Average Percentage Improvement Research Articles

ClassifyTE: a stacking-based prediction of hierarchical classification of transposable elements.

Transposable Elements (TEs) or jumping genes are DNA sequences that have an intrinsic capability to move within a host genome from one genomic location to another. Studies show that the presence of a TE within or adjacent to a functional gene may alter its expression. TEs can also cause an increase in the rate of mutation and can even mediate duplications and large insertions and deletions in the genome, promoting gross genetic rearrangements. The proper classification of identified jumping genes is important for analyzing their genetic and evolutionary effects. An effective classifier, which can explain the role of TEs in germline and somatic evolution more accurately, is needed. In this study, we examine the performance of a variety of machine learning (ML) techniques and propose a robust method, ClassifyTE, for the hierarchical classification of TEs with high accuracy, using a stacking-based ML method. We propose a stacking-based approach for the hierarchical classification of TEs. When trained on three different benchmark datasets, our proposed system achieved 4%, 10.68% and 10.13% average percentage improvement (using the hF measure) compared to several state-of-the-art methods. We developed an end-to-end automated hierarchical classification tool based on the proposed approach, ClassifyTE, to classify TEs up to the super-family level. We further evaluated our method on a new TE library generated by a homology-based classification method and found relatively high concordance at higher taxonomic levels. Thus, ClassifyTE paves the way for a more accurate analysis of the role of TEs. The source code and data are available at https://github.com/manisa/ClassifyTE. Supplementary data are available at Bioinformatics online.

Abstract P209: Community-Based Outpatient Stroke Rehabilitation Program Achieves Excellent Return to Work Outcomes: Characteristics and Rehabilitation Outcomes of Stroke Survivors Who Return to Work

Background: Return to employment and driving are goals for many stroke survivors. There are few reports of patient centered outcomes like return to work or the characteristics of those who achieve these goals. Methods: Prospective observational study of stroke patients treated between 12/2011-7/2019 in an interdisciplinary outpatient rehabilitation program addressing physical, cognitive/communicative, risk factors, psychosocial, driving and vocational issues. Results: Of 190 consecutive patients,104(55%) were working prior to their stroke. 50% (73% men, 27% women) of those working pre-stroke returned to work upon program completion with 90% returning to their prior job, and 65% at prior hours. 24% of returning workers owned their own business, and 76% had other occupations. Returning workers had average age of 54 years, average NIHSS score of 4(range 0-11), average{mRS} of 2.7 (6% with 1; 25% with 2; 58% with 3; 10% with 4), and 31% were aphasic. 39 suffered ischemic strokes and 13 had hemorrhages. Locations were: 18 Left, 15 Right,11 Bilateral hemisphere, 8 Brainstem). 93% of patients working pre-stroke were driving. 80% who returned to work also resumed driving. Returning workers had average percent improvement in 6” walk and Berg Balance scores of 97% and 50% respectively; this group also improved in all 9 Stroke Impact scale domains with SIS total score improving an average of 22%. In comparison, of those not returning to work 34 had ischemic strokes, 18 had hemorrhages. This group had average age 55.6, average NIHSS score 7.3(range2-16), average {mRS} of 3.3(11% with 2, 54% with 3, 33% with 4, 2% with 5), 42% were aphasic, and 20% resumed driving. Returning workers had average admit Montreal Cognitive Assessment score of 25/30 and Stroke Self Efficacy Score of 85%; with average scores of 21/30 and 71% for those not returning to work. Average length of stay was 4.6(range .5-24) and 7.4(range1.7-18) months for those returning and not returning to work. Other demographic and outcome measures are collected. Conclusions: Despite significant disability, stroke survivors can achieve successful recovery, including return to work and driving. Further study of patients returning to work can help direct rehabilitation efforts for patients with return to work goals.

Binary chemical reaction optimization based feature selection techniques for machine learning classification problems

Feature selection is an important pre-processing technique for dimensionality reduction of high-dimensional data in machine learning (ML) field. In this paper, we propose a binary chemical reaction optimization (BCRO) and a hybrid binary chemical reaction optimization-binary particle swarm optimization (HBCRO-BPSO) based feature selection techniques to optimize the number of selected features and improve the classification accuracy. Three objective functions have been used for the proposed feature selection techniques to compare their performances with a BPSO and advanced binary ant colony optimization (ABACO) along with an implemented GA based feature selection approach called as binary genetic algorithm (BGA). Five ML algorithms including K-nearest neighbor (KNN), logistic regression, Naïve Bayes, decision tree, and random forest are considered for classification tasks. Experimental results tested on eleven benchmark datasets from UCI ML repository show that the proposed HBCRO-BPSO algorithm improves the average percentage of reduction in features (APRF) and average percentage of improvement in accuracy (APIA) by 5.01% and 3.83%, respectively over the existing BPSO based feature selection method; 4.58% and 3.12% over BGA; and 4.15% and 2.27% over ABACO when used with a KNN classifier.

Solving routing and spectrum allocation problems in flexgrid optical networks using pre-computing strategies

Flexible optical network architectures are considered a very promising solution where spectrum resources are allocated within flexible frequency grids. This paper presents a minimum spectrum utilization (SU) and average path length (APL) approach to solve the (off-line) routing and spectrum allocation problem (RSA) based on combining a simple ordering pre-computation strategy, namely most subcarriers first (MSF) with three nature-inspired algorithms. These algorithms are ant colony optimization, differential evolution based relative position indexing (DE-RPI), and differential evolution general combinatorial (DE-GC). We begin by showing that MSF is the most effective ordering pre-computation strategy when compared to other well-known typical heuristics in the literature, such as first-fit, and longest path first. Then, we apply MSF in combination with the three nature-inspired algorithms to simultaneously optimize the SU and APL. The usefulness of MSF ordering pre-computation strategy is presented via a comparison of results obtained when using and not using MSF under the same scenarios. The algorithms are evaluated in benchmark optical networks, such as the NSFNet, the European optical network, and the 40-node USA network. We show that DE-RPI with MSF ordering pre-computation is the best option to solve the RSA problem, obtaining an average improvement percentage in the range of 0.9772–4.4086% on the SU and from $$-0.1668$$ to 0.8511% on the APL when compared to other meta-heuristics, either with or without the MSF ordering policy.

Reversible data hiding exploiting Huffman encoding with dual images for IoMT based healthcare

Continuous technological progressions and huge investments are made for the realization of the various goals in the Internet of Things (IoT) driven networks. Internet of Medical Things (IoMT) being a part of IoT has made human living smarter. It is revolutionizing the healthcare industry and is providing a smarter healthcare framework to the people. The generic IoMT framework consists of the major components i.e., data acquisition, communication gateways, and servers. Once the data is acquired, it is sent over the insecure channel where its authentication is essential before diagnosis. In this work, a dual image reversible data hiding technique with high capacity is proposed for IoMT based networks. First of all, the acquired secret data is preprocessed using the Huffman encoding strategy. Once Huffman coding is applied, a codebook of ‘d’ bits is generated for encoding the converted decimal values using indices. The value of these indices is divided into 2 parts and embedded into two visually similar images to obtain dual stego images. The scheme has provided a very high payload while maintaining good perceptual quality. The results obtained depict significant improvement compared to the state-of-the-art. The scheme provides an average (percentage) improvement in embedding capacity by 33.2%, with the improvisation of Peak Signal to Noise (PSNR) Ratio by 1.32%. The average value of the Structural Similarity Index (SSIM) is found to be 0.8873. The scheme is computationally efficient which makes it a better candidate to be used in IoMT driven networks.

A Combination Prediction Model of Long-Term Ionospheric foF2 Based on Entropy Weight Method.

It is critically meaningful to accurately predict the ionospheric F2 layer critical frequency (foF2), which greatly limits the efficiency of communications, radar, and navigation systems. This paper introduced the entropy weight method to develop the combination prediction model (CPM) for long-term foF2 at Darwin (12.4° S, 131.5° E) in Australia. The weight coefficient of each individual model in the CPM is determined by using the entropy weight method after completing the simulation of the individual model in the calibration period. We analyzed two sets of data to validate the method used in this study: One set is from 2000 and 2009, which are included in the calibration period (1998–2016), and the other set is outside the calibration cycle (from 1997 and 2017). To examine the performance, the root mean square error (RMSE) of the observed monthly median foF2 value, the proposed CPM, the Union Radio Scientifique Internationale (URSI), and the International Radio Consultative Committee (CCIR) are compared. The yearly RMSE average values calculated from CPM were less than those calculated from URSI and CCIR in 1997, 2000, 2009, and 2017. In 2000 and 2009, the average percentage improvement between CPM and URSI is 9.01%, and the average percentage improvement between CPM and CCIR is 13.04%. Beyond the calibration period, the average percentage improvement between CPM and URSI is 13.2%, and the average percentage improvement between CPM and CCIR is 12.6%. The prediction results demonstrated that the proposed CPM has higher precision of prediction and stability than that of the URSI and CCIR, both within the calibration period and outside the calibration period.

A Combined Regimen of Dietary Supplements Containing Omega-3 Fish Oil and Standardized Maqui Berry Extract Improves Dry Eye Related Symptoms in a Pilot Randomized Interventional Clinical Study

Purpose: To determine if a dietary supplement consisting of a re-esterified triglyceride form of omega-3 fish oil combined with a standardized maqui berry extract is more efficacious for the treatment of dry eye disease than omega-3 fish oil alone. Methods: This was an institutional review board-approved, single-site, randomized, interventional, placebocontrolled, comparative, pilot clinical study. Patients with mild to moderate dry eye disease (DED) were enrolled and divided into four study groups. Group 1 (O3M) received dietary supplements of a re-esterified triglyceride form of omega-3 fish oil (O3) and a standardized maqui berry extract (M); group 2 (O3) received only the omega-3 supplements; group 3 (M) received only the maqui berry extract supplements; and group 4 received placebo capsules. Thirteen participants aged 26–78 years with DED were included. Patients were evaluated for DED by corneal staining, changes in tear osmolarity values, matrix metalloproteinase-9 (MMP-9) levels, Schirmer's test, and tear breakup time (TBUT) at baseline, week 2, week 4, week 8, and a final assessment at week 12. In addition, at each visit patients were given the Ocular Surface Disease Index (OSDI) and Dry Eye questionnaires to complete in order to assess their level of DED. Adverse events (AEs) were recorded on AE forms at each patient visit, and any serious AEs were to be reported to the sponsor within 24 hours via telephone. Results: 13 patients participated in the study (77% females, 23% males), with an average age of 55 years. The highest average percent improvement in OSDI & DED scores from baseline through week 12 of study treatment was demonstrated in the O3M group (77.6% and 70.5% respectively), compared to 39.2% improvement in OSDI assessment and 44.9% in DED score for the O3 group. Schirmer's tests demonstrated an average increase in tear production over baseline of ~3.4 mm in OD and ~0.8 mm in OS after 3 months for the O3M group. In comparison, for the O3 group, the tear production decreased by ~0.6 mm in OD and increased by ~1.67 mm in OS from baseline to 3 months. TBUT improved significantly in the O3M group, from baseline to 3 months it went up by ~3.2-2.8 sec in both eyes (OD & OS), while for the O3 group TBUT values went up by ~1.66-2.0 sec in both eyes (OD&OS). Notably, there were no adverse events reported due to dietary supplement usage during the study. Conclusion: Dietary supplements of omega-3 fish oil and standardized maqui berry extract outperformed omega-3 fish oil alone as a treatment for DED.

Abstract WP188: Community-Based Outpatient Stroke Rehabilitation Program Achieves Excellent Outcomes Including Return to Work, Driving, Stroke Knowledge, and Other Rehabilitation Outcomes

Background: Return to driving and employment are goals for many stroke survivors. There are few reports of patient centered outcomes including return to employment, driving, self-efficacy, functional outcomes, stroke knowledge or characteristics of survivors who achieve these goals. Methods: Prospective observational study of stroke patients treated in an interdisciplinary outpatient rehabilitation program addressing physical, cognitive, communicative, risk factor/stroke knowledge, self-efficacy, psychosocial, driving and vocational issues. Results: 190 consecutive patients(117 men, 73 women, average age 62(18-90), 66 aphasic) with baseline modified Rankin Scores of 5(1%); 4(25%); 3(56%); 2(15%); 1(3%), and average NIHSS score of 6(range 0-18) were treated between 12/2011-7/2019. 136 Patients suffered ischemic strokes and 54 had hemorrhages. Locations were: 65 left, 69 right, 26 bilateral hemisphere, 30 brainstem. 41% had MCA territory strokes. Of 104 patients working pre-stroke, 50% returned to work and an additional 7% were work capable upon completion of program. 95% of patients driving pre-stroke could not drive on admission. After training, 43%(67/157) returned to driving. Patients had improvements in all 9 domains of Stroke Impact Scale with SIS total score improving an average of 23%. Patients received an average of 33 physical therapy visits with average percent improvement of 72% and 41% on 6” walk and Berg Balance scores. Average percent improvement in Stroke Self Efficacy Scores was 29%. After individualized cardiovascular risk, stroke and medication education sessions with the Nurse Practitioner based on AHA guidelines and Life’s Simple 7’s curriculum, stroke and risk factor knowledge quiz scores improved an average of 29%. Average length of stay was 5(range .5-24) months. Multiple other demographic and outcome measures are collected. Conclusions: Outcome data show a community-based team rehabilitation program can successfully combine CV/stroke education with rehabilitation services to maximize patient centered outcomes including return to work, driving, overall stroke recovery, physical functioning, stroke knowledge, and self-efficacy for many types of stroke survivors with moderate to severe disability.

Color measurement of single yarn based on hyperspectral imaging system

AbstractThis article proposed a novel approach to color measurement of a single yarn using hyperspectral imaging system (HIS). Due to the size of a single yarn, it is impossible for spectrophotometers to measure its color directly. The HIS can acquire the spectral reflectance of continuous bands within a region of interest on a yarn sample, which can achieve color measurement of a single yarn compared with traditional spectrophotometers. A single yarn is segmented from the background by a spectral matching method through adaptively setting threshold of Fréchet distance values. The spectral reflectance of single yarn is specified by a method that lightness of pixels used as weight. The experiment based on Pantone Cotton Chip Set shows that the interinstrument agreement between the HIS and a standard spectrophotometer Datacolor SF650 has a significant improvement after using the R‐Model, and the average percentage improvement of the color difference is up to 54.99%. The yarn segmentation comparative experimental results show that the proposed method to segment single yarn from background is better in retaining the edge information of the yarn than the modified K‐means clustering method, and the color of the yarn segmented by the proposed method is more similar to the actual color of single yarn.

HDSR-Flor: A Robust End-to-End System to Solve the Handwritten Digit String Recognition Problem in Real Complex Scenarios

Automatic handwriting recognition systems are of interest for academic research fields and for commercial applications. Recent advances in deep learning techniques have shown dramatic improvement in relation to classic computer vision problems, especially in Handwritten Text Recognition (HTR). However, several approaches try to solve the problem of deep learning applied to Handwritten Digit String Recognition (HDSR), where it has to deal with the low number of trainable data, while learning to ignore any writing symbol around the digits (noise). In this context, we present a new optical model architecture (Gated-CNN-BGRU), based on HTR workflow, applied to HDSR. The International Conference on Frontiers of Handwriting Recognition (ICFHR) 2014 competition on HDSR were used as baselines to evaluate the effectiveness of our proposal, whose metrics, datasets and recognition methods were adopted for fair comparison. Furthermore, we also use a private dataset (Brazilian Bank Check - Courtesy Amount Recognition), and 11 different approaches from the state-of-the-art in HDSR, as well as 2 optical models from the state-of-the-art in HTR. Finally, the proposed optical model demonstrated robustness even with low data volume (126 trainable data, for example), surpassing the results of existing methods with an average precision of 96.50%, which is equivalent to an average percentage of improvement of 3.74 points compared to the state-of-the-art in HDSR. In addition, the result stands out in the competition’s CVL HDS set, where the proposed optical model achieved a precision of 93.54%, while the best result so far had been from Beijing group (from the competition itself), with 85.29%.

EFEKTIFITAS PELATIHAN PERAWATAN LUKA TINGKAT DASAR DI BAPELKES BATAM DALAM MENINGKATKAN KEMAMPUANTENAGA KEPERAWATAN DI PROVINSI KEPULAUAN RIAU

The improvement of wound care capabilityof nurses needs to be continually conducted due to the development in science and technology of the wound care. One way to improve the capabilityof the nurse can be achievedby conducting a training program. This paper is aimed to assess the effectiveness of the basic level wound care training program at Health Training Center (Bapelkes) of Batam in the improvement of the nurses’ capabilities in the Kepulauan Riau Province. Thisstudy implements qualitative descriptive method by using questionnaires as research instrument which were distributed to three groups of respondents, i.e. the alumni, the alumni’s mentors, and the alumni’s working partners,out of three training batches between 2011-2013. The variables evaluated in this study include performance, working motivation, professional competency, and training competency. Descriptive statistical analysis methods and analytic statistics (simple linear regression) are also used in data analysis. The results of the study show that the training program is effective in improving the nurses’ capability in all evaluated variables with the average improvement percentage of 96,88% (based on response of the alumni), 90,63% (based on the response of the alumni’s mentors), and 92,19% (based on the response of the alumni’s working partners). The study also concludes that the correlation of responsesamong the group of respondents showing that the correlation of the responses between the alumni and the alumni’s mentors is strongly significant with the coefficient correlation (R2) of 0.804 and strong with the R2 of 0,516 for the correlation of the responses between the alumni and the alumni’s working partners. This means that the assessment given by training alumni, alumni superiors and alumni colleagues is not biased and the three groups of respondents suppot each other/ have in common, assessments given by training alumni and above alumni have a very strong relationship with the correlation coefficient (R2) 0.804. 
 
 Keywords : Post-training evaluation; wound care; nurse.

Botox for chronic migraine: Local experience from UAE

Chronic migraine (CM) is a disabling neurological condition characterized by episodes of pulsating unilateral or bilateral headache. Food and Drug Administration (FDA) approved onabotulinumtoxinA (Botox) for the prophylactic treatment of CM in 2010. It has been shown that onabotulinumtoxinA is effective in the reduction of headache frequency and severity in patients with CM. To show our own experience of treatment a patients with CM with Botox. We retrospectively assessed a patient’s with chronic migraine who received a Botox treatment in the last one year. Diagnosis established on ICH-III classification. A 19 patients had a Botox treatment according to the PREEMPT study protocol (31 injection sites with a total dose of 155 units). Average age was 40.9 years, average headache duration of 10.9 years. Average number of headache/month before Botox treatment was 18.7 days. Average number of medication used for migraine prophylaxis before Botox were 2.7 (maximum - 6). Botox injection sessions ranged from 1 to 10 with an average of 5 sessions. After Botox an average number of headache / month drop to 3.65. Average percentage of improvement was 71.5% (from 20% up to 100%). The complications were encountered only in one patient with mild temporary ptosis. Botox is an efficient treatment option for patients with chronic migraine. Botox treatment well tolerable and safe procedure leading to significant reduction of headache spells per month. Proper section of patients is crucial to achieve a good results of treatment.

Second ejaculation: a simple, cost free mechanism to deal with high sperm DNA fragmentation

High sperm DNA fragmentation is a controversial subject. However, many physicians test for DNA fragmentation and feel it is important. If high, methods of dealing with DNA fragmentation include testicular sperm aspiration, Anexin sperm wash and ICSI. These procedures add cost, pain after surgery and are of undetermined value. Sperm DNA fragmentation is felt to occur in the epididymis while waiting to be expelled. This study was undertaken to determine if a second ejaculation 3-hours after the first could improve sperm DNA fragmentation, by limiting time in the epidydimis. A prospective cohort study where males were requested to wait 3-days without an ejaculation at which point a semen analysis and DNA fragmentation was performed and repeated 3-hours latter on a 2nd specimen. 112 subjects underwent the two semen analysis protocol as part of the fertility evaluation. All ejaculations were performed at the fertility center. DNA fragmentation was evaluated using the halo test. Data was compared by intra-subject t-test. Data is presented as % or mean±SD. Power analysis suggested ≥73 subjects were required for an 80% power and an alpha of 5% with a 2 unit mean difference with SD of 6 units. High DNA fragmentation is >35%. Male age was 36±7 years (range 29-65). DNA fragmentation decreased from 34.6±19.4 to 23.7±16.0% (p<0.0001) in the 1st and 2nd specimen respectively. Average percentage improvement 23%±30%. Among subjects with high fragmentation 22/49 (45%) failed to improve into the normal range. Regarding subjects with initial DNA fragmentation>35%, comparison of 1st and second 2nd fragmentation were 52±16% & 36±17% (p<0.0001), respectively. Greatest improvement was 97%-28% DNA fragmentation. 7/112 had worse DNA fragmentation in the second specimen and of those only 2 fell above the normal range, both with a first specimen above the normal range as well. Among semen parameters volume went from 3.1±3.3ml to 1.9±0.8ml, p=0.0001, concentration from 41±39 to 32±31 million/ml, p=0.001 & progressive motility increased from 57±21% to 60±21%, p=0.06. In none of the cases where total motile sperm count was greater than 5 million did the quality of the second semen specimen convert the subject to ICSI. The first 10 subjects had both 1st and 2nd DNA fragmentation confirmed with the TUNEL assay and equivalent improvements were seen r=0.97 (p<0.05), this was not continued due to cost assumed by the clinic. High DNA sperm fragmentation can often be managed with a second ejaculation 3 hours after the first. Changes in sperm quality are not clinically significant and none of the ICSI specimens from ejaculation 1 would have required ICSI based on the ejaculation 3 hours latter. 55% improve into the normal range. Therefore, a second ejaculation represents a safe, cost free mechanism to deal with this issue in many patients.

Acoustic power transfer for biomedical implants using piezoelectric receivers: effects of misalignment and misorientation

Implantable medical devices (IMDs) can be powered wirelessly using acoustics with no need for a battery. In an acoustic power transfer system, which consists of a transmitter, medium, and a receiver, the power that the receiver generates is a function of its position (depth, orientation, and alignment relative to the transmitter). The power delivered to the implant should remain stable and reliable even with possible uncertainties in the location of the implant. In this paper, we compare two common designs for piezoelectric ultrasonic transducers that can be used for acoustically powering IMDs, and study their generated power sensitivity to any change in their location. Although commercial off-the-shelf (COTS) transducers are widely being used in the literature, they may not be the best candidate for powering small implants since they may not be able to provide sufficient power in the presence of location uncertainties. Piezoelectric micromachined ultrasonic transducers (pMUTs) are diaphragm structures and are also suitable for wirelessly powering implants. We present a pMUT receiver and study the sensitivity of the generated power of the pMUT to changes in its position. We then perform a comparative study between power generation capability of our pMUT and a COTS transducer with the same lateral dimensions as the pMUT. We observed that the generated power from a pMUT structure is less sensitive to misorientation and misalignment of the device. The average percentage improvement in the generated power from pMUT compared to COTS are 86%, 917%, and 111% for depth, alignment, and orientation, respectively.

An Improved Radial Basis Function Networks Based on Quantum Evolutionary Algorithm for Training Nonlinear Datasets

In neural networks, the accuracies of its networks are mainly relying on two important factors which are the centers and spread value. Radial basis function network (RBFN) is a type of feedforward network that capable of perform nonlinear approximation on unknown dataset. It has been widely used in classification, pattern recognition, nonlinear control and image processing. Thus, with the increases in RBFN application, some problems and weakness of RBFN network is identified. Through the combination of quantum computing and RBFN provides a new research idea in design and performance improvement of RBFN system. This paper describes the theory and application of quantum computing and cloning operators, and discusses the superiority of these theories and the feasibility of their optimization algorithms. This proposed improved RBFN (I-RBFN) that combined with cloning operator and quantum computing algorithm demonstrated its ability in global search and local optimization to effectively speed up learning and provides better accuracy in prediction results. Both the algorithms that combined with RBFN optimize the centers and spread value of RBFN. The proposed I-RBFN was tested against the standard RBFN in predictions. The experimental models were tested on four literatures nonlinear function and four real-world application problems, particularly in Air pollutant problem, Biochemical Oxygen Demand (BOD) problem, Phytoplankton problem, and forex pair EURUSD. The results are compared to I-RBFN for root mean square error (RMSE) values with standard RBFN. The proposed I-RBFN yielded better results with an average improvement percentage more than 90 percent in RMSE.

Distance Weighted K-Means Algorithm for Center Selection in Training Radial Basis Function Networks

The accuracies rates of the neural networks mainly depend on the selection of the correct data centers. The K-means algorithm is a widely used clustering algorithm in various disciplines for centers selection. However, the method is known for its sensitivity to initial centers selection. It suffers not only from a high dependency on the algorithm's initial centers selection but, also from data points. The performance of K-means has been enhanced from different perspectives, including centroid initialization problem over the years. Unfortunately, the solution does not provide a good trade-off between quality and efficiency of the centers produces by the algorithm. To solve this problem, a new method to find the initial centers and improve the sensitivity to the initial centers of K-means algorithm is proposed. This paper presented a training algorithm for the radial basis function network (RBFN) using improved K-means (KM) algorithm, which is the modified version of KM algorithm based on distance-weighted adjustment for each centers, known as distance-weighted K-means (DWKM) algorithm. The proposed training algorithm, which uses DWKM algorithm select centers for training RBFN obtained better accuracy in predictions and reduced network architecture compared to the standard RBFN. The proposed training algorithm was implemented in MATLAB environment; hence, the new network was undergoing a hybrid learning process. The network called DWKM-RBFN was tested against the standard RBFN in predictions. The experimental models were tested on four literatures nonlinear function and four real-world application problems, particularly in Air pollutant problem, Biochemical Oxygen Demand (BOD) problem, Phytoplankton problem, and forex pair EURUSD. The results are compared to proposed method for root mean square error (RMSE) in radial basis function network (RBFN). The proposed method yielded a promising result with an average improvement percentage more than 50 percent in RMSE.

An Improved Radial Basis Function Networks in Networks Weights Adjustment for Training Real-World Nonlinear Datasets

In neural networks, the accuracies of its networks are mainly relying on two important factors which are the centers and the networks weight. The gradient descent algorithm is a widely used weight adjustment algorithm in most of neural networks training algorithm. However, the method is known for its weakness for easily trap in local minima. It suffers from a random weight generated for the networks during initial stage of training at input layer to hidden layer networks. The performance of radial basis function networks (RBFN) has been improved from different perspectives, including centroid initialization problem to weight correction stage over the years. Unfortunately, the solution does not provide a good trade-off between quality and efficiency of the weight produces by the algorithm. To solve this problem, an improved gradient descent algorithm for finding initial weight and improve the overall networks weight is proposed. This improved version algorithm is incorporated into RBFN training algorithm for updating weight. Hence, this paper presented an improved RBFN in term of algorithm for improving the weight adjustment in RBFN during training process. The proposed training algorithm, which uses improved gradient descent algorithm for weight adjustment for training RBFN, obtained significant improvement in predictions compared to the standard RBFN. The proposed training algorithm was implemented in MATLAB environment. The proposed improved network called IRBFN was tested against the standard RBFN in predictions. The experimental models were tested on four literatures nonlinear function and four real-world application problems, particularly in Air pollutant problem, Biochemical Oxygen Demand (BOD) problem, Phytoplankton problem, and forex pair EURUSD. The results are compared to IRBFN for root mean square error (RMSE) values with standard RBFN. The IRBFN yielded a promising result with an average improvement percentage more than 40 percent in RMSE.

Abstract TP181: Community Based Outpatient Stroke Rehabilitation Program Achieve Excellent Outcomes Including Return to Work, Driving, Stroke Education Knowledge, and Other Rehabilitation Outcomes

Background: Return to driving and employment are goals for many stroke survivors. There are few reports of patient centered outcomes including return to employment, driving, self efficacy, functional outcomes, stroke knowledge or characteristics of patients who achieve these goals. Methods: Prospective observational study of stroke patients treated in an interdisciplinary outpatient rehabilitation program addressing physical, cognitive, communicative, risk factors/stroke knowledge, self efficacy, psychosocial, driving and vocational issues. Results: 168 consecutive patients(104 men, 64 women, average age 62(18-90),61 aphasic)with baseline modified Rankin Scores of 5( 1%); 4(27%); 3(55%); 2(13%); 1(4%), and average NIHSS score of 6(range 0-16) were treated between 12/2011-7/2018. 116 patients suffered ischemic strokes(53 Left, 63 right, 24 bilateral hemisphere, 28 brainstem) and 52 had hemorrhages. 41% of patients had MCA territory strokes. Of 91 patients working pre-stroke, (47%) returned to work and and an additional 6% were work capable upon completion of program. 96% of patients driving pre-stroke could not drive at time of admission. After training, 45%(65/143) returned to driving. Patients had improvement in all 9 domains of Stroke Impact Scale with SIS total score improving an average of 26%. Patients received an average of 33 physical therapy visits with average percent improvement of 70% and 76% on 6 " walk and Berg Balance scores. Average Stroke Self Efficacy Scores improved from 74% to 82%. After individualized cardiovascular/stroke, and medication education sessions with the Nurse Practitioner based on AHA guidelines and Life's Simple 7's curriculum, Stroke Knowledge Quiz and risk factor knowledge scores improved 13% and 18%. Average length of stay was 5(range .5-20) months. Multiple other demographic and outcome measures are collected. Conclusions: Outcome data show a community-based team rehabilitation program can successfully combine cardiovascular education and rehabilitation services to maximize patient centered outcomes including return to work, driving, overall stroke recovery, physical functioning, stroke knowledge, and self efficacy for many types of stroke survivors with moderate to severe disability.

Semi-permutation-based genetic algorithm for order acceptance and scheduling in two-stage assembly problem

The joint decision-making of order acceptance and scheduling has recently gained increasing attention. Besides, the two-stage assembly scheduling problem has various real-life applications. The current paper considers an integrated model for order acceptance and scheduling decisions in two-stage assembly problem. The objective is maximizing profit which is the sum of revenues minus total weighted tardiness of the accepted orders. A mixed-integer linear programming model is developed based on time-index variables. Also, a new concept of semi-permutation scheduling is introduced assuming that positions of each job on all of the machines have no significant difference in the optimal solution. This problem is NP-hard, and therefore, a genetic algorithm (GA)-based heuristic is proposed to apply semi-permutation concept, named semi-permutation GA (SPGA), to solve the problem efficiently. The solutions of SPGA are compared with those of CPLEX and non-semi permutation GA (N-SPGA). Computational experiments are conducted in a diverse range of problem instances indicating that the SPGA performs much better than CPLEX regarding the average percentage of improvement, ranging from 1.4 to 168.84%, and run time. The results revealed that an increasing number of machines and orders could lead to a dramatic decrease in the performance of CPLEX and N-SPGA than SPGA. Also, the effect of semi-permutation scheduling is investigated. According to the result, semi-permutation scheduling had a strong effect on the performance of the algorithm. As a result, the SPGA algorithm outperformed the non-semi permutation version of GA completely. Moreover, SPGA could represent the better performance of 36.27% in average in comparison with N-SPGA.

Experience with definitive instrumented final fusion after posterior-based distraction lengthening in patients with early-onset spinal deformity: single center results.

Limited reports exist with regard to clinical and radiographic details of patients with early-onset spinal deformity (EOSD) undergoing definitive instrumented final fusion (FF) following implantation of a growing construct. Charts and radiographs were queried for all patients who underwent FF after management of EOSD with a distraction-based posterior construct at a single institution from 2006 to 2017. Patients managed during the growth modulation period with either proximal rib or spinal fixation were included. Thirteen patients qualified for inclusion. Spinal deformity etiologies were varied (neuromuscular: 5, idiopathic: 4, congenital: 4). Average age at implantation was 6.5 years, and patients underwent an average of 8.15 lengthening procedures over an average of 69 months. After the growing program, modest correction in main coronal Cobb was obtained at FF (average 52.4° before FF, 37.6° following FF, P<0.001; average percent improvement of 27%). Five patients required posterior column osteotomies. Using criteria proposed by Flynn and colleagues, 'minimal' coronal correction was achieved in 23% of patients and 'moderate' in 77%. No patient achieved 'substantial' (>50%) correction. Subjective poor bone quality was appreciated in 67% of patients, and 85% were noted to have areas of autofusion over previously spanned levels. Two (15%) patients sustained a total of four complications. After growth modulation procedures, EOSD deformities are generally stiff, bone quality is often subjectively poor, and autofusion is common, even after use of proximally rib-based systems. Modestly improved coronal Cobb correction can be obtained at FF but may require single or multiple posterior osteotomies. Intraoperative difficulties can be expected, and clinicians should be alert for intraoperative and postoperative complications.