning approaches. Soil fertility status was assessed in 180 farms and paired with the surveyed data on maize yield, socio-economic conditions, and agronomic management. The C&RT relative variable importance plot identified farm size, total labor, soil factors, seed rate, fertilizer, and organic manure as influential factors. Among the three approaches compared for classifying maize yield, the artificial neural network (ANN) yielded the least (25%) misclassification on validation samples. The random forest partial dependence plots revealed a positive association between farm size and maize productivity. Nonlinear support vector machine boundary analysis for the eight top important variables revealed complex interactions underpinning maize yield response. Notably, farm size and total labor synergistically increased maize yield. Future research integrating these algorithms with empirical crop growth models and crop simulation models for ex-ante yield estimations could result in further improvement. [ABSTRACT FROM AUTHOR]

Background: This study evaluated the risk factors for delayed graft function (DGF) in a country where its incidence is high, detailing donor maintenance-related (DMR) variables and using machine learning (ML) methods beyond the traditional regression-based models. Methods: A total of 443 brain dead deceased donor kidney transplants (KT) from two Brazilian centers were retrospectively analyzed and the following DMR were evaluated using predictive modeling: arterial blood gas pH, serum sodium, blood glucose, urine output, mean arterial pressure, vasopressors use, and reversed cardiac arrest. Results: Most patients (95.7%) received kidneys from standard criteria donors. The incidence of DGF was 53%. In multivariable logistic regression analysis, DMR variables did not impact on DGF occurrence. In post-hoc analysis including only KT with cold ischemia time<21h (n = 220), urine output in 24h prior to recovery surgery (OR = 0.639, 95%CI 0.444–0.919) and serum sodium (OR = 1.030, 95%CI 1.052–1.379) were risk factors for DGF. Using elastic net regularized regression model and ML analysis (decision tree, neural network and support vector machine), urine output and other DMR variables emerged as DGF predictors: mean arterial pressure, ≥ 1 or high dose vasopressors and blood glucose. Conclusions: Some DMR variables were associated with DGF, suggesting a potential impact of variables reflecting poor clinical and hemodynamic status on the incidence of DGF. [ABSTRACT FROM AUTHOR]

This paper focuses on the application of machine learning algorithms for predicting spinal abnormalities. As a data preprocessing step, univariate feature selection as a filter based feature selection, and principal component analysis (PCA) as a feature extraction algorithm are considered. A number of machine learning approaches namely support vector machine (SVM), logistic regression (LR), bagging ensemble methods are considered for the diagnosis of spinal abnormality. The SVM, LR, bagging SVM and bagging LR models are applied on a dataset of 310 samples publicly available in Kaggle repository. The performance of classification of abnormal and normal spinal patients is evaluated in terms of a number of factors including training and testing accuracy, recall, and miss rate. The classifier models are also evaluated by optimizing the kernel parameters, and by using the results of receiver operating characteristic (ROC) and precision-recall curves. Results indicate that when 78% data are used for training, the observed training accuracies for SVM, LR, bagging SVM and bagging LR are 86.30%, 85.47%, 86.72% and 85.06%, respectively. On the other hand, the accuracies for the test dataset for SVM, LR, bagging SVM and bagging LR are the same being 86.96%. However, bagging SVM is the most attractive as it has a higher recall value and a lower miss rate compared to others. Hence, bagging SVM is suitable for the classification of spinal patients when applied on the most five important features of spinal samples. [ABSTRACT FROM AUTHOR]

This paper presents a procedure for the patient-specific prediction of epileptic seizures. To this end, a combination of nonnegative matrix factorization (NMF) and smooth basis functions with robust regression is applied to power spectra of intracranial electroencephalographic (iEEG) signals. The resulting time and frequency components capture the dominant information from power spectra, while removing outliers and noise. This makes it possible to detect structure in preictal states, which is used for classification. Linear support vector machines (SVM) with L1 regularization are used to select and weigh the contributions from different number of not equally informative channels among patients. Due to class imbalance in data, synthetic minority over-sampling technique (SMOTE) is applied. The resulting method yields a computationally and conceptually simple, interpretable model of EEG signals of preictal and interictal states, which shows a good performance for the task of seizure prediction on two datasets (the EPILEPSIAE and on the public Epilepsyecosystem dataset). [ABSTRACT FROM AUTHOR]

We investigated whether the integration of machine learning (ML) into MRI interpretation can provide accurate decision rules for the management of suspicious breast masses. A total of 173 consecutive patients with suspicious breast masses upon complementary assessment (BI-RADS IV/V: n = 100/76) received standardized breast MRI prior to histological verification. MRI findings were independently assessed by two observers (R1/R2: 5 years of experience/no experience in breast MRI) using six (semi-)quantitative imaging parameters. Interobserver variability was studied by ICC (intraclass correlation coefficient). A polynomial kernel function support vector machine was trained to differentiate between benign and malignant lesions based on the six imaging parameters and patient age. Ten-fold cross-validation was applied to prevent overfitting. Overall diagnostic accuracy and decision rules (rule-out criteria) to accurately exclude malignancy were evaluated. Results were integrated into a web application and published online. Malignant lesions were present in 107 patients (60.8%). Imaging features showed excellent interobserver variability (ICC: 0.81–0.98) with variable diagnostic accuracy (AUC: 0.65–0.82). Overall performance of the ML algorithm was high (AUC = 90.1%; BI-RADS IV: AUC = 91.6%). The ML algorithm provided decision rules to accurately rule-out malignancy with a false negative rate <1% in 31.3% of the BI-RADS IV cases. Thus, integration of ML into MRI interpretation can provide objective and accurate decision rules for the management of suspicious breast masses, and could help to reduce the number of potentially unnecessary biopsies. [ABSTRACT FROM AUTHOR]

Background: Atrial Fibrillation is the most common arrhythmia worldwide with a global age adjusted prevalence of 0.5% in 2010. Anticoagulation treatment using warfarin or direct oral anticoagulants is effective in reducing the risk of AF-related stroke by approximately two-thirds and can provide a 10% reduction in overall mortality. There has been increased interest in detecting AF due to its increased incidence and the possibility to prevent AF-related strokes. Inexpensive consumer devices which measure the ECG may have the potential to accurately detect AF but do not generally incorporate diagnostic algorithms. Machine learning algorithms have the potential to improve patient outcomes particularly where diagnoses are made from large volumes or complex patterns of data such as in AF. Methods: We designed a novel AF detection algorithm using a de-correlated Lorenz plot of 60 consecutive RR intervals. In order to reduce the volume of data, the resulting images were compressed using a wavelet transformation (JPEG200 algorithm) and the compressed images were used as input data to a Support Vector Machine (SVM) classifier. We used the Massachusetts Institute of Technology (MIT)—Beth Israel Hospital (BIH) Atrial Fibrillation database and the MIT-BIH Arrhythmia database as training data and verified the algorithm performance using RR intervals collected using an inexpensive consumer heart rate monitor device (Polar-H7) in a case-control study. Results: The SVM algorithm yielded excellent discrimination in the training data with a sensitivity of 99.2% and a specificity of 99.5% for AF. In the validation data, the SVM algorithm correctly identified AF in 79/79 cases; sensitivity 100% (95% CI 95.4%-100%) and non-AF in 328/336 cases; specificity 97.6% (95% CI 95.4%-99.0%). Conclusions: An inexpensive wearable heart rate monitor and machine learning algorithm can be used to detect AF with very high accuracy and has the capability to transmit ECG data which could be used to confirm AF. It could potentially be used for intermittent screening or continuously for prolonged periods to detect paroxysmal AF. Further work could lead to cost-effective and accurate estimation of AF burden and improved risk stratification in AF. [ABSTRACT FROM AUTHOR]

Previous work has shown that proteins that have the potential to be vaccine candidates can be predicted from features derived from their amino acid sequences. In this work, we make an empirical comparison across various machine learning classifiers on this sequence-based inference problem. Using systematic cross validation on a dataset of 200 known vaccine candidates and 200 negative examples, with a set of 525 features derived from the AA sequences and feature selection applied through a greedy backward elimination approach, we show that simple classification algorithms often perform as well as more complex support vector kernel machines. The work also includes a novel cross validation applied across bacterial species, i.e. the validation proteins all come from a specific species of bacterium not represented in the training set. We termed this type of validation Leave One Bacteria Out Validation (LOBOV). [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to investigate the effectiveness of using delta-radiomics to predict overall survival (OS) for patients with recurrent malignant gliomas treated by concurrent stereotactic radiosurgery and bevacizumab, and to investigate the effectiveness of machine learning methods for delta-radiomics feature selection and building classification models. Methods: The pre-treatment, one-week post-treatment, and two-month post-treatment T1 and T2 fluid-attenuated inversion recovery (FLAIR) MRI were acquired. 61 radiomic features (intensity histogram-based, morphological, and texture features) were extracted from the gross tumor volume in each image. Delta-radiomics were calculated between the pre-treatment and post-treatment features. Univariate Cox regression and 3 multivariate machine learning methods (L1-regularized logistic regression [L1-LR], random forest [RF] or neural networks [NN]) were used to select a reduced number of features, and 7 machine learning methods (L1-LR, L2-LR, RF, NN, kernel support vector machine [KSVM], linear support vector machine [LSVM], or naïve bayes [NB]) was used to build classification models for predicting OS. The performances of the total 21 model combinations built based on single-time-point radiomics (pre-treatment, one-week post-treatment, and two-month post-treatment) and delta-radiomics were evaluated by the area under the receiver operating characteristic curve (AUC). Results: For a small cohort of 12 patients, delta-radiomics resulted in significantly higher AUC than pre-treatment radiomics (p-value<0.01). One-week/two-month delta-features resulted in significantly higher AUC (p-value<0.01) than the one-week/two-month post-treatment features, respectively. 18/21 model combinations were with higher AUC from one-week delta-features than two-month delta-features. With one-week delta-features, RF feature selector + KSVM classifier and RF feature selector + NN classifier showed the highest AUC of 0.889. Conclusions: The results indicated that delta-features could potentially provide better treatment assessment than single-time-point features. The treatment assessment is substantially affected by the time point for computing the delta-features and the combination of machine learning methods for feature selection and classification. [ABSTRACT FROM AUTHOR]

Disease causing gene identification is considered as an important step towards drug design and drug discovery. In disease gene identification and classification, the main aim is to identify disease genes while identifying non-disease genes are of less or no significant. Hence, this task can be defined as a one-class classification problem. Existing machine learning methods typically take into consideration known disease genes as positive training set and unknown genes as negative samples to build a binary-class classification model. Here we propose a new One-class Classification Support Vector Machines (OCSVM) method to precisely classify candidate disease genes. Our aim is to build a model that concentrate its focus on detecting known disease-causing gene to increase sensitivity and precision. We investigate the impact of our proposed model using a benchmark consisting of the gene expression dataset for Acute Myeloid Leukemia (AML) cancer. Compared with the traditional methods, our experimental result shows the superiority of our proposed method in terms of precision, recall, and F-measure to detect disease causing genes for AML. OCSVM codes and our extracted AML benchmark are publicly available at: https://github.com/imandehzangi/OCSVM. [ABSTRACT FROM AUTHOR]

Diabetes is a large healthcare burden worldwide. There is substantial evidence that lifestyle modifications and drug intervention can prevent diabetes, therefore, an early identification of high risk individuals is important to design targeted prevention strategies. In this paper, we present an automatic tool that uses machine learning techniques to predict the development of type 2 diabetes mellitus (T2DM). Data generated from an oral glucose tolerance test (OGTT) was used to develop a predictive model based on the support vector machine (SVM). We trained and validated the models using the OGTT and demographic data of 1,492 healthy individuals collected during the San Antonio Heart Study. This study collected plasma glucose and insulin concentrations before glucose intake and at three time-points thereafter (30, 60 and 120 min). Furthermore, personal information such as age, ethnicity and body-mass index was also a part of the data-set. Using 11 OGTT measurements, we have deduced 61 features, which are then assigned a rank and the top ten features are shortlisted using minimum redundancy maximum relevance feature selection algorithm. All possible combinations of the 10 best ranked features were used to generate SVM based prediction models. This research shows that an individual's plasma glucose levels, and the information derived therefrom have the strongest predictive performance for the future development of T2DM. Significantly, insulin and demographic features do not provide additional performance improvement for diabetes prediction. The results of this work identify the parsimonious clinical data needed to be collected for an efficient prediction of T2DM. Our approach shows an average accuracy of 96.80% and a sensitivity of 80.09% obtained on a holdout set. [ABSTRACT FROM AUTHOR]

The reaction-diffusion equation serves to model systems in the diffusion regime with sources. Specific applications include diffusion processes in chemical reactions, as well as the propagation of species, diseases, and populations in general. In some of these applications the location of an outbreak, for instance, the source point of a disease or the nest of a vector spreading a virus is important. Also important are the environmental parameters of the domain where the process diffuses, namely the space-dependent diffusion coefficient and the proliferation parameter of the process. Determining both, the location of a source and the environmental parameters, define an inverse problem that in turn, involves a partial differential equation. In this paper we classify the values of these parameters using Support Vector Machines (SVM) trained with numerical solutions of the reaction-diffusion problem. Our set up has accuracy of classifying the outbreak location above 90% and 77% of classifying both, the location and the environmental parameters. The approach presented in our analysis can be directly implemented by measuring the population under study at specific locations in the spatial domain as function of time. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is poised as a transformational approach uniquely positioned to discover the hidden biological interactions for better prediction and diagnosis of complex diseases. In this work, we integrated ML-based models for feature selection and classification to quantify the risk of individual susceptibility to asthma using single nucleotide polymorphism (SNP). Random forest (RF) and recursive feature elimination (RFE) algorithm were implemented to identify the SNPs with high implication to asthma. K-nearest neighbor (kNN) and support vector machine (SVM) algorithms were trained to classify the identified SNPs whether associated with non-asthmatic or asthmatic samples. Feature selection step showed that RF outperformed RFE and the feature importance score derived from RF was consistently high for a subset of SNPs, indicating the robustness of RF in selecting relevant features associated with asthma. Model comparison showed that the integration of RF-SVM obtained the highest model performance with an accuracy, precision, and sensitivity of 62.5%, 65.3%, and 69%, respectively, when compared to the baseline, RF-kNN, and an external MeanDiff-kNN models. Furthermore, results show that the occurrence of asthma can be predicted with an Area under the Curve (AUC) of 0.62 and 0.64 for RF-SVM and RF-kNN models, respectively. This study demonstrates the integration of ML models to augment traditional methods in predicting genetic predisposition to multifactorial diseases such as asthma. [ABSTRACT FROM AUTHOR]

Among different post-translational modifications (PTMs), one of the most important one is the lysine crotonylation in proteins. Its importance cannot be undermined related to different diseases and essential biological practice. The key step for finding the hidden mechanisms of crotonylation along with their occurrence sites is to completely apprehend the mechanism behind this biological process. In previously reported studies, researchers have used different techniques, like position weighted matrix (PWM), support vector machine (SVM), k nearest neighbors (KNN), and many others. However, the maximum prediction accuracy achieved was not such high. To address this, herein, we propose an improved predictor for lysine crotonylation sites named iCrotoK-PseAAC, in which we have incorporated various position and composition relative features along with statistical moments into PseAAC. The results of self-consistency testing were 100% accurate, while the 10-fold cross validation gave 99.0% accuracy. Based on the validation and comparison of model, it is concluded that the iCrotoK-PseAAC is more accurate than the previously proposed models. [ABSTRACT FROM AUTHOR]

We aimed to assess the ability of deep learning (DL) and support vector machine (SVM) to detect a nonperfusion area (NPA) caused by retinal vein occlusion (RVO) with optical coherence tomography angiography (OCTA) images. The study included 322 OCTA images (normal: 148; NPA owing to RVO: 174 [128 branch RVO images and 46 central RVO images]). Training to construct the DL model using deep convolutional neural network (DNN) algorithms was provided using OCTA images. The SVM used a scikit-learn library with a radial basis function kernel. The area under the curve (AUC), sensitivity and specificity for detecting an NPA were examined. We compared the diagnostic ability (sensitivity, specificity and average required time) between the DNN, SVM and seven ophthalmologists. Heat maps were generated. With regard to the DNN, the mean AUC, sensitivity, specificity and average required time for distinguishing RVO OCTA images with an NPA from normal OCTA images were 0.986, 93.7%, 97.3% and 176.9 s, respectively. With regard to SVM, the mean AUC, sensitivity, and specificity were 0.880, 79.3%, and 81.1%, respectively. With regard to the seven ophthalmologists, the mean AUC, sensitivity, specificity and average required time were 0.962, 90.8%, 89.2%, and 700.6 s, respectively. The DNN focused on the foveal avascular zone and NPA in heat maps. The performance of the DNN was significantly better than that of SVM in all parameters (p < 0.01, all) and that of the ophthalmologists in AUC and specificity (p < 0.01, all). The combination of DL and OCTA images had high accuracy for the detection of an NPA, and it might be useful in clinical practice and retinal screening. [ABSTRACT FROM AUTHOR]

Atrial fibrillation (AF) is the most common sustained heart arrhythmia. However, as many cases are asymptomatic, a large proportion of patients remain undiagnosed until serious complications arise. Efficient, cost-effective detection of the undiagnosed may be supported by risk-prediction models relating patient factors to AF risk. However, there exists a need for an implementable risk model that is contemporaneous and informed by routinely collected patient data, reflecting the real-world pathology of AF. This study sought to develop and evaluate novel and conventional statistical and machine learning models for risk-predication of AF. This was a retrospective, cohort study of adults (aged ≥30 years) without a history of AF, listed on the Clinical Practice Research Datalink, from January 2006 to December 2016. Models evaluated included published risk models (Framingham, ARIC, CHARGE-AF), machine learning models, which evaluated baseline and time-updated information (neural network, LASSO, random forests, support vector machines), and Cox regression. Analysis of 2,994,837 individuals (3.2% AF) identified time-varying neural networks as the optimal model achieving an AUROC of 0.827 vs. 0.725, with number needed to screen of 9 vs. 13 patients at 75% sensitivity, when compared with the best existing model CHARGE-AF. The optimal model confirmed known baseline risk factors (age, previous cardiovascular disease, antihypertensive medication usage) and identified additional time-varying predictors (proximity of cardiovascular events, body mass index (both levels and changes), pulse pressure, and the frequency of blood pressure measurements). The optimal time-varying machine learning model exhibited greater predictive performance than existing AF risk models and reflected known and new patient risk factors for AF. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is a powerful tool for identifying and structuring several informative variables for predictive tasks. Here, we investigated how ML algorithms may assist in echocardiographic pulmonary hypertension (PH) prediction, where current guidelines recommend integrating several echocardiographic parameters. In our database of 90 patients with invasively determined pulmonary artery pressure (PAP) with corresponding echocardiographic estimations of PAP obtained within 24 hours, we trained and applied five ML algorithms (random forest of classification trees, random forest of regression trees, lasso penalized logistic regression, boosted classification trees, support vector machines) using a 10 times 3-fold cross-validation (CV) scheme. ML algorithms achieved high prediction accuracies: support vector machines (AUC 0.83; 95% CI 0.73–0.93), boosted classification trees (AUC 0.80; 95% CI 0.68–0.92), lasso penalized logistic regression (AUC 0.78; 95% CI 0.67–0.89), random forest of classification trees (AUC 0.85; 95% CI 0.75–0.95), random forest of regression trees (AUC 0.87; 95% CI 0.78–0.96). In contrast to the best of several conventional formulae (by Aduen et al.), this ML algorithm is based on several echocardiographic signs and feature selection, with estimated right atrial pressure (RAP) being of minor importance. Using ML, we were able to predict pulmonary hypertension based on a broader set of echocardiographic data with little reliance on estimated RAP compared to an existing formula with non-inferior performance. With the conceptual advantages of a broader and unbiased selection and weighting of data our ML approach is suited for high level assistance in PH prediction. [ABSTRACT FROM AUTHOR]

Although electrogastrography (EGG) could be a critical tool in the diagnosis of patients with gastrointestinal (GI) disease, it remains under-utilized. The lack of spatial and temporal resolution using current EGG methods presents a significant roadblock to more widespread usage. Human and preclinical studies have shown that GI myoelectric electrodes can record signals containing significantly more information than can be derived from abdominal surface electrodes. The current study sought to assess the efficacy of multi-electrode arrays, surgically implanted on the serosal surface of the GI tract, from gastric fundus-to-duodenum, in recording myoelectric signals. It also examines the potential for machine learning algorithms to predict functional states, such as retching and emesis, from GI signal features. Studies were performed using ferrets, a gold standard model for emesis testing. Our results include simultaneous recordings from up to six GI recording sites in both anesthetized and chronically implanted free-moving ferrets. Testing conditions to produce different gastric states included gastric distension, intragastric infusion of emetine (a prototypical emetic agent), and feeding. Despite the observed variability in GI signals, machine learning algorithms, including k-nearest neighbors and support vector machines, were able to detect the state of the stomach with high overall accuracy (>75%). The present study is the first demonstration of machine learning algorithms to detect the physiological state of the stomach and onset of retching, which could provide a methodology to diagnose GI diseases and symptoms such as nausea and vomiting. [ABSTRACT FROM AUTHOR]

Objectives: An extended-wavelength diffuse reflectance spectroscopy (EWDRS) technique was evaluated for its ability to differentiate between and classify different skin and tissue types in an in vivo pig model. Materials and methods: EWDRS recordings (450–1550 nm) were made on skin with different degrees of pigmentation as well as on the pig snout and tongue. The recordings were used to train a support vector machine to identify and classify the different skin and tissue types. Results: The resulting EWDRS curves for each skin and tissue type had a unique profile. The support vector machine was able to classify each skin and tissue type with an overall accuracy of 98.2%. The sensitivity and specificity were between 96.4 and 100.0% for all skin and tissue types. Conclusion: EWDRS can be used in vivo to differentiate between different skin and tissue types with good accuracy. Further development of the technique may potentially lead to a novel diagnostic tool for e.g. non-invasive tumor margin delineation. [ABSTRACT FROM AUTHOR]