Needle electromyography (EMG) is a common technique used in clinical neurophysiology to record the electrical activity of muscles at different levels of activation. It can be used to diagnose... Show moreNeedle electromyography (EMG) is a common technique used in clinical neurophysiology to record the electrical activity of muscles at different levels of activation. It can be used to diagnose various neurological/muscular disorders, as the EMG signals of patients with both nerve diseases (neuropathies) andmuscle diseases (myopathies) differ from the signal in healthy controls. A major drawback of this examination is that it relies on visual inspection and as such, it is highly subjective and prone to errors. Based on EMG time series of 65 individuals (40 with ALS/IBM and 25 healthy), we aim to develop an automated machine-learning pipeline for the classification of EMG recordings of muscles in either disease or healthy (muscle-level). The automated pipeline consists of feature extraction, feature selection, modelling algorithm, and optimization, in which the most significant features are automatically selected from the feature space and the hyperparameters of the model are optimized by a Bayesian technique as part of the automatedapproach. Aside from the muscle-level approach, we also explore a patient-level approach, which uses the output of the muscle-level automated pipeline in a post-processing manner to classify patients in being either disease or healthy, based on their muscle recordings. The resulting two approaches yield an AUC scoreof 81.7% (muscle-level) and 81.5% (patient-level), indicating that such approaches can assist clinicians in diagnosing if a patient has a neuropathy/myopathy or is healthy. Show less
After the recent groundbreaking results of AlphaGo and AlphaZero, we have seen strong interests in deep reinforcement learning and artificial general intelligence (AGI) in game playing. However,... Show moreAfter the recent groundbreaking results of AlphaGo and AlphaZero, we have seen strong interests in deep reinforcement learning and artificial general intelligence (AGI) in game playing. However, deep learning is resource-intensive and the theory is not yet well developed. For small games, simple classical table-based Q-learning might still be the algorithm of choice. General Game Playing (GGP) provides a good testbed for reinforcement learning to research AGI. Q-learning is one of the canonical reinforcement learning methods, and has been used by (Banerjee & Stone, IJCAI 2007) in GGP. In this paper we implement Q-learning in GGP for three small-board games (Tic-Tac-Toe, Connect Four, Hex), to allow comparison to Banerjee et al. We find that Q-learning converges to a high win rate in GGP. For the ϵ" role="presentation" style="display: inline-table; line-height: normal; letter-spacing: normal; word-spacing: normal; overflow-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; border-width: 0px; border-style: initial; position: relative;">ϵ-greedy strategy, we propose a first enhancement, the dynamic ϵ" role="presentation" style="display: inline-table; line-height: normal; letter-spacing: normal; word-spacing: normal; overflow-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; border-width: 0px; border-style: initial; position: relative;">ϵ algorithm. In addition, inspired by (Gelly & Silver, ICML 2007) we combine online search (Monte Carlo Search) to enhance offline learning, and propose QM-learning for GGP. Both enhancements improve the performance of classical Q-learning. In this work, GGP allows us to show, if augmented by appropriate enhancements, that classical table-based Q-learning can perform well in small games. Show less
Kriging or Gaussian Process Regression is applied in many fields as a non-linear regression model as well as a surrogate model in the field of evolutionary computation. However, the computational... Show moreKriging or Gaussian Process Regression is applied in many fields as a non-linear regression model as well as a surrogate model in the field of evolutionary computation. However, the computational and space complexity of Kriging, that is cubic and quadratic in the number of data points respectively, becomes a major bottleneck with more and more data available nowadays. In this paper, we propose a general methodology for the complexity reduction, called cluster Kriging, where the whole data set is partitioned into smaller clusters and multiple Kriging models are built on top of them. In addition, four Kriging approximation algorithms are proposed as candidate algorithms within the new framework. Each of these algorithms can be applied to much larger data sets while maintaining the advantages and power of Kriging. The proposed algorithms are explained in detail and compared empirically against a broad set of existing state-of-the-art Kriging approximation methods on a well-defined testing framework. According to the empirical study, the proposed algorithms consistently outperform the existing algorithms. Moreover, some practical suggestions are provided for using the proposed algorithms. Show less
Shao, S.; Li, B.; Cautun, M.; Wang, H.; Wang, J. 2019
ImportanceAlthough IgA nephropathy (IgAN) is the most common glomerulonephritis in the world, there is no validated tool to predict disease progression. This limits patient-specific risk... Show moreImportanceAlthough IgA nephropathy (IgAN) is the most common glomerulonephritis in the world, there is no validated tool to predict disease progression. This limits patient-specific risk stratification and treatment decisions, clinical trial recruitment, and biomarker validation. ObjectiveTo derive and externally validate a prediction model for disease progression in IgAN that can be applied at the time of kidney biopsy in multiple ethnic groups worldwide. Design, Setting, and ParticipantsWe derived and externally validated a prediction model using clinical and histologic risk factors that are readily available in clinical practice. Large, multi-ethnic cohorts of adults with biopsy-proven IgAN were included from Europe, North America, China, and Japan. Main Outcomes and MeasuresCox proportional hazards models were used to analyze the risk of a 50% decline in estimated glomerular filtration rate (eGFR) or end-stage kidney disease, and were evaluated using the R-D(2) measure, Akaike information criterion (AIC), C statistic, continuous net reclassification improvement (NRI), integrated discrimination improvement (IDI), and calibration plots. ResultsThe study included 3927 patients; mean age, 35.4 (interquartile range, 28.0-45.4) years; and 2173 (55.3%) were men. The following prediction models were created in a derivation cohort of 2781 patients: a clinical model that included eGFR, blood pressure, and proteinuria at biopsy; and 2 full models that also contained the MEST histologic score, age, medication use, and either racial/ethnic characteristics (white, Japanese, or Chinese) or no racial/ethnic characteristics, to allow application in other ethnic groups. Compared with the clinical model, the full models with and without race/ethnicity had better R-D(2) (26.3% and 25.3%, respectively, vs 20.3%) and AIC (6338 and 6379, respectively, vs 6485), significant increases in C statistic from 0.78 to 0.82 and 0.81, respectively (Delta C, 0.04; 95% CI, 0.03-0.04 and Delta C, 0.03; 95% CI, 0.02-0.03, respectively), and significant improvement in reclassification as assessed by the NRI (0.18; 95% CI, 0.07-0.29 and 0.51; 95% CI, 0.39-0.62, respectively) and IDI (0.07; 95% CI, 0.06-0.08 and 0.06; 95% CI, 0.05-0.06, respectively). External validation was performed in a cohort of 1146 patients. For both full models, the C statistics (0.82; 95% CI, 0.81-0.83 with race/ethnicity; 0.81; 95% CI, 0.80-0.82 without race/ethnicity) and R-D(2) (both 35.3%) were similar or better than in the validation cohort, with excellent calibration. Conclusions and RelevanceIn this study, the 2 full prediction models were shown to be accurate and validated methods for predicting disease progression and patient risk stratification in IgAN in multi-ethnic cohorts, with additional applications to clinical trial design and biomarker research. Show less
surface. These results unveiled the dynamic interactions between amended biochar and soil oxide minerals which can significantly affect the immobilization pathways of metals in contaminated soils.
Identifying speakers by their spoken output is a specialist task for forensic investigators. In the present study we focused on cross-linguistic speaker (Chinese, English, Dutch) identification... Show moreIdentifying speakers by their spoken output is a specialist task for forensic investigators. In the present study we focused on cross-linguistic speaker (Chinese, English, Dutch) identification based on (components of) English stops and fricatives, /p, b, t, d, k, g/ and the fricatives /f, v, θ, ð, s, z, ʃ, ʒ/. English noise bursts’ contribution to native language identification is presented and the special tokens which contribute the most were analyzed. Show less
Although the class-imbalance classification problem has caught a huge amount of attention, hyperparameter optimisation has not been studied in detail in this field. Both classification algorithms... Show moreAlthough the class-imbalance classification problem has caught a huge amount of attention, hyperparameter optimisation has not been studied in detail in this field. Both classification algorithms and resampling techniques involve some hyperparameters that can be tuned. This paper sets up several experiments and draws the conclusion that, compared to using default hyperparameters, applying hyperparameter optimisation for both classification algorithms and resampling approaches can produce the best results for classifying the imbalanced datasets. Moreover, this paper shows that data complexity, especially the overlap between classes, has a big impact on the potential improvement that can be achieved through hyperparameter optimisation. Results of our experiments also indicate that using resampling techniques cannot improve the performance for some complex datasets, which further emphasizes the importance of analyzing data complexity before dealing with imbalanced datasets. Show less
Continuous optimization is never easy: the exact solution is always a luxury demand and the theory of it is not always analytical and elegant. Continuous optimization, in practice, is... Show moreContinuous optimization is never easy: the exact solution is always a luxury demand and the theory of it is not always analytical and elegant. Continuous optimization, in practice, is essentially about the efficiency: how to obtain the solution with same quality using as minimal resources (e.g., CPU time or memory usage) as possible? In this thesis, the number of function evaluations is considered as the most important resource to save. To achieve this goal, various efforts have been implemented and applied successfully. One research stream focuses on the so-called stochastic variation (mutation) operator, which conducts an (local) exploration of the search space. The efficiency of those operator has been investigated closely, which shows a good stochastic variation should be able to generate a good coverage of the local neighbourhood around the current search solution. This thesis contributes on this issue by formulating a novel stochastic variation that yields good space coverage. Show less
We continue recent work on the definition of multimodality in multiobjective optimization (MO) and the introduction of a test bed for multimodal MO problems. This goes beyond well-known diversity... Show moreWe continue recent work on the definition of multimodality in multiobjective optimization (MO) and the introduction of a test bed for multimodal MO problems. This goes beyond well-known diversity maintenance approaches but instead focuses on the landscape topology induced by the objective functions. More general multimodal MO problems are considered by allowing ellipsoid contours for single-objective subproblems. An experimental analysis compares two MO algorithms, one that explicitly relies on hypervolume gradient approximation, and one that is based on local search, both on a selection of generated example problems. We do not focus on performance but on the interaction induced by the problems and algorithms, which can be described by means of specific characteristics explicitly designed for the multimodal MO setting. Furthermore, we widen the scope of our analysis by additionally applying visualization techniques in the decision space. This strengthens and extends the foundations for Exploratory Landscape Analysis (ELA) in MO. Show less
Noordam, R.; Bos, M.M.; Wang, H.; Mook-Kanamori, D.; Heemst, D. van; Redline, S. 2018
The Chinese medicine Qiliqiangxin (QL) has been shown to have a protective role in heart failure. Here, we explore the underlying working mechanism of the key therapeutic component in QL using a... Show moreThe Chinese medicine Qiliqiangxin (QL) has been shown to have a protective role in heart failure. Here, we explore the underlying working mechanism of the key therapeutic component in QL using a rat model of heart failure. Heart failure after myocardial infarction was induced surgically and confirmed using echocardiography; a separate group of rats underwent sham surgery. The rats with heart failure were randomly assigned to receive QL, the angiotensin-converting enzyme inhibitor benazepril, or placebo groups. Blood samples were collected from the rats at four time points for up to 8 weeks and used for biochemical analysis and mass spectrometry‒based metabolomics profiling. In total, we measured nine well-known biochemical parameters of heart failure and 147 metabolites. In the rats with heart failure, QL significantly improved these biochemical parameters and metabolomics profiles, significantly increasing the cardioprotective parameter angiopoietin-like 4 and significantly lowering inflammation-related oxylipins and lysophosphatidic acids compared to benazepril. Mechanistically, QL may improve outcome in heart failure by controlling inflammatory process and cardiac hypertrophy. Clinical studies should be designed in order to investigate these putative mechanisms in patients. Show less
Niu, K.; Xu, Y.; Wang, H.; Ye, R.; Xin, H.L.; Lin, F.; ... ; Zheng, H. 2017
Two hypotheses have been advanced in the recent literature with respect to the so-called Interlanguage Speech Intelligibility Benefit (ISIB): a nonnative speaker will be better understood by a... Show moreTwo hypotheses have been advanced in the recent literature with respect to the so-called Interlanguage Speech Intelligibility Benefit (ISIB): a nonnative speaker will be better understood by a another nonnative listener than a native speaker of the target language will be (a) only when the nonnatives share the same native language (matched interlanguage) or (b) even when the nonnatives have different mother tongues (non-matched interlanguage). Based on a survey of published experimental materials, the present article will demonstrate that both the restricted (a) and the generalized (b) hypotheses are false when the ISIB effect is evaluated in terms of absolute intelligibility scores.We will then propose a simple way to compute a relative measure for the ISIB (R-ISIB), which we claim is a more insightful way of evaluating the interlanguage benefit, and test the hypotheses in relative (R-ISIB) terms on the same literature data. We then find that our R-ISIB measure only supports the more restricted hypothesis (a) while rejecting the more general hypothesis (b). This finding shows that the native language shared by the interactants biases the listener toward interpreting sounds in terms of the phonology of the shared mother tongue. Show less
We determined the mutual intelligibility Mandarin Chinese, Dutch and American speakers of English in all nine logically possible combinations of speaker and listener native language backgrounds.... Show moreWe determined the mutual intelligibility Mandarin Chinese, Dutch and American speakers of English in all nine logically possible combinations of speaker and listener native language backgrounds. Designated speakers (one male, one female per language group) were selected from larger sets of 20 speakers so as to be optimally representative of their peer groups. All non-native speakers and listeners were university students who did not specialize in English and had never lived in an English speaking community. Intelligibility was tested in separate tests targeting vowels, onset consonants, onset consonant clusters, words in syntactically correct but semantically empty sentences (SUS test), and words in meaningful sentences in which they appeared in either low or high predictability contexts. We test the hypotheses that mutual intelligibility between speaker and listener is better as (i) their native languages resemble each other more, and (ii) if speaker and listener share the same native language. In order to test the second hypothesis we propose a new method for quantifying the so-called interlanguage speech intelligibility benefit (ISIB). Show less
