Keywords

1 Introduction

Most of human mechanical interactions with the surrounding world are performed by the hands [1]. There are many tools and devices which are used by hands, from everyday life (scissors, hammers, mobile phones, mouses) to areas of expertise (laparoscope, rackets, VR controllers). Extensive research has shown that hand related products and handheld devices would have a poor use experience without good ergonomic design [2,3,4]. Lack of ergonomic principles in these hand related products have been proved to be a contributing factor in hand diseases, such as tenosynovitis [5,6,7], neuropraxia [8] and etc.

Anthropometry, as an important branch of ergonomics, collects large amounts of anthropometric dimensions, analyses the measured data, and studies their relationship with the work systems, thus provide indication for the design of products and systems [9,10,11]. A considerable amount of literature has been published on hand anthropometry of varies countries around the world. Furthermore, different methods have been proposed to obtain hand anthropometric data. Direct measurement is the most traditional and common approach, and it is involved in hand anthropometric study of Czech population [12], Korean [13], Iranian male workers [14], Nigerian farm workers [15], northern Colombian [11] and Bangladeshi agricultural farm workers [16]. Except this, some researchers developed hand anthropometric database by photogrammetric technique, such as the hand anthropometry of Spain [17], and San Francisco, USA [18]. In recent years, a new progressing method to collect human dimensions data, three-dimensional scanning method, has adopted by several research. For example, the hand anthropometric examinations of Americans [19], and people from Taiwan, China [20].

The 3D scanning technology have been developing over years, which facilitates the investigation of anthropometry. Using a 3D laser or optical scanner, 3D scanning collects surface measurements including not only length and circumference but also complex geometry including shape, surface area, and volume [21,22,23,24,25]. Compared with direct measurement or photogrammetry, 3D scanning shows better reliability if there is good scanning quality [21,22,23].

On the one hand, data for hand dimensions of Chinese are scarce. On the other hand, there has been a significant rising of indigenous technological companies who have started to explore their own design, including the design of hand related products and handheld devices [26]. There is high demand for these companies to improve the ergonomics of their products to achieve good wearing comfort for hand products. Therefore, a representative and up-to-date hand anthropometric database is needed to help improve the ergonomics design of hand products. In this study, a hand anthropometry survey was created by recruiting 937 participants from 6 main geographic areas of the China (North China, Northeast China, Northwest China, Southwest China, South China, and East China). The aim of this study includes:

  • To build a database including scanned hand models and hand dimensions of Chinese;

  • To investigate the differences in hand dimensions between left and right hands, and between males and females;

  • To compare obtained hand dimensions with corresponding dimensions from other populations.

2 Methodology

2.1 Participants

According to the procedures outlined in the ISO 15535 [27]: general requirements for establishing anthropometric databases, the minimum sample size per region to build a national hand database was calculated to be 100. The accuracy of measurements depends on the sample size. In this study, a larger sample size involving 937 subjects was chosen in order to obtain sufficient data quality and generalization.

The proportional stratified random sampling method was adopted for this survey, through which the population was split into six strata. After the strata were established, a simple random sample was captured from every stratum separately according to proportion. This survey has three demographic variables: gender, age, and geographical area. With the proportional stratified random sampling method, this anthropometric survey is considered to be representative in gender, age and geographical area (Table 1).

Table 1. Basic information of all participants.
Full size table

2.2 Experiment

In this study, a novel 3D scanning approach was adopted after a pilot scanning trial. This trial was conducted to obtain good hand images, and avoid errors resulted from hidden hand regions and sway of the hand that were found in previous studies [24, 25, 28]. Two methods were used to improve the scanning quality. Firstly, change reflecting angle effectively to avoid the occurrence of hidden hand regions. Secondly, reduce scanning time by using Artec Eva scanner (Artec 3D, Luxembourg), which has a scanning speed with million points per second. Thirdly, enhance the proficiency of operators who were selected by the pilot hand scanning trial. In this way, the required time for scanning was shortened from 8 min to 2 min, result in the decrease of hand sway [21].

Each participant was shown a brief overview on the aim and processes before the experiment. Marking stickers (ϕ = 3 mm, thickness = 0.5 mm) were attached on the hand of each participant as hand identified landmarks. Both hands were scanned in the splayed and closed postures respectively with an accuracy of up to 0.1 mm. Artec Studio 15 (Artec 3D, Luxembourg) was used as the 3D data post-processing software to export high-quality 3D hand models. Geomagic Wrap (Artec 3D, Luxembourg) were chosen to extracted measurements and features of 3D hand scans.

2.3 Measurement

Hand landmarks and dimensions were identified and extracted according to a quantity of literature [11, 13, 16, 17, 29,30,31,32,33]. Definition of 10 dimensions was based on ISO 7250 [32], GBT 5703 [33] and other previous studies [34], as shown in Table 2. All measurements were calculated and extracted using Geomagic Wrap.

Table 2. Definition of hand dimensions.
Full size table

2.4 Statistical Analysis

SPSS 25 (IBM Corp., Armonk, New York) was used for the statistical analysis of all dimensions for males and females respectively. Statistics like mean value, standard deviation, and various percentiles (5th, 25th, 50th, 75th and 95th) were calculated, as shown in Table 3. All hand measurements were analyzed using Kolmogorov-Smirnov test. Independent sample T-tests was conducted to compare the differences between left and right hands. Besides, to investigate the diversity between males and females, another independent sample T-tests was taken for right hand dimensions. At the aspect of ethnography, T-test were also conducted to compare the hand features from six areas of China. Moreover, differences of hand dimensions between Chinese and other populations were analyzed.

3 Result

The results of anthropometric hand dimensions of left and right including the mean and standard deviation, taken from Chinese male and female of six geographic areas, are shown in Table 3 and Table 4. These two tables also show a matched samples t-test results between left and right hands for male and female respectively. In this paper, hand dimensions comparisons mainly focus on right hands. Tables 5 and 6 lists the 5th, 25th, 50th, 75th, and 95th percentile for right hand measurements. Table 7 depicts a comparison of right hand measurements between male and female.

Table 3. Statistics of left and right hand dimensions of Chinese males.
Full size table
Table 4. Statistics of left and right hand dimensions of Chinese females.
Full size table
Table 5. Percentiles of right hand dimensions of Chinese males.
Full size table
Table 6. Percentiles of right hand dimensions of Chinese females.
Full size table
Table 7. Comparison of right hand measurements between Chinese males and females.
Full size table

4 Discussion

In this study, ethnic did not be chosen as a criterion in participants recruitment. After 3D hand scanning and data collection, it was found that 98% of the participants are Han nationality. Han Nationality is the largest nationality among the 56 ethnic groups of China, accounting for 91.59% of China’s population. So, the sample of this study could be thought to be in accordance with the ethnic group distribution of China.

A three-dimensional scanning method was adopted in this study. Duplicated hand models were obtained through 3D scanning, after which 10 hand dimensions were extracted. Consequently, the result of this study is expandable that more dimensions could be extracted and analyzed according in the future. It is worth noting that unlike conventional direct measurement, 3D scanning doesn’t have skin deformation issues resulted from measurement tools, so the figure length dimensions of this study might be smaller than that of conventional direct measurement method, and the figures of width and thickness dimensions might be bigger [21]. With the development of computer-aided design (CAD), more and more products would be developed through CAD technology. Compared with conventional direct measurement method, three-dimension scanning method could match more with CAD, hence might gradually gain its popularity.

The result of this survey could be used to determine dimensions of hand related products and hand-held devices. When designing a dimension-adjustable product which covers 90% of Chinese population, product dimensions should range from the 5th to the 95th percentile. If a glove is developed for Chinese women, then it is not appropriate for the part of palm to be shorter than 94.7 mm (5th percentile value of Chinese female palm length). Besides, percentiles of finger breadth could be used to determine the diameter of rings, and percentiles of hand breadth could provide design reference to the breadth of mouses.

4.1 Left-Right Hand Comparison

Chinese males have significant differences (p < 0.001) between left and right hand in the measurement of hand thickness, hand breadth, wrist circumference, and hand circumference. Females have significant differences (p < 0.001) between left and right hand in the measurement of hand thickness, hand breadth, hand length, thumb length, thumb breadth, and hand circumference. Males and females both have considerable differences in the measurement of hand circumference, hand thickness and hand breadth. Moreover, for right hands, all length dimensions are shorter than left hand, and all breadth and width dimensions are bigger than left hand. This might be result from higher frequency of right-hand use.

It is also worth noting that although the p values of these measurements are below 0.001, the difference percentages of them only range from 0.1 to 4%. So, when producing products of both hands, whether to consider the differences between left and right hand should base on the specific scenery.

4.2 Male-Female Comparison

T-test results and percentage differences suggest a considerable differences (p < 0.001) between Chinese men and women. As expect, males have larger values in all dimensions, and the percentage differences are large (vary from 5.86 to 11.36%). This result indicates a strong necessity to design different product sizes for males and females respectively. For example, mouses for females should be shorter in length at about 7.5% (the percentage difference of hand length between males and females is 7.5%), be narrowed in breadth at about 11.5% (the percentage difference of hand breadth between males and females is 7.5%); straps of wrist-wearing products for females should have 10% shorter than males (the percentage difference of wrist circumference between males and females is 10%); rings for males and females should have an approximate 10% difference in diameter between them (the finger breadth of females are about 10% less than males).

5 Conclusion

In this study, 937 males and females from six representative cities of China were recruited and 3D scanned. Both hands of them were scanned as 3D models and 10 dimensions were measured and analyzed. The result shows basic hands features and percentiles of Chinese male and female. This is the first large-scale measurement of Chinese hands using 3D scanning, which could provide design references for various types of hand related products aimed for Chinese population.

This study compares the differences of hand dimensions between left and right hands. The result reveals that both male and female hands differ significantly in terms of hand thickness, hand breadth, middle finger middle phalanx length, and hand circumference. However, the percentage differences in these measurements are only between 0.1% and 4%, so it is worth considering whether to ignore the differences between the left and right hands when designing products.

In addition, a comparison of hand dimensions between male and female shows that all values are significantly larger for men compared with women. This suggests that in some hand product designs involving both male and female users, it is important to differentiate between them in separate designs. Furthermore, the hand characteristics of the Chinese differ significantly from those of European and American countries, but not so much from those of Asian countries such as Korea and Jordan. The results of this study can be utilized to design hand related products for other Asian countries.