Wonsup Lee

Classification of a Driver's cognitive workload levels using artificial neural network on ECG signals

An artificial neural network (ANN) model was developed in the present study to classify the level of a drivers cognitive workload based on electrocardiography (ECG). ECG signals were measured on 15 male participants while they performed a simulated driving task as a primary task with/without an N-back task as a secondary task. Three time-domain ECG measures (mean inter-beat interval (IBI), standard deviation of IBIs, and root mean squared difference of adjacent IBIs) and three frequencydomain ECG measures (power in low frequency, power in high frequency, and ratio of power in low and high frequencies) were calculated. To compensate for individual differences in heart response during the driving tasks, a three-step data processing procedure was performed to ECG signals of each participant: (1) selection of two most sensitive ECG measures, (2) definition of three (low, medium, and high) cognitive workload levels, and (3) normalization of the selected ECG measures. An ANN model was constructed using a feed-forward network and scaled conjugate gradient as a back-propagation learning rule. The accuracy of the ANN classification model was found satisfactory for learning data (95%) and testing data (82%).

Development of an Evaluation Method for a Driver’s Cognitive Workload Using ECG Signal

Objective: The present study is to develop an effective evaluation method for a driver’s cognitive workload using electrocardiography (ECG) signal. Background: ECG measures such as heart rate (HR) and heart rate variability (HRV) have been used for cognitive workload evaluation. Since ECG changes by cognitive workload vary largely depending on personal characteristics, an optimal analysis protocol of ECG needs to be tailored to each individual accordingly; however, existing studies have not considered personal characteristics in ECG analysis for cognitive workload evaluation. Method: The proposed evaluation method uses the area under the receiver operating characteristic (ROC) curve (AUC). A preliminary analysis was conducted with ECG data collected in a driving simulator while an n-back task was conducted. AUC analysis was performed for four ECG metrics (mean IBI, SDNN, RMSSD, and RMSE), three window spans (20, 30, and 40 seconds), and three update rates (1, 2, and 3 seconds). Results: It was identified that the optimal ECG analysis parameters of metric, window span, and update rate maximizing the discriminability of cognitive workload evaluation varied between individual drivers. Conclusion: The finding of the present study supports the use of an individually customized ECG analysis protocol for better evaluation accuracy of a drivers’ cognitive workload. Application: The proposed ECG analysis method for cognitive workload evaluation can be applied to development of a safe driving support system.

Development of a usability evaluation method using natural product-use motion

The present study developed and tested a new usability evaluation method which considers natural product-use motions. The proposed method measures both natural product-use motions (NMs) and actual product-use motions (AMs) for a product using an optical motion capture system and examines the usability of the product based on motion similarity (MS; %) between NMs and AMs. The proposed method was applied to a usability test of four vacuum cleaners (A, B, C, and D) with 15 participants and their MSs were compared with EMG measurements and subjective discomfort ratings. Cleaners A (44.6%) and C (44.2%) showed higher MSs than cleaners B (42.9%) and D (41.7%); the MSs mostly corresponded to the EMG measurements, which could indicate that AMs deviated from NMs may increase muscular efforts. However, the MSs were slightly different from the corresponding discomfort ratings. The proposed method demonstrated its usefulness in usability testing, but further research is needed with various products to generalize its effectiveness.

Measurement and Application of 3D Ear Images For Earphone Design

Detail anthropometric dimensions and a 3D shape of the outer-ear are applicable to design ear-related products such as an earphone. However, 3D scanning of the ear part is quite difficult due to a complex shape of the ear, also detailed ear dimensions which are needed to be measured for earphone design were not identified in previous studies. This study collected 3D scan images of the whole outer-ear from 100 Korean participants (50 females and 50 males) aged 20 to 59, then measured their detailed ear dimensions for earphone design. The pinna part was directly 3D scanned; and complex shape of the concha and acoustic canal parts were cast by applying an ear casting tool, then the cast was scanned in 3D. 13 ear dimensions were measured by applying an ear measurement system coded using Matlab. Both 3D ear scans and ear measurements were applied to design some earphone parts (earphone-head, ear-band, ear-tip) in this study.

Anlysis Methods of the Variation of Facial Size and Shape Based on 3d Face Scan Images

3D scan images have been successfully applied in ergonomic product design. Features of human body parts (e.g., landmarks, measurements, curvatures, surfaces, volumes) extracted from 3D body scan images can be used to analyze variations of the size and shape of human bodies. The information of size and shape variations can be applied in product design to support technical ideas regarding accommodation, tolerance, and adjustability. This study is aimed to briefly introduce a few analysis methods of body shape variation using 3D facial scan images of Dutch children in order to acquire useful features for the design of a children’s facial mask.

Development of HeadForms and an Anthropometric Sizing Analysis System for Head-Related Product Designs:

The present study developed a sizing analysis system based on the Civilian American and European Surface Anthropometry Resource (CAESAR) database of North Americans (n = 2,299) for head-related product designs. To find representative heads from a huge amount of 3D human scan database, a sizing analysis system is required for efficient analysis of sizing systems based on anthropometric measurements. The head of the CAESAR 3D scan were manually edited to improve a quality for better use to the product design. Twenty one anthropometric landmarks were marked on the edited 3D heads to measure 40 anthropometric dimensions related to the head product designs. All head and face dimensions were automatically measured by applying a measurement system coded using Matlab. Fifteen representative headforms were generated in terms of 5 ethnic groups (composite group, Caucasians, American Africans, Asians, and Hispanics) and 3 gender groups (composite gender, males, and females). Finally, the sizing analysis system was developed based on the measurement of the CAESAR for analysis of head and facial measurements and generation of sizing systems.

Estimation of Instantaneous Hand Joint Centers of Rotation Using 3D Reconstructed Hand Skeleton Motion from CT Scans

The present study estimated instantaneous hand joint centers of rotation (CoR) using 3D reconstructed hand skeleton motions captured from CT scan. We proposed a novel method for estimation of instantaneous joint CoR using the same bone surfaces for different hand postures. Each bone in a template hand posture was registered to the corresponding bone of different hand postures. The registered hand postures (having the same bone surfaces as the template hand posture but different postures) with the template hand posture were then used for estimation of instantaneous joint CoR. The proposed method performed better than the existing methods in estimation of instantaneous joint CoR. Consistency of instantaneous joint CoRs determined in the same rotation angle range was improved by 31.7% to 51.0% in the proposed method. The present study focused on distal interphalangeal (DIP) and proximal interphalangeal (PIP) joints of the index finger of a participant. Joints of the whole hands of more participants will be studied for further generalization of the findings.

Anthropometric analysis of 3D ear scans of Koreans and Caucasians for ear product design

Abstract The present study measured 25 dimensions of the ear including the concha and ear canal for ergonomic design of ear products and compared with existing ear measurement studies. Scanning and casting methods were employed to produce 3D ear images for 230 Koreans and 96 Caucasians and measurements of the ear dimensions were obtained by identifying 21 landmarks on individual ear scan image. The Korean ear measurements were found significantly larger (mean difference  = 0.4–3.7 mm) and more varied (ratio of SDs =1.01–1.55) than those of Caucasians in most of ear dimensions. The average ear length and ear breadth of male were significantly longer ( = 1.3–7.0 mm) and wider ( = 0.8–3.0 mm) than those of female. Use of gender- and ethnicity-composite ear data is recommended in product design due to the much larger intra-population variations (7.5–22.2 mm) than the corresponding inter-population variations. Practitioner Summary: The 3D ear measurements of Koreans and Caucasians were collected and compared with those of different ethnic populations. The distinct ear features of the populations identified in this study are applicable to ergonomic design of ear products with better fit and comfort. Abbreviations: CCW: cavum concha width; CV: coefficient of variation; EB: ear breadth; EL: ear length; SD: standard deviation; SE: sampling error; 3D: 3 dimensional

A Comparison Between Representative 3D Faces Based on Bi- and Multi-variate and Shape Based Analysis

In Ergonomic product design, designers need to translate anthropometric data of the target population into product dimensions or sizing systems. Currently, sizing systems are often based on traditional anthropometric data and generally use the variation of one or two key body dimensions directly related to the product. For products that need to closely fit a certain part of the body it is relevant to incorporate multiple key dimensions. This can be realized by a multivariate approach such as a Principal Component Analysis. Over the past decades, there has been an increase in incorporating 3D imaging in anthropometric surveys. In order to integrate the use of 3D anthropometry in product sizing, representative models are used to visualize the variability of the target population. For the development of a ventilation mask for children, this study compares representative models of 3D faces based on a bivariate, multivariate and shape based analysis of 303 children’s faces.

Contact Pressure Analysis for Wearable Product Design

3D body scanning has been used broadly including digital human modeling, simulation, ergonomic product design, and so forth. This research used template-registered faces of 336 Koreans in order to use them to design an oxygen mask that provides good fit to Korean faces. The finite element analysis method is applied onto the template-registered faces to predict the contact pressure of a mask design onto different faces. The average and variation of the estimated contact pressure values among all the Korean faces were analyzed for evaluation of the appropriateness of a mask design for Koreans. The proposed method can be usefully applied to find an optimal shape of wearable products for a specific target population.