A new approach to applying feedforward neural networks to the prediction of musculoskeletal disorder risk

Chen, Chuen-Lung; Kaber, David B.; Dempsey, Patrick G.

doi:10.1016/s0003-6870(99)00055-1

Cited by 34 publications

(34 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, ANNs have been investigated as to their effectiveness for modeling LBD risk (Chen, Kaber, & Dempsey, 2000;Kaber & Chen, 1998;Karwowski, Zurada, Marras, & Gaddie, 1994;Zurada, Karwowski, & Marras, 1997). These methods do not often require strict assumptions such as those required by logistic regression, and the results of the cited investigations indicate that ANNs are a promising method for modeling LBD risk.…”

Section: Introductionmentioning

confidence: 99%

“…In two previous studies (Chen et al, 2000;Kaber & Chen, 1998), the authors developed a unique method for architecting feedforward neural networks (FNNs) and applying them to problems of classifying industrial jobs in terms of risk for LBDs. Historically, FNNs used for ergonomic applications have been developed using an error backpropagation (EBP) algorithm (Killough et al, 1995;Nussbaum & Chaffin, 1996;Zurada et al, 1997).…”

Section: Introductionmentioning

confidence: 99%

“…This can degrade validation performance of the FNN because the network may be "overfit" to the training data and produce poor prediction accuracies for data patterns that differ from those in the training dataset. Overfitting typically occurs when the number of connections among neurons (and network connection weights) is large in comparison to the number of patterns in a training dataset (Chen et al, 2000). If a validation dataset contains patterns different from those in the training dataset, weights may be adjusted to specific training patterns (i.e., overfit to the training data).…”

Section: Introductionmentioning

confidence: 99%

“…As an example, if a classification problem has five independent variables and its response can be classified into two classes, Masters would recommend constructing an initial FNN with three hidden neurons (Int ((2 ϫ 5)1/2) ϭ 3). Masters' equation has been successfully applied to define parsimonious FNN architectures, yielding higher prediction accuracies in validation than arbitrarily defined architectures (Chen et al, 2000).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Using feedforward neural networks and forward selection of input variables for an ergonomics data classification problem

Chen¹,

Kaber²,

Dempsey³

2003

Hum Ftrs & Erg Mfg Svc

Self Cite

View full text Add to dashboard Cite

A method was developed to accurately predict the risk of injuries in industrial jobs based on datasets not meeting the assumptions of parametric statistical tools, or being incomplete. Previous research used a backward-elimination process for feedforward neural network (FNN) input variable selection. Simulated annealing (SA) was used as a local search method in conjunction with a conjugate-gradient algorithm to develop an FNN. This article presents an incremental step in the use of FNNs for ergonomics analyses, specifically the use of forward selection of input variables. Advantages to this approach include enhancing the effectiveness of the use of neural networks when observations are missing from ergonomics datasets, and preventing overspecification or overfitting of an FNN to training data. Classification performance across two methods involving the use of SA combined with either forward selection or backward elimination of input variables was comparable for complete datasets, and the forward-selection approach produced results superior to previously used methods of FNN development, including the error back-propagation algorithm, when dealing with incomplete data.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Using feedforward neural networks and forward selection of input variables for an ergonomics data classification problem

Chen¹,

Kaber²,

Dempsey³

2003

Hum Ftrs & Erg Mfg Svc

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the context of the model, a low-workload classification represented the experimental condition in which participants were required to manage three aircraft at any time, and a high-workload classification represented the experiment condition in which operators were presented with seven aircraft. We initially used multiple linear regression for selection of input variables for the NN (see Chen, Kaber, & Dempsey, 2000, for a detailed example) among all workload measures recorded during the ATC simulation. We sought to investigate only those predictors that appeared to be significant in explaining objective workload states as inputs to the NN models.…”

Section: Development Of a Functional State Classification Toolmentioning

confidence: 99%

Workload State Classification With Automation During Simulated Air Traffic Control

Kaber

Perry

Segall

et al. 2007

The International Journal of Aviation Psychology

Self Cite

View full text Add to dashboard Cite

Real-time operator workload assessment and state classification may be useful for decisions about when and how to dynamically apply automation to information processing functions in aviation systems. This research examined multiple cognitive workload measures, including secondary task performance and physiological (cardiac) measures, as inputs to a neural network for operator functional state classification during a simulated air traffic control (ATC) task. Twenty-five participants performed a low-fidelity simulation under manual control or 1 of 4 different forms of automation. Traffic volume was either low (3 aircraft) or high (7 aircraft). Participants also performed a secondary (gauge) monitoring task. Results demonstrated significant effects of traffic volume (workload) on aircraft clearances (p < .01) and trajectory conflicts (p < .01), secondary task performance (p < .01), and subjective ratings of task workload (p < .01). The form of ATC automation affected the number of aircraft collisions (p < .05), secondary task performance (p < .01), and heart rate (HR; p < .01). However, heart rate and heart rate variability measures were not sensitive to the traffic manipulation. Neural network models of controller workload (defined in terms of traffic volume) were developed using the secondary task performance and simple heart rate measure as inputs. The best workload classification accuracy using a genetic algorithm (across all forms of ATC automation) was 64%,

show abstract

Ant colony optimization approach for classification of occupational low back disorder risks

Akay

Toksarı

2008

Hum Ftrs & Erg Mfg Svc

View full text Add to dashboard Cite

Manual material handling (MMH) tasks, the leading cause of low back disorders (LBDs), are still extensively used in industry in spite of the advanced technology. ClassiÞcation of industrial jobs related to LBD risks has great signiÞcance in preventing injuries and designing workplaces. In this article, industrial jobs have been classiÞed into two categories, low risk and high risk, using ant colony optimization (ACO). ACO classiÞcation (ACOCLASS) has obtained better results than studies that used the same experimental data. Ergonomic interventions can be done by means of obtained classiÞcation schema for future reductions in low back injuries. C 2008 Wiley Periodicals, Inc.

show abstract

A new approach to applying feedforward neural networks to the prediction of musculoskeletal disorder risk

Cited by 34 publications

References 16 publications

Using feedforward neural networks and forward selection of input variables for an ergonomics data classification problem

Using feedforward neural networks and forward selection of input variables for an ergonomics data classification problem

Workload State Classification With Automation During Simulated Air Traffic Control

Ant colony optimization approach for classification of occupational low back disorder risks

Contact Info

Product

Resources

About