Search

This paper describes the INSIDE system, a networked robot system designed to allow the use of mobile robots as active players in the therapy of children with autism spectrum disorders (ASD). While a significant volume of work has explored the impact of robots in ASD therapy, most such work comprises remotely operated robots and/or well-structured interaction dynamics. In contrast, the INSIDE system allows for complex, semi-unstructured interaction in ASD therapy while featuring a fully autonomous robot. In this paper we describe the hardware and software infrastructure that supports such rich form of interaction, as well as the design methodology that guided the development of the INSIDE system. We also present some results on the use of our system both in pilot and in a long-term study comprising multiple therapy sessions with children at Hospital Garcia de Orta, in Portugal, highlighting the robustness and autonomy of the system as a whole. [ABSTRACT FROM AUTHOR]

This paper proposes a stochastic shortest path approach to find an optimal sequence of tests to confirm or discard a disease, for any prescribed optimality criterion. The idea is to select the best sequence in which to apply a series of available tests, with a view at reaching a diagnosis with minimum expenditure of resources. The proposed approach derives an optimal policy whereby the decision maker is provided with a test strategy for each a priori probability of disease, aiming to reach posterior probabilities that warrant either immediate treatment or a not-ill diagnosis. [ABSTRACT FROM AUTHOR]

Clustering white matter (WM) tracts from diffusion tensor imaging (DTI) is primarily important for quantitative analysis on pediatric brain development. A recently developed algorithm, density peaks (DP) clustering, demonstrates great robustness to the complex structural variations of WM tracts without any prior templates. Nevertheless, the calculation of densities, the core step of DP, is time consuming especially when the number of WM fibers is huge. In this paper, we propose a fast algorithm that accelerates the density computation about 50 times over the original one. We convert the global calculation for the density as well as critical parameter in the process into local computations, and develop a binary tree structure to orderly store the neighbors for these local computations. Hence, the density computation turns out to be a direct access of the structure, rendering significantly computational saving. Performing experiments on synthetic point data and the JHU-DTI data set and comparing results of our fast DP algorithm and existing clustering methods, we can validate the efficiency and effectiveness of our fast DP algorithm. Finally, we demonstrate the application of the proposed algorithm on the analysis of pediatric WM tract development. [ABSTRACT FROM AUTHOR]

This paper discusses the potential of fuzzy logic methods within medical imaging. Technical advances have produced imaging techniques that can visualize structures and their functions in the living human body. The interpretation of these images plays a prominent role in diagnostic and therapeutic decisions, so physicians must deal with a variety of image processing methods and their applications. This paper describes three different sources of medical imagery that allow the visualization of nerve fibers in the human brain: (1) an algorithm for automatic segmentation of some parts of the thalamus in magnetic resonance images based on the differences in myelin content in various thalamic subnuclei; (2) polarized light for classifying the 3D orientation of the nerve fibers at each point; and (3) confocal laser scanning microscopy (CLSM) for calculating semiquantitative variables for myelin content. Fuzzy logic methods were applied to analyze these pictures from low- to high-level image processing. The solutions presented here are motivated by problems of routine neuroanatomic research demonstrating fuzzy-based methods to be valuable tools in medical image processing. [Copyright &y& Elsevier]

The intuitionistic fuzzy set, as a generation of fuzzy set, can express and process uncertainty much better. Distance measures between intuitionistic fuzzy sets are used to indicate the difference degree between the information carried by intuitionistic fuzzy sets. Although some distance measures have been proposed in previous studies, they can not satisfy the axioms of distance measure, or exist counter-intuitive cases. In this paper, we give a new distance measure between intuitionistic fuzzy sets, which is based on a matrix norm and a strictly increasing (or decreasing) binary function. The new distance measure not only satisfies the axiomatic definition of distance measure, but also overcomes the counter-intuitive cases. It is proved that the new distance measure is reasonable by numerical examples. Moreover, we give the algorithms for pattern recognition and use it to solve medical diagnosis problems. [ABSTRACT FROM AUTHOR]

The brain connections in the different regions demonstrate the characteristics of brain activities. In addition, in various conditions and with neuropsychological disorders, the brain has special patterns in different regions. This paper presents a model to show and compare the connection patterns in different brain regions of children with autism (53 boys and 36 girls) and control children (61 boys and 33 girls). The model is designed by cellular neural networks and it uses the proper features of electroencephalography. The results show that there are significant differences and abnormalities in the left hemisphere, (p < 0.05) at the electrodes AF3, F3, P7, T7, and O1 in the children with autism compared with the control group. Also, the evaluation of the obtained connections values between brain regions demonstrated that there are more abnormalities in the connectivity of frontal and parietal lobes and the relations of the neighboring regions in children with autism. It is observed that the proposed model is able to distinguish the autistic children from the control subjects with an accuracy rate of 95.1% based on the obtained values of CNN using the SVM method. [ABSTRACT FROM AUTHOR]

Lung sounds convey relevant information related to pulmonary disorders, and to evaluate patients with pulmonary conditions, the physician or the doctor uses the traditional auscultation technique. However, this technique suffers from limitations. For example, if the physician is not well trained, this may lead to a wrong diagnosis. Moreover, lung sounds are non-stationary, complicating the tasks of analysis, recognition, and distinction. This is why developing automatic recognition systems can help to deal with these limitations. In this paper, we compare three machine learning approaches for lung sounds classification. The first two approaches are based on the extraction of a set of handcrafted features trained by three different classifiers (support vector machines, k-nearest neighbor, and Gaussian mixture models) while the third approach is based on the design of convolutional neural networks (CNN). In the first approach, we extracted the 12 MFCC coefficients from the audio files then calculated six MFCCs statistics. We also experimented normalization using zero mean and unity variance to enhance accuracy. In the second approach, the local binary pattern (LBP) features are extracted from the visual representation of the audio files (spectrograms). The features are normalized using whitening. The dataset used in this work consists of seven classes (normal, coarse crackle, fine crackle, monophonic wheeze, polyphonic wheeze, squawk, and stridor). We have also experimentally tested dataset augmentation techniques on the spectrograms to enhance the ultimate accuracy of the CNN. The results show that CNN outperformed the handcrafted feature based classifiers. [ABSTRACT FROM AUTHOR]

Diagnostic information regarding the health status of the corneal endothelium may be obtained by analyzing the size and the shape of the endothelial cells in specular microscopy images. Prior to the analysis, the endothelial cells need to be extracted from the image. Up to today, this has been performed manually or semi-automatically. Several approaches to automatic segmentation of endothelial cells exist; however, none of them is perfect. Therefore this paper proposes to perform cell segmentation using a U-Net-based convolutional neural network. Particularly, the network is trained to discriminate pixels located at the borders between cells. The edge probability map outputted by the network is next binarized and skeletonized in order to obtain one-pixel wide edges. The proposed solution was tested on a dataset consisting of 30 corneal endothelial images presenting cells of different sizes, achieving an AUROC level of 0.92. The resulting DICE is on average equal to 0.86, which is a good result, regarding the thickness of the compared edges. The corresponding mean absolute percentage error of cell number is at the level of 4.5% which confirms the high accuracy of the proposed approach. The resulting cell edges are well aligned to the ground truths and require a limited number of manual corrections. This also results in accurate values of the cell morphometric parameters. The corresponding errors range from 5.2% for endothelial cell density, through 6.2% for cell hexagonality to 11.93% for the coefficient of variation of the cell size. [ABSTRACT FROM AUTHOR]

The Golgi Apparatus (GA) is a key organelle for protein synthesis within the eukaryotic cell. The main task of GA is to modify and sort proteins for transport throughout the cell. Proteins permeate through the GA on the ER (Endoplasmic Reticulum) facing side (cis side) and depart on the other side (trans side). Based on this phenomenon, we get two types of GA proteins, namely, cis-Golgi protein and trans-Golgi protein. Any dysfunction of GA proteins can result in congenital glycosylation disorders and some other forms of difficulties that may lead to neurodegenerative and inherited diseases like diabetes, cancer and cystic fibrosis. So, the exact classification of GA proteins may contribute to drug development which will further help in medication. In this paper, we focus on building a new computational model that not only introduces easy ways to extract features from protein sequences but also optimizes classification of trans-Golgi and cis-Golgi proteins. After feature extraction, we have employed Random Forest (RF) model to rank the features based on the importance score obtained from it. After selecting the top ranked features, we have applied Support Vector Machine (SVM) to classify the sub-Golgi proteins. We have trained regression model as well as classification model and found the former to be superior. The model shows improved performance over all previous methods. As the benchmark dataset is significantly imbalanced, we have applied Synthetic Minority Over-sampling Technique (SMOTE) to the dataset to make it balanced and have conducted experiments on both versions. Our method, namely, identification of sub-Golgi Protein Types (isGPT), achieves accuracy values of 95.4%, 95.9% and 95.3% for 10-fold cross-validation test, jackknife test and independent test respectively. According to different performance metrics, isGPT performs better than state-of-the-art techniques. The source code of isGPT, along with relevant dataset and detailed experimental results, can be found at https://github.com/srautonu/isGPT. [ABSTRACT FROM AUTHOR]

Targeted intervention and resource allocation are essential for effective malaria control, particularly in remote areas, with predictive models providing important information for decision making. While a diversity of modeling technique have been used to create predictive models of malaria, no work has made use of Bayesian networks. Bayes nets are attractive due to their ability to represent uncertainty, model time lagged and nonlinear relations, and provide explanations. This paper explores the use of Bayesian networks to model malaria, demonstrating the approach by creating village level models with weekly temporal resolution for Tha Song Yang district in northern Thailand. The networks are learned using data on cases and environmental covariates. Three types of networks are explored: networks for numeric prediction, networks for outbreak prediction, and networks that incorporate spatial autocorrelation. Evaluation of the numeric prediction network shows that the Bayes net has prediction accuracy in terms of mean absolute error of about 1.4 cases for 1 week prediction and 1.7 cases for 6 week prediction. The network for outbreak prediction has an ROC AUC above 0.9 for all prediction horizons. Comparison of prediction accuracy of both Bayes nets against several traditional modeling approaches shows the Bayes nets to outperform the other models for longer time horizon prediction of high incidence transmission. To model spread of malaria over space, we elaborate the models with links between the village networks. This results in some very large models which would be far too laborious to build by hand. So we represent the models as collections of probability logic rules and automatically generate the networks. Evaluation of the models shows that the autocorrelation links significantly improve prediction accuracy for some villages in regions of high incidence. We conclude that spatiotemporal Bayesian networks are a highly promising modeling alternative for prediction of malaria and other vector-borne diseases. [ABSTRACT FROM AUTHOR]

Motivation: Patients under radiotherapy for head-and-neck cancer often suffer of long-term xerostomia, and/or consistent shrinkage of parotid glands. In order to avoid these drawbacks, adaptive therapy can be planned for patients at risk, if the prediction is obtained timely, before or during the early phase of treatment. Artificial intelligence can address the problem, by learning from examples and building classification models. In particular, fuzzy logic has shown its suitability for medical applications, in order to manage uncertain data, and to build transparent rule-based classifiers. In previous works, clinical, dosimetric and image-based features were considered separately, to find different possible predictors of parotid shrinkage. On the other hand, a few works reported possible image-based predictors of xerostomia, while the combination of different types of features has been little addressed.Objective: This paper proposes the application of a novel machine learning approach, based on both statistics and fuzzy logic, aimed at the classification of patients at risk of i) parotid gland shrinkage and ii) 12-months xerostomia. Both problems are addressed with the aim of individuating predictors and models to classify respective outcomes.Methods: Knowledge is extracted from a real dataset of radiotherapy patients, by means of a recently developed method named Likelihood-Fuzzy Analysis, based on the representation of statistical information by fuzzy rule-based models. This method enables to manage heterogeneous variables and missing data, and to obtain interpretable fuzzy models presenting good generalization power (thus high performance), and to measure classification confidence. Numerous features are extracted to characterize patients, coming from different sources, i.e. clinical features, dosimetric parameters, and radiomics-based measures obtained by texture analysis of Computed Tomography images. A learning approach based on the composition of simple models in a more complicated one allows to consider the features separately, in order to identify predictors and models to use when only some data source is available, and obtaining more accurate results when more information can be combined.Results: Regarding parotid shrinkage, a number of good predictors is detected, some already known and confirmed here, and some others found here, in particular among radiomics-based features. A number of models are also designed, some using single features and others involving models composition to improve classification accuracy. In particular, the best model to be used at the initial treatment stage, and another one applicable at the half treatment stage are identified. Regarding 12-months toxicity, some possible predictors are detected, in particular among radiomics-based features. Moreover, the relation between final parotid shrinkage rate and 12-months xerostomia is evaluated. The method is compared to the naïve Bayes classifier, which reveals similar results in terms of classification accuracy and best predictors. The interpretable fuzzy rule-based models are explicitly presented, and the dependence between predictors and outcome is explained, thus furnishing in some cases helpful insights about the considered problems.Conclusion: Thanks to the performance and interpretability of the fuzzy classification method employed, predictors of both parotid shrinkage and xerostomia are detected, and their influence on each outcome is revealed. Moreover, models for predicting parotid shrinkage at initial and half radiotherapy stages are found. [ABSTRACT FROM AUTHOR]

Purpose: Explore how efficient intelligent decision support systems, both easily understandable and straightforwardly implemented, can help modern hospital managers to optimize both bed occupancy and utilization costs.Methods and Materials: This paper proposes a hybrid genetic algorithm-queuing multi-compartment model for the patient flow in hospitals. A finite capacity queuing model with phase-type service distribution is combined with a compartmental model, and an associated cost model is set up. An evolutionary-based approach is used for enhancing the ability to optimize both bed management and associated costs. In addition, a "What-if analysis" shows how changing the model parameters could improve performance while controlling costs. The study uses bed-occupancy data collected at the Department of Geriatric Medicine - St. George's Hospital, London, period 1969-1984, and January 2000.Results: The hybrid model revealed that a bed-occupancy exceeding 91%, implying a patient rejection rate around 1.1%, can be carried out with 159 beds plus 8 unstaffed beds. The same holding and penalty costs, but significantly different bed allocations (156 vs. 184 staffed beds, and 8 vs. 9 unstaffed beds, respectively) will result in significantly different costs (£755 vs. £1172). Moreover, once the arrival rate exceeds 7 patient/day, the costs associated to the finite capacity system become significantly smaller than those associated to an Erlang B queuing model (£134 vs. £947).Conclusion: Encoding the whole information provided by both the queuing system and the cost model through chromosomes, the genetic algorithm represents an efficient tool in optimizing the bed allocation and associated costs. The methodology can be extended to different medical departments with minor modifications in structure and parameterization. [ABSTRACT FROM AUTHOR]

Objective: Antimicrobial stewardship programs have been shown to limit the inappropriate use of antimicrobials. Hospitals are increasingly relying on clinical decision support systems to assist in the demanding prescription reviewing process. In previous work, we have reported on an emerging clinical decision support system for antimicrobial stewardship that can learn new rules supervised by user feedback. In this paper, we report on the evaluation of this system.Methods: The evaluated system uses a knowledge base coupled with a supervised learning module that extracts classification rules for inappropriate antimicrobial prescriptions using past recommendations for dose and dosing frequency adjustments, discontinuation of therapy, early switch from intravenous to oral therapy, and redundant antimicrobial spectrum. Over five weeks, the learning module was deployed alongside the baseline system to prospectively evaluate its ability to discover rules that complement the existing knowledge base for identifying inappropriate prescriptions of piperacillin-tazobactam, a frequently used antimicrobial.Results: The antimicrobial stewardship pharmacists reviewed 374 prescriptions, of which 209 (56% of 374) were identified as inappropriate leading to 43 recommendations to optimize prescriptions. The baseline system combined with the learning module triggered alerts in 270 prescriptions with a positive predictive value of identifying inappropriate prescriptions of 74%. Of these, 240 reviewed prescriptions were identified by the alerts of the baseline system with a positive predictive value of 82% and 105 reviewed prescriptions were identified by the alerts of the learning module with a positive predictive value of 62%. The combined system triggered alerts for all 43 recommendations, resulting in a rate of actionable alerts of 16% (43 recommendations of 270 reviewed alerts); the baseline system triggered alerts for 38 interventions, resulting in a rate of actionable alerts of 16% (38 of 240 reviewed alerts); and the learning module triggered alerts for 17 interventions, resulting in a rate of actionable alerts of 16% (17 of 105 reviewed alerts). The learning module triggered alerts for every inappropriate prescription missed by the knowledge base of the baseline system (n=5).Conclusions: The learning module was able to extract clinically relevant rules for multiple types of antimicrobial alerts. The learned rules were shown to extend the knowledge base of the baseline system by identifying pharmacist interventions that were missed by the baseline system. The learned rules identified inappropriate prescribing practices that were not supported by local experts and were missing from its knowledge base. However, combining the baseline system and the learning module increased the number of false positives. [ABSTRACT FROM AUTHOR]

Objective: This paper aims at developing an automated gastroscopic video summarization algorithm to assist clinicians to more effectively go through the abnormal contents of the video.Methods and Materials: To select the most representative frames from the original video sequence, we formulate the problem of gastroscopic video summarization as a dictionary selection issue. Different from the traditional dictionary selection methods, which take into account only the number and reconstruction ability of selected key frames, our model introduces the similar-inhibition constraint to reinforce the diversity of selected key frames. We calculate the attention cost by merging both gaze and content change into a prior cue to help select the frames with more high-level semantic information. Moreover, we adopt an image quality evaluation process to eliminate the interference of the poor quality images and a segmentation process to reduce the computational complexity.Results: For experiments, we build a new gastroscopic video dataset captured from 30 volunteers with more than 400k images and compare our method with the state-of-the-arts using the content consistency, index consistency and content-index consistency with the ground truth. Compared with all competitors, our method obtains the best results in 23 of 30 videos evaluated based on content consistency, 24 of 30 videos evaluated based on index consistency and all videos evaluated based on content-index consistency.Conclusions: For gastroscopic video summarization, we propose an automated annotation method via similar-inhibition dictionary selection. Our model can achieve better performance compared with other state-of-the-art models and supplies more suitable key frames for diagnosis. The developed algorithm can be automatically adapted to various real applications, such as the training of young clinicians, computer-aided diagnosis or medical report generation. [ABSTRACT FROM AUTHOR]

Objective: Feature selection is a technique widely used in data mining. The aim is to select the best subset of features relevant to the problem being considered. In this paper, we consider feature selection for the classification of gene datasets. Gene data is usually composed of just a few dozen objects described by thousands of features. For this kind of data, it is easy to find a model that fits the learning data. However, it is not easy to find one that will simultaneously evaluate new data equally well as learning data. This overfitting issue is well known as regards classification and regression, but it also applies to feature selection.Methods and Materials: We address this problem and investigate its importance in an empirical study of four feature selection methods applied to seven high-dimensional gene datasets. We chose datasets that are well studied in the literature-colon cancer, leukemia and breast cancer. All the datasets are characterized by a significant number of features and the presence of exactly two decision classes. The feature selection methods used are ReliefF, minimum redundancy maximum relevance, support vector machine-recursive feature elimination and relaxed linear separability.Results: Our main result reveals the existence of positive feature selection bias in all 28 experiments (7 datasets and 4 feature selection methods). Bias was calculated as the difference between validation and test accuracies and ranges from 2.6% to as much as 41.67%. The validation accuracy (biased accuracy) was calculated on the same dataset on which the feature selection was performed. The test accuracy was calculated for data that was not used for feature selection (by so called external cross-validation).Conclusions: This work provides evidence that using the same dataset for feature selection and learning is not appropriate. We recommend using cross-validation for feature selection in order to reduce selection bias. [ABSTRACT FROM AUTHOR]

The purpose of this paper is to provide a review of the current state of fuzzy logic theory in epidemioloy, which is a recent area of research. We present four applications of fuzzy logic theory in epidemic problems, using linguistic fuzzy models, possibility measure, probability of fuzzy events and fuzzy decision making techniques. The results demonstrate that the applications of fuzzy sets in epidemiology is a very promising area of research. The final discussion sets the future stage of fuzzy sets application in epidemiology. [Copyright &y& Elsevier]

This paper develops a methodology for the extraction of multisource brain activity using only single channel recordings of electromagnetic (EM) brain signals. Measured electroencephalogram (EEG) and magnetoencephalogram (MEG) signals are used to demonstrate the utility of the method on extracting multisource activity from a single channel recording. At the heart of the method is dynamical embedding (DE) where first an appropriate embedding matrix is constructed out of a series of delay vectors from the measured signal. The embedding matrix contains the information we require, but in a mixed form which therefore needs to be deconstructed. In particular, we demonstrate how one form of independent component analysis (ICA) performed on the embedding matrix can deconstruct the single channel recording into its underlying informative components. The components are treated as a convenient expansion basis and subjective methods are then used to identify components of interest relevant to the application. The framework has been applied to single channels of both EEG and MEG recordings and is shown to isolate multiple sources of activity which includes: (i) artifactual components such as ocular, electrocardiographic and electrode artefact, (ii) seizure components in epileptic EEG recordings, and (iii) theta band, tumour related, activity in MEG recordings. The results are intuitive and meaningful in a neurophysiological setting. [Copyright &y& Elsevier]

This paper presents an approach to active mining of patient records aimed at discovering patient groups at high risk for coronary heart disease (CHD). The approach proposes active expert involvement in the following steps of the knowledge discovery process: data gathering, cleaning and transformation, subgroup discovery, statistical characterization of induced subgroups, their interpretation, and the evaluation of results. As in the discovery and characterization of risk subgroups, the main risk factors are made explicit, the proposed methodology has high potential for patient screening and early detection of patient groups at risk for CHD. [Copyright &y& Elsevier]

This paper describes a brainstem auditory evoked potentials (BAEPs) detection method based on supervised pattern recognition. A previously used pattern recognition technique relying on cross-correlation with a template was modified in order to include a priori information allowing detection accuracy. Reference is made to the patient’s audiogram and to the latency–intensity (LI) curve with respect to physiological mechanisms. Flexible and adaptive constraints are introduced in the optimization procedure by means of eight rules. Several data samples were used in this study. The determination of parameters was performed through 270 BAEPs from 20 subjects with normal and high audiometric thresholds and through additional BAEPs from 123 normal ears and 14 ears showing prominent wave VI BAEPs. The evaluation of the detection performance was performed in two steps: first, the sensitivity, specificity and accuracy were estimated using 283 BAEPs from 20 subjects showing normal and high audiometric thresholds and secondly, the sensitivity, specificity and accuracy of the detection and the accuracy of the response threshold were estimated using 213 BAEPs from 18 patients in clinic.Taking into account some a priori information, the accuracy in BAEPs detection was enhanced from 76 to 90%. The patient response thresholds were determined with a mean error of 5 dB and a standard deviation error of 8.3 dB. Results were obtained using experimental data; therefore, they are promising for routine use in clinic. [Copyright &y& Elsevier]

The majority of current systems for automatic diagnosis considers the detection of a unique and previously known pathology. Considering specifically the diagnosis of lesions in the small bowel using endoscopic capsule images, very few consider the possible existence of more than one pathology and when they do, they are mainly detection based systems therefore unable to localize the suspected lesions. Such systems do not fully satisfy the medical community, that in fact needs a system that detects any pathology and eventually more than one, when they coexist. In addition, besides the diagnostic capability of these systems, localizing the lesions in the image has been of great interest to the medical community, mainly for training medical personnel purposes. So, nowadays, the inclusion of the lesion location in automatic diagnostic systems is practically mandatory. Multi-pathology detection can be seen as a multi-object detection task and as each frame can contain different instances of the same lesion, instance segmentation seems to be appropriate for the purpose. Consequently, we argue that a multi-pathology system benefits from using the instance segmentation approach, since classification and segmentation modules are both required complementing each other in lesion detection and localization. According to our best knowledge such a system does not yet exist for the detection of WCE pathologies. This paper proposes a multi-pathology system that can be applied to WCE images, which uses the Mask Improved RCNN (MI-RCNN), a new mask subnet scheme which has shown to significantly improve mask predictions of the high performing state-of-the-art Mask-RCNN and PANet systems. A novel training strategy based on the second momentum is also proposed for the first time for training Mask-RCNN and PANet based systems. These approaches were tested using the public database KID, and the included pathologies were bleeding, angioectasias, polyps and inflammatory lesions. Experimental results show significant improvements for the proposed versions, reaching increases of almost 7% over the PANet model when the new proposed training approach was employed. [ABSTRACT FROM AUTHOR]

Subcutaneous Implantable Cardioverter-Defibrillators (S-ICDs) are used for prevention of sudden cardiac death triggered by ventricular arrhythmias. T Wave Over Sensing (TWOS) is an inherent risk with S-ICDs which can lead to inappropriate shocks. A major predictor of TWOS is a high T:R ratio (the ratio between the amplitudes of the T and R waves). Currently, patients' Electrocardiograms (ECGs) are screened over 10 s to measure the T:R ratio to determine the patients' eligibility for S-ICD implantation. Due to temporal variations in the T:R ratio, 10 s is not a long enough window to reliably determine the normal values of a patient's T:R ratio. In this paper, we develop a convolutional neural network (CNN) based model utilising phase space reconstruction matrices to predict T:R ratios from 10-second ECG segments without explicitly locating the R or T waves, thus avoiding the issue of TWOS. This tool can be used to automatically screen patients over a much longer period and provide an in-depth description of the behavior of the T:R ratio over that period. The tool can also enable much more reliable and descriptive screenings to better assess patients' eligibility for S-ICD implantation. [ABSTRACT FROM AUTHOR]

No comprehensive review of Bayesian networks (BNs) in healthcare has been published in the past, making it difficult to organize the research contributions in the present and identify challenges and neglected areas that need to be addressed in the future. This unique and novel scoping review of BNs in healthcare provides an analytical framework for comprehensively characterizing the domain and its current state. A literature search of health and health informatics literature databases using relevant keywords found 3810 articles that were reduced to 123. This was after screening out those presenting Bayesian statistics, meta-analysis or neural networks, as opposed to BNs and those describing the predictive performance of multiple machine learning algorithms, of which BNs were simply one type. Using the novel analytical framework, we show that: (1) BNs in healthcare are not used to their full potential; (2) a generic BN development process is lacking; (3) limitations exist in the way BNs in healthcare are presented in the literature, which impacts understanding, consensus towards systematic methodologies, practice and adoption; and (4) a gap exists between having an accurate BN and a useful BN that impacts clinical practice. This review highlights several neglected issues, such as restricted aims of BNs, ad hoc BN development methods, and the lack of BN adoption in practice and reveals to researchers and clinicians the need to address these problems. To map the way forward, the paper proposes future research directions and makes recommendations regarding BN development methods and adoption in practice. [ABSTRACT FROM AUTHOR]

Due to low tissue contrast, irregular shape, and large location variance, segmenting the objects from different medical imaging modalities (e.g., CT, MR) is considered as an important yet challenging task. In this paper, a novel method is presented for interactive medical image segmentation with the following merits. (1) Its design is fundamentally different from previous pure patch-based and image-based segmentation methods. It is observed that during delineation, the physician repeatedly check the intensity from area inside-object to outside-object to determine the boundary, which indicates that comparison in an inside-out manner is extremely important. Thus, the method innovatively models the segmentation task as learning the representation of bi-directional sequential patches, starting from (or ending in) the given central point of the object. This can be realized by the proposed ConvRNN network embedded with a gated memory propagation unit. (2) Unlike previous interactive methods (requiring bounding box or seed points), the proposed method only asks the physician to merely click on the rough central point of the object before segmentation, which could simultaneously enhance the performance and reduce the segmentation time. (3) The method is utilized in a multi-level framework for better performance. It has been systematically evaluated in three different segmentation tasks, including CT kidney tumor, MR prostate, and PROMISE12 challenge, showing promising results compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Apart from the need for superior accuracy, healthcare applications of intelligent systems also demand the deployment of interpretable machine learning models which allow clinicians to interrogate and validate extracted medical knowledge. Fuzzy rule-based models are generally considered interpretable that are able to reflect the associations between medical conditions and associated symptoms, through the use of linguistic if-then statements. Systems built on top of fuzzy sets are of particular appealing to medical applications since they enable the tolerance of vague and imprecise concepts that are often embedded in medical entities such as symptom description and test results. They facilitate an approximate reasoning framework which mimics human reasoning and supports the linguistic delivery of medical expertise often expressed in statements such as 'weight low' or 'glucose level high' while describing symptoms. This paper proposes an approach by performing data-driven learning of accurate and interpretable fuzzy rule bases for clinical decision support. The approach starts with the generation of a crisp rule base through a decision tree learning mechanism, capable of capturing simple rule structures. The crisp rule base is then transformed into a fuzzy rule base, which forms the input to the framework of adaptive network-based fuzzy inference system (ANFIS), thereby further optimising the parameters of both rule antecedents and consequents. Experimental studies on popular medical data benchmarks demonstrate that the proposed work is able to learn compact rule bases involving simple rule antecedents, with statistically better or comparable performance to those achieved by state-of-the-art fuzzy classifiers. [ABSTRACT FROM AUTHOR]

Despite advances in machine learning-based clinical prediction models, only few of such models are actually deployed in clinical contexts. Among other reasons, this is due to a lack of validation studies. In this paper, we present and discuss the validation results of a machine learning model for the prediction of acute kidney injury in cardiac surgery patients initially developed on the MIMIC-III dataset when applied to an external cohort of an American research hospital. To help account for the performance differences observed, we utilized interpretability methods based on feature importance, which allowed experts to scrutinize model behavior both at the global and local level, making it possible to gain further insights into why it did not behave as expected on the validation cohort. The knowledge gleaned upon derivation can be potentially useful to assist model update during validation for more generalizable and simpler models. We argue that interpretability methods should be considered by practitioners as a further tool to help explain performance differences and inform model update in validation studies. [ABSTRACT FROM AUTHOR]

Automated detection of dynamical change in EEG signals has been a long-standing problem in a wide range of clinic applications. It is essential to extract an effective and accurate EEG rhythm indicator that can reflect the dynamical behavior of a given EEG signal. Time-frequency analysis is a promising method to achieve this end, but existing methods still have limitations in real implementation making this kind of methods still progressive until the present day. In this paper, along the line of ongoing research on time-frequency methods, we present a new method based on graph-based modeling. By virtue of this method, an effective and accurate EEG rhythm indicator can be extracted to characterize the dynamical EEG time series. Together with the extracted EEG rhythm indicator, an automatic analysis of continuous monitoring of EEG signal, is developed by means of a null hypothesis testing to inspect whether an EEG change occurs or not during a monitoring period. The proposed framework is applied to both simulated data and real signals respectively to validate its effectiveness. Experimental results, together with theoretical interpretation and discussions, suggest its promising potentials in practice. [ABSTRACT FROM AUTHOR]

In this paper, we embed two types of attention modules in the dilated fully convolutional network (FCN) to solve biomedical image segmentation tasks efficiently and accurately. Different from previous work on image segmentation through multiscale feature fusion, we propose the fully convolutional attention network (FCANet) to aggregate contextual information at long-range and short-range distances. Specifically, we add two types of attention modules, the spatial attention module and the channel attention module, to the Res2Net network, which has a dilated strategy. The features of each location are aggregated through the spatial attention module, so that similar features promote each other in space size. At the same time, the channel attention module treats each channel of the feature map as a feature detector and emphasizes the channel dependency between any two channel maps. Finally, we weight the sum of the output features of the two types of attention modules to retain the feature information of the long-range and short-range distances, to further improve the representation of the features and make the biomedical image segmentation more accurate. In particular, we verify that the proposed attention module can seamlessly connect to any end-to-end network with minimal overhead. We perform comprehensive experiments on three public biomedical image segmentation datasets, i.e., the Chest X-ray collection, the Kaggle 2018 data science bowl and the Herlev dataset. The experimental results show that FCANet can improve the segmentation effect of biomedical images. The source code models are available at https://github.com/luhongchun/FCANet. [ABSTRACT FROM AUTHOR]

Patients with advanced cancer are burdened physically and psychologically, so there is an urgent need to pay more attention to their health-related quality of life (HRQOL). With an expected clinical endpoint prediction, over-treatment can be effectively eliminated by the means of palliative care at the right time. This paper develops a deep learning based approach for cancer clinical endpoint prediction based on patient's electronic health records (EHR). Due to the pervasive existence of categorical information in EHR, it brings unavoidably obstacles to the effective numerical learning algorithms. To address this issue, we propose a novel cross-field categorical attributes embedding (CCAE) model to learn a vectorized representation for cancer patients in attribute-level by orders, in which the strong semantic coupling among categorical variables are well exploited. By transforming the order-dependency modeling into a sequence learning task in an ingenious way, recurrent neural network is adopted to capture the semantic relevance among multi-order representations. Experimental results from the SEER-Medicare EHR dataset have illustrated that the proposed model can achieve competitive prediction performance compared with other baselines. [ABSTRACT FROM AUTHOR]

Regular medical records are useful for medical practitioners to analyze and monitor patient's health status especially for those with chronic disease. However, such records are usually incomplete due to unpunctuality and absence of patients. In order to resolve the missing data problem over time, tensor-based models have been developed for missing data imputation in recent papers. This approach makes use of the low-rank tensor assumption for highly correlated data in a short-time interval. Nevertheless, when the time intervals are long, data correlation may not be high between consecutive time stamps so that such assumption is not valid. To address this problem, we propose to decompose matrices with missing data over time into their latent factors. Then, the locally linear constraint is imposed on the latent factors for temporal matrix completion. By using three publicly available medical datasets and two medical datasets collected from Prince of Wales Hospital in Hong Kong, experimental results show that the proposed algorithm achieves the best performance compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The most severe form of kidney disease, End-Stage Renal Disease (ESRD) is treated with various forms of dialysis - artificial blood cleansing. Dialysis patients suffer many health burdens including high mortality and hospitalization rates, and symptomatic anemia: a low red blood cell count as indicated by a low hemoglobin (Hgb) level. ESRD-induced anemia is treated, with variable patient response, by erythropoiesis stimulating agents (ESAs): expensive injectable medications typically administered during dialysis sessions. The dosing protocol is typically a population level protocol based on original clinical trials, the use of which often results in Hgb cycling. This cycling phenomenon occurs primarily due to the mismatch in the time between dosing decisions and the time it takes for the effects of a dosing change to be fully realized. In this paper we develop a recurrent neural network approach that uses historic data together with future ESA and iron dosing data to predict the 1, 2, and 3 month Hgb levels of patients with ESRD-induced anemia. The results of extensive experimentation indicate that this approach generates predictions that are clinically relevant: the mean absolute error of the predictions is comparable to estimates of the intra-individual variability of the laboratory test for Hgb. [ABSTRACT FROM AUTHOR]

Objective: According to a meta-analysis of 7 studies, the median number of patients with at least one adverse event during the surgery is 14.4%, and a third of those adverse events were preventable. The occurrence of adverse events forces surgeons to implement corrective strategies and, thus, deviate from the standard surgical process. Therefore, it is clear that the automatic identification of adverse events is a major challenge for patient safety. In this paper, we have proposed a method enabling us to identify such deviations. We have focused on identifying surgeons' deviations from standard surgical processes due to surgical events rather than anatomic specificities. This is particularly challenging, given the high variability in typical surgical procedure workflows.Methods: We have introduced a new approach designed to automatically detect and distinguish surgical process deviations based on multi-dimensional non-linear temporal scaling with a hidden semi-Markov model using manual annotation of surgical processes. The approach was then evaluated using cross-validation.Results: The best results have over 90% accuracy. Recall and precision for event deviations, i.e. related to adverse events, are respectively below 80% and 40%. To understand these results, we have provided a detailed analysis of the incorrectly-detected observations.Conclusion: Multi-dimensional non-linear temporal scaling with a hidden semi-Markov model provides promising results for detecting deviations. Our error analysis of the incorrectly-detected observations offers different leads in order to further improve our method.Significance: Our method demonstrated the feasibility of automatically detecting surgical deviations that could be implemented for both skill analysis and developing situation awareness-based computer-assisted surgical systems. [ABSTRACT FROM AUTHOR]

Adverse drug events (ADEs) may occur and lead to severe consequences for the public, even though clinical trials are conducted in the stage of pre-market. Computational methods are still needed to fulfil the task of pharmacosurveillance. In post-market surveillance, the spontaneous reporting system (SRS) has been widely used to detect suspicious associations between medicines and ADEs. However, the passive mechanism of SRS leads to the hysteresis in ADE detection by SRS based methods, not mentioning the acknowledged problem of under-reporting and duplicate reporting in SRS. Therefore, there is a growing demand for other complementary methods utilising different types of healthcare data to assist with global pharmacosurveillance. Among those data sources, prescription data is of proved usefulness for pharmacosurveillance. However, few works have used prescription data for signalling ADEs. In this paper, we propose a data-driven method to discover medicines that are responsible for a given ADE purely from prescription data. Our method uses a logistic regression model to evaluate the associations between up to hundreds of suspected medicines and an ADE spontaneously and selects the medicines possessing the most significant associations via Lasso regularisation. To prepare data for training the logistic regression model, we adapt the design of the case-crossover study to construct case time and control time windows for the extraction of medicine use information. While the case time window can be readily determined, we propose several criteria to select the suitable control time windows providing the maximum power of comparisons. In order to address confounding situations, we have considered diverse factors in medicine utilisation in terms of the temporal effect of medicine and the frequency of prescription, as well as the individual effect of patients on the occurrence of an ADE. To assess the performance of the proposed method, we conducted a case study with a real-world prescription dataset. Validated by the existing domain knowledge, our method successfully traced a wide range of medicines that are potentially responsible for the ADE. Further experiments were also carried out according to a recognised gold standard, our method achieved a sensitivity of 65.9% and specificity of 96.2%. [ABSTRACT FROM AUTHOR]

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays an important role in diagnosis and grading of brain tumors. Although manual DCE biomarker extraction algorithms boost the diagnostic yield of DCE-MRI by providing quantitative information on tumor prognosis and prediction, they are time-consuming and prone to human errors. In this paper, we propose a fully-automated, end-to-end system for DCE-MRI analysis of brain tumors. Our deep learning-powered technique does not require any user interaction, it yields reproducible results, and it is rigorously validated against benchmark and clinical data. Also, we introduce a cubic model of the vascular input function used for pharmacokinetic modeling which significantly decreases the fitting error when compared with the state of the art, alongside a real-time algorithm for determination of the vascular input region. An extensive experimental study, backed up with statistical tests, showed that our system delivers state-of-the-art results while requiring less than 3 min to process an entire input DCE-MRI study using a single GPU. [ABSTRACT FROM AUTHOR]

The decompressive laminectomy is a common operation for treatment of lumbar spinal stenosis. The tools for grinding and drilling are used for fenestration and internal fixation, respectively. The state recognition is one of the main technologies in robot-assisted surgery, especially in tele-surgery, because surgeons have limited perception during remote-controlled robot-assisted surgery. The novelty of this paper is that a state recognition system is proposed for the robot-assisted tele-surgery. By combining the learning methods and traditional methods, the robot from the slave-end can think about the current operation state like a surgeon, and provide more information and decision suggestions to the master-end surgeon, which aids surgeons work safer in tele-surgery. For the fenestration, we propose an image-based state recognition method that consists a U-Net derived network, grayscale redistribution and dynamic receptive field assisting in controlling the grinding process to prevent the grinding-bit from crossing the inner edge of the lamina to damage the spinal nerves. For the internal fixation, we propose an audio and force-based state recognition method that consists signal features extraction methods, LSTM-based prediction and information fusion assisting in monitoring the drilling process to prevent the drilling-bit from crossing the outer edge of the vertebral pedicle to damage the spinal nerves. Several experiments are conducted to show the reliability of the proposed system in robot-assisted surgery. [ABSTRACT FROM AUTHOR]