Comparative evaluations of the Monte Carlo-based light propagation simulation packages for optical imaging

Monte Carlo simulation of light propagation in turbid medium has been studied for years.A number of software packages have been developed to handle with such issue.However,it is hard to compare these simulation packages,especially for tissues with complex heterogeneous structures.Here,we first designed a group of mesh datasets generated by Iso2Mesh software,and used them to cross-validate the accuracy and to evaluate the performance of four Monte Carlo-based simulation packages,including Monte Carlo model of steady-state light transport in multi-layered tissues(MCML),tetrahedron-based inhomogeneous Monte Carlo optical simulator(TIMOS),Molecular Optical Simulation Environment(MOSE),and Mesh-based Monte Carlo(MMC).The performance of each package was evaluated based on the designed mesh datasets.The merits and demerits of each package were also discussed.Comparative results showed that the TIMOS package provided the best performance,which proved to be a reliable,efficient,and stable MC simulation package for users.

A comparative evaluation of 3 kinds of cecocolon urinary reservoir

Objective To evaluate and compare 3 kinds of cecocloln urinary reservoir. Methods Perm pouch, indiana pouch or detenial cecocolon reservoir has been performed for 37 patients. The outcome was evaluated by urodynamic and clinically. Results Good continence has been achieved in all patients with an appendiceal continence mechanism. The capacity and intrareservoir pressure of detenial cecocolon was significantly different from penn or indiana pouch 3 months postoperatively (P【0.01) whereas the difference became in no significant 12 months postoperativelly (P】0. 05). The intrareservoir pressure on contraction was lower with penn pouch. Conclusion The penn or modified indiana pouch is better and more ideal as a urinary reservoir whereas the detenial cecocolon pouch is technically simpler. 6 refs,1 tab.

Comparative Evaluation of Machine Learning Models and Input Feature Space for Non-intrusive Load Monitoring

Recent advancement in computational capabilities has accelerated the research and development of non-intrusive load disaggregation.Non-intrusive load monitoring(NILM)offers many promising applications in the context of energy efficiency and conservation.Load classification is a key component of NILM that relies on different artificial intelligence techniques,e.g.,machine learning.This study employs different machine learning models for load classification and presents a comprehensive performance evaluation of the employed models along with their comparative analysis.Moreover,this study also analyzes the role of input feature space dimensionality in the context of classification performance.For the above purposes,an event-based NILM methodology is presented and comprehensive digital simulation studies are carried out on a low sampling real-world electricity load acquired from four different households.Based on the presented analysis,it is concluded that the presented methodology yields promising results and the employed machine learning models generalize well for the invisible diverse testing data.The multi-layer perceptron learning model based on the neural network approach emerges as the most promising classifier.Furthermore,it is also noted that it significantly facilitates the classification performance by reducing the input feature space dimensionality.

Comparative evaluation of a rapid diagnostic test,an antibody ELISA,and a pLDH ELISA in detecting asymptomatic malaria parasitaemia in blood donors in Buea,Cameroon

Background:In malaria endemic areas,infected blood donors serve as a source of infection to blood recipients,which may adversely affect their prognosis.This necessitates the proper screening of blood to be used for transfusion in these areas.The purpose of this study was to determine the prevalence of malaria parasitaemia in blood donors in Buea,Cameroon,and to evaluate the performance of a rapid diagnostic test(RDT),a malaria antibody enzyme-linked immunosorbent assay(ELISA),and a Plasmodium lactate dehydrogenase(pLDH)ELISA in the detection of asymptomatic malaria parasitaemia in the target population.Methods:In a prospective study conducted between September 2015 and June 2016,1240 potential blood donors were enrolled.The donors were screened for malaria parasites using Giemsa microscopy(GM)and a RDT.A sub-sample of 184 samples,comprising 88 positive and 96 negative samples,were selected for the evaluation of the pLDH ELISA and the antibody ELISA.The chi-square test and correlation analysis were performed as part of the statistical analyses.The statistical significance cut-off was set at P<0.05.Results:The prevalence of malaria parasitaemia in this study was found to be 8.1%(95%CI:6.6-9.7).The prevalence was not observed to be dependent on the age or sex of the participants.The RDT had a sensitivity(88.0%),specificity(99.1%),and negative predictive value(99.0%)higher than the ELISAs.The performance of the pLDH ELISA,which demonstrated the highest positive predictive value(91.6%),was generally comparable to the RDT.The sensitivity was lowest with the antibody ELISA(69.9%),which also demonstrated the highest false positive and false negative rates.The detection threshold for the pLDH(three parasites/μl)was lower compared to the RDT(50-60 parasites/μl).Non-significant positive correlations were observed between the parasite density and the pLDH titers and malaria antibody titers.Conclusions:Overall,the RDT and the pLDH ELISA demonstrated a perfectly correlated agreement with GM,meanwhile the antibody ELISA demonstrated a substantially correlated agreement with GM.The pLDH is therefore recommended for mass screening of blood(to detect malaria parasitaemia)for transfusions in the study area.However,where this is not feasible,an RDT will suffice.

Comparative Evaluation and Comprehensive Analysis of Machine Learning Models for Regression Problems

Artificial intelligence and machine learning applications are of significant importance almost in every field of human life to solve problems or support human experts. However, the determination of the machine learning model to achieve a superior result for a particular problem within the wide real-life application areas is still a challenging task for researchers. The success of a model could be affected by several factors such as dataset characteristics, training strategy and model responses. Therefore, a comprehensive analysis is required to determine model ability and the efficiency of the considered strategies. This study implemented ten benchmark machine learning models on seventeen varied datasets. Experiments are performed using four different training strategies 60:40, 70:30, and 80:20 hold-out and five-fold cross-validation techniques. We used three evaluation metrics to evaluate the experimental results: mean squared error, mean absolute error, and coefficient of determination(R2score). The considered models are analyzed, and each model’s advantages, disadvantages, and data dependencies are indicated. As a result of performed excess number of experiments, the deep Long-Short Term Memory(LSTM) neural network outperformed other considered models, namely, decision tree, linear regression, support vector regression with a linear and radial basis function kernels, random forest, gradient boosting, extreme gradient boosting, shallow neural network, and deep neural network. It has also been shown that cross-validation has a tremendous impact on the results of the experiments and should be considered for the model evaluation in regression studies where data mining or selection is not performed.

Multislice spiral computed tomography imaging in evaluating hemophilic arthropathy

Background:Certain hemophilia patients are unable to cooperate with or afford magnetic resonance imaging(MRI)examinations.The purpose of our study was to explore the value of multislice spiral computed tomography(MSCT)in evaluating hemophilic arthropathy(HA).Methods:Thirty-eight patients with 73 joints of HA were consecutively selected from January 2016 to May 2018 for this prospective study.All 73 joints were examined by X-ray,CT,and MRI within 2 days.The MRI scores of the joints were determined by the International Prophylaxis Study Group(IPSG)standard.The CT findings were quantified according to the IPSG standard,except for cartilage injury,which was quantified by joint space narrowing using the X-ray Pettersson score.The CT and MRI scores were compared by the paired Wilcoxon signed-rank test.The correlations between the CT score of joint space narrowing and MRI score of cartilage injury and the total CT and MRI scores were analyzed by Spearman rank correlation.The kappa test was used to compare the consistency of CT and MRI scores.Results:MRI was superior to CT based on the scores for small amount of effusion(P<0.05),synovial hypertrophy and hemosiderin deposition in the mild groups(P<0.05).The CT and MRI scores were not significantly different for moderate and massive effusion,synovial hypertrophy,and hemosiderin deposition in the moderate and severe groups,bone erosion or cystic changes(P>0.05),and there was a high degree of consistency between the two scores(kappa>0.81).The consistency between the Pettersson scores of joint space narrowing on CT and the IPSG scores of cartilage injury on MRI was high(kappa=0.774,P<0.05).Conclusion:The image scores of MSCT are generally consistent with MRI except for mild synovitis,which can be used as an alternative for the evaluation of HA.

Research methods in complementary and alternative medicine： an integrative review

The scientific literature presents a modest amount ot evidence m the use or complementary ana al[erna- tire medicine （CAM）. On the other hand, in practice, relevant results are common. The debates among CAM practitioners about the quality and execution of scientific research are important, Therefore, the aim of this review is to gather, synthesize and describe the differentiated methodological models that encompass the complexity of therapeutic interventions. The process of bringing evidence-based medicine into clinical practice in CAM is essential for the growth and strengthening of complementary medicines worldwide.