Simulating the wrinkling and aging of skin with a multi-layer finite element model

Abstract

One of the outward signs of the aging process of human skin is the increased appearance of wrinkles on its surface. Clinical studies show that the increased frequency of wrinkles with age may be attributed to changes in the composition of the various layers of skin, leading to a change in mechanical properties. A parameter study was performed on a previously proposed multi-layer finite element model of skin. A region of skin was subject to an in-plane compression, resulting in wrinkling. A number of physical properties of the skin model were changed and the effects these changes had on the size of the subsequent wrinkles were measured. Reducing the moisture content of the stratum corneum by 11% produces wrinkles 25–85% larger. Increasing the dermal collagen fibre density by 67%, results in wrinkles, which are 25–50% larger. A reduction and change in the pre-stress distribution in the skin model, which represents the natural tension and relaxed skin tension lines in real skin, also influences the wrinkle height in a similar manner to real aging skin. Typically, there can be up to a 100% increase in the height of wrinkles as skin ages. This model would be of benefit in the development of cosmetic moisturisers and plastic-surgery techniques to reduce the appearance of aging.

Introduction

As human skin ages, each layer undergoes several biological changes. One of the most visible results of these changes is an increase in the frequency of skin wrinkles as we get older. In the stratum corneum, the ability to bind water within the layer decreases with age (Ramos-e-Silva and da Silva Carneiro, 2007). It is the water that breaks down the hydrogen bonds in the keratin and thus plasticises the stratum corneum (Park and Baddiel, 1972). Reducing the moisture content of this layer results in a stiffer stratum corneum, which leads to a progression of skin wrinkling with age (Batisse et al., 2002). There appears to be a lack of understanding of the process of change in the dermis in aging skin. Some research in the literature report on a decrease in fibre density (Ramos-e-Silva and da Silva Carneiro, 2007). Lovell et al. (1987) conclude that the total collagen content in the dermis remains unchanged during normal aging, while it has also been observed that the collagen fibre density in the dermis as a whole increases as skin ages (Lavker et al. cited in Silver et al. (2001)). Oba and Edwards (2006) showed a correlation between the destruction of the collagen fibres and an increase in wrinkling. According to Piérard et al. (2003), as skin ages there is a loss of collagen fibres in the reticular dermis and hypodermis and this leads to wrinkles. Batisse et al. (2002) report a decrease in fibre density in the upper or papillary dermis but found the whole dermis to be stiffer in aged skin. Biologically, a stiffer dermis may be attributed to an increased density of collagen fibres. There is also a reported increase in collagen cross-links with age, which may also contribute to the increase in the stiffness of the dermis (Wulf et al., 2004). Human skin in vivo is in a state of natural tension, which varies on different parts of the body. The tension is also direction dependent—the direction in which it is greatest is indicated by the Langer lines on the body. As skin ages, it loses some of this natural tension or tone, due in part to the thinning of the underlying hypodermis from weight loss (Quatresooz et al., 2006). This age-related loss of skin tension results in the appearance of more wrinkles (Piérard et al., 2003). The Langer lines or directionality of skin is also reported to change in skin with age. Infant skin is significantly more isotropic than adult skin so that growth can be accommodated (Ruvolo Jr. et al., 2007). This regression in isotropy with age arises as a result of stronger collagen fibre alignment in skin (Lovell et al., 1987; Ruvolo Jr. et al., 2007).

Due to the high visibility and importance of skin, it has been studied in many areas of science and technology such as biomechanics, medicine, forensics, cosmetology and computer animation (Batisse et al., 2002; Bischoff et al., 2000; Bro-Nielsen, 1998; Evans, 2009; Sang and Hodgins, 2006; Thali et al., 2002). Much of this effort has been in the development of mathematical and computer models of human skin. The complexity and quality of these models have evolved significantly over the last few decades, such that several models that can simulate accurately the non-linear, orthotropic and viscoelastic characteristics of skin measured in various in vivo and in vitro experiments (Bischoff, 2006; Shoemaker et al., 1986). Most models, however, assume skin to be a single-layer homogenous material. Previous multi-layer models have also ignored the underlying hypodermis (Magnenat-Thalmann et al., 2002). In addition, few models have been developed to simulate wrinkling.

It has been shown that a three-layer model, consisting of the stratum corneum, dermis and hypodermis, better simulates more complex and realistic skin deformations such as wrinkling as compared with models of fewer layers (Flynn and McCormack, 2008a). The benefits of the three-layer model would be strengthened if it was shown that it could simulate the aging of skin by adjusting the properties of each layer in accordance to what has been biologically observed in real aging skin. A model that realistically simulates aging could aid in developing cosmetic moisturisers and plastic-surgery techniques in order to reduce the appearance of the aging process. Such a model could also improve the quality of animations. Many computer animation models are purely geometrically based (Sang and Hodgins, 2006) and do not take into account the mechanical properties of skin. As a result, the more subtle details such as the wrinkling and furrowing are more difficult to replicate (Terzopoulos and Waters, 1990).

This paper presents further developments of a three-layer model of human skin (Flynn and McCormack, 2008a). The ability of this model to simulate realistic skin behaviour such as wrinkling has already been demonstrated by validation against in vivo skin experiments. The effects of changing the moisture content of the stratum corneum, the collagen fibre density of the dermis and the natural tension inherent in skin on the wrinkle formation are investigated. These changes are compared to what is observed in real aging skin.