 Research
 Open access
 Published:
Fixed points of multivalued contraction mappings in modular metric spaces
Fixed Point Theory and Applications volume 2014, Article number: 249 (2014)
Abstract
The purpose of this paper is to study the existence of fixed points for contractivetype multivalued maps in the setting of modular metric spaces. The notion of a modular metric on an arbitrary set and the corresponding modular spaces, generalizing classical modulars over linear spaces like Orlicz spaces, were recently introduced. In this paper we investigate the existence of fixed points of multivalued modular contractive mappings in modular metric spaces. Consequently, our results either generalize or improve fixed point results of Nadler (Pac. J. Math. 30:475488, 1969) and Edelstein (Proc. Am. Math. Soc. 12:710, 1961).
MSC:47H09, 46B20, 47H10, 47E10.
1 Introduction
The aim of this paper is to give an outline of a fixed point theory for multivalued Lipschitzian mappings defined on some subsets of modular metric spaces. Modular metric spaces were introduced in [1, 2]. The way we approached the concept of modular metric spaces is different. Indeed we look at these spaces as the nonlinear version of the classical modular spaces as introduced by Nakano [3] on vector spaces and modular function spaces introduced by Musielak [4] and Orlicz [5]. In [6] the authors have defined and investigated the fixed point property in the framework of modular metric space and introduced the analog of the Banach contraction principle theorem in modular metric space.
As is well known, a fixed point theorem for multivalued contraction mappings was established by Nadler [7]. In 1961 Edelstein [8] has generalized the Banach contraction principle to mappings satisfying a less restrictive Lipschitz inequality such as local contraction. This result has been generalized to a multivalued version by Nadler [7]. On the other hand Mizoguchi and Takahashi [9] have improved Reich’s result [10] and proved the existence of fixed points for multivalued maps in the case when values of mappings are closed bounded instead of compact.
In this paper we define the Hausdorff modular metric and obtain a multivalued version of the result [[6], Theorem 3.1] in modular metric spaces. We also extend the results of Nadler [7], Mizoguchi and Takahashi [9] to modular metric spaces. The linear version of some of our results may be found in the work of Kutbi and Latif [11].
For more on metric fixed point theory, the reader may consult the book [12].
2 Basic definitions and properties
Let X be a nonempty set. Throughout this paper for a function \omega :(0,\mathrm{\infty})\times X\times X\to (0,\mathrm{\infty}), we will write
for all \lambda >0 and x,y\in X.
A function \omega :(0,\mathrm{\infty})\times X\times X\to [0,\mathrm{\infty}] is said to be a modular metric on X if it satisfies the following axioms:

(i)
x=y if and only if {\omega}_{\lambda}(x,y)=0, for all \lambda >0;

(ii)
{\omega}_{\lambda}(x,y)={\omega}_{\lambda}(y,x), for all \lambda >0, and x,y\in X;

(iii)
{\omega}_{\lambda +\mu}(x,y)\le {\omega}_{\lambda}(x,z)+{\omega}_{\mu}(z,y), for all \lambda ,\mu >0 and x,y,z\in X.
If instead of (i), we have only the condition
then ω is said to be a pseudomodular (metric) on X. A modular metric ω on X is said to be regular if the following weaker version of (i) is satisfied:
Finally, ω is said to be convex if for \lambda ,\mu >0 and x,y,z\in X, it satisfies the inequality
Note that for a metric pseudomodular ω on a set X, and any x,y\in X, the function \lambda \to {\omega}_{\lambda}(x,y) is nonincreasing on (0,\mathrm{\infty}). Indeed, if 0<\mu <\lambda, then
Let ω be a pseudomodular on X. Fix {x}_{0}\in X. The two sets
and
are said to be modular spaces (around {x}_{0}).
We obviously have {X}_{\omega}\subset {X}_{\omega}^{\ast}. In general this inclusion may be proper. It follows from [1, 2] that if ω is a modular on X, then the modular space {X}_{\omega} can be equipped with a (nontrivial) metric, generated by ω and given by
for any x,y\in {X}_{\omega}. If ω is a convex modular on X, according to [1, 2] the two modular spaces coincide, i.e. {X}_{\omega}^{\ast}={X}_{\omega}, and this common set can be endowed with the metric {d}_{\omega}^{\ast} given by
for any x,y\in {X}_{\omega}. These distances will be called Luxemburg distances.
First attempts to generalize the classical function spaces of the Lebesgue type {L}^{p} were made in the early 1930s by Orlicz and Birnbaum in connection with orthogonal expansions. Their approach consisted in considering spaces of functions with some growth properties different from the power type growth control provided by the {L}^{p}norms. Namely, they considered the function spaces defined as follows:
where \phi :[0,\mathrm{\infty}]\to [0,\mathrm{\infty}] was assumed to be a convex function increasing to infinity, i.e. the function which to some extent behaves similarly to power functions \phi (t)={t}^{p}. Modular function spaces {L}^{\phi} furnishes a wonderful example of a modular metric space. Indeed define the function ω by
for all \lambda >0, and f,g\in {L}^{\phi}. Then ω is a modular metric on {L}^{\phi}. Moreover, the distance {d}_{\omega}^{\ast} is exactly the distance generated by the Luxemburg norm on {L}^{\phi}.
For more examples on modular function spaces, the reader may consult the book of Kozlowski [13] and for modular metric spaces [1, 2].
Definition 2.3 Let {X}_{\omega} be a modular metric space.

(1)
The sequence {\{{x}_{n}\}}_{n\in \mathbb{N}} in {X}_{\omega} is said to be ωconvergent to x\in {X}_{\omega} if and only if {\omega}_{1}({x}_{n},x)\to 0, as n\to \mathrm{\infty}. x will be called the ωlimit of \{{x}_{n}\}.

(2)
The sequence {\{{x}_{n}\}}_{n\in N} in {X}_{\omega} is said to be ωCauchy if {\omega}_{1}({x}_{m},{x}_{n})\to 0, as m,n\to \mathrm{\infty}.

(3)
A subset M of {X}_{\omega} is said to be ωclosed if the ωlimit of a ωconvergent sequence of M always belongs to M.

(4)
A subset M of {X}_{\omega} is said to be ωcomplete if any ωCauchy sequence in M is a ωconvergent sequence and its ωlimit is in M.

(5)
A subset M of {X}_{\omega} is said to be ωbounded if we have
{\delta}_{\omega}(M)=sup\{{\omega}_{1}(x,y);x,y\in M\}<\mathrm{\infty}. 
(6)
A subset M of {X}_{\omega} is said to be ωcompact if for any \{{x}_{n}\} in M there exists a subset sequence \{{x}_{{n}_{k}}\} and x\in M such that {\omega}_{1}({x}_{{n}_{k}},x)\to 0.

(7)
ω is said to satisfy the Fatou property if and only if for any sequence {\{{x}_{n}\}}_{n\in \mathbb{N}} in {X}_{\omega} ωconvergent to x, we have
{\omega}_{1}(x,y)\le \underset{n\to \mathrm{\infty}}{lim\hspace{0.17em}inf}{\omega}_{1}({x}_{n},y),
for any y\in {X}_{\omega}.
In general if {lim}_{n\to \mathrm{\infty}}{\omega}_{\lambda}({x}_{n},x)=0, for some \lambda >0, then we may not have {lim}_{n\to \mathrm{\infty}}{\omega}_{\lambda}({x}_{n},x)=0, for all \lambda >0. Therefore, as is done in modular function spaces, we will say that ω satisfies the {\mathrm{\Delta}}_{2}condition if this is the case, i.e. {lim}_{n\to \mathrm{\infty}}{\omega}_{\lambda}({x}_{n},x)=0, for some \lambda >0 implies {lim}_{n\to \mathrm{\infty}}{\omega}_{\lambda}({x}_{n},x)=0, for all \lambda >0. In [1] and [2], one will find a discussion as regards the connection between ωconvergence and metric convergence with respect to the Luxemburg distances. In particular, we have
for any \{{x}_{n}\}\in {X}_{\omega} and x\in {X}_{\omega}. In particular we see that ωconvergence and {d}_{\omega}convergence are equivalent if and only if the modular ω satisfies the {\mathrm{\Delta}}_{2}condition. Moreover, if the modular ω is convex, then we know that {d}_{\omega}^{\ast} and {d}_{\omega} are equivalent, which implies
for any \{{x}_{n}\}\in {X}_{\omega} and x\in {X}_{\omega} [1, 2].
Definition 2.4 Let (X,\omega ) be a modular metric space. We will say that ω satisfies the {\mathrm{\Delta}}_{2}type condition if, for any \alpha >0, there exists C>0 such that
for any \lambda >0, x,y\in {X}_{\omega}, with x\ne y.
Note that if ω satisfies {\mathrm{\Delta}}_{2}type condition, then ω satisfies the {\mathrm{\Delta}}_{2}condition. The above definition will allow us to introduce the growth function in the modular metric spaces as was done in the linear case.
Definition 2.5 Let (X,\omega ) be a modular metric space. Define the growth function Ω by
for any \alpha >0.
The following properties were proved in the linear case in [14].
Lemma 2.1 Let (X,\omega ) be a modular metric space. Assume that ω is a convex regular modular which satisfies the {\mathrm{\Delta}}_{2}type condition. Then

(1)
\mathrm{\Omega}(\alpha )<\mathrm{\infty}, for any \alpha >0,

(2)
Ω is a strictly increasing function, and \mathrm{\Omega}(1)=1,

(3)
\mathrm{\Omega}(\alpha \beta )\le \mathrm{\Omega}(\alpha )\mathrm{\Omega}(\beta ), for any \alpha ,\beta \in (0,\mathrm{\infty}),

(4)
{\mathrm{\Omega}}^{1}(\alpha ){\mathrm{\Omega}}^{1}(\beta )\le {\mathrm{\Omega}}^{1}(\alpha \beta ), where {\mathrm{\Omega}}^{1} is the function inverse of Ω,

(5)
for any x,y\in {X}_{\omega}, x\ne y, we have
{d}_{\omega}^{\ast}(x,y)\le \frac{1}{{\mathrm{\Omega}}^{1}(1/{\omega}_{1}(x,y))}.
Proof It is clear that the {\mathrm{\Delta}}_{2}type condition implies (1) and from the definition of Ω we have \mathrm{\Omega}(1)=1. Next we prove that Ω is strictly increasing. Let \alpha <\beta. Since ω is convex, we get
which implies
for any x,y\in {X}_{\omega}. From this inequality, we can easily deduce the following relation:
for any \alpha <\beta. The properties (3) and (4) follow the same line as the proofs developed in [14]. As for the property (5), note that
The inequality in (5) follows from the definition of the distance {d}_{\omega}^{\ast}. □
The following technical lemma will be useful later on in this work.
Lemma 2.2 Let (X,\omega ) be a modular metric space. Assume that ω is a convex regular modular which satisfies the {\mathrm{\Delta}}_{2}type condition. Let \{{x}_{n}\} be a sequence in {X}_{\omega} such that
where K is an arbitrary nonzero constant and \alpha \in (0,1). Then \{{x}_{n}\} is Cauchy for both ω and {d}_{\omega}^{\ast}.
Proof Under the above assumptions, we have
where we assumed that {x}_{n}\ne {x}_{n+1}. Hence
for n\ge 1, which implies
for any n\ge 1. Note that this inequality is still valid when {x}_{n+1}={x}_{n}. Since \alpha <1, and {\mathrm{\Omega}}^{1} is strictly increasing, we get
which implies \frac{1}{{\mathrm{\Omega}}^{1}(\frac{1}{\alpha})}<1. Classical analysis on metric spaces implies that {\sum}_{n\ge 1}{d}_{\omega}^{\ast}({x}_{n+1},{x}_{n}) is convergent which implies that \{{x}_{n}\} is Cauchy for {d}_{\omega}^{\ast}. Since ω satisfies the {\mathrm{\Delta}}_{2}type condition, \{{x}_{n}\} is Cauchy for ω. □
Note that this lemma is crucial since the main assumption on \{{x}_{n}\} will not be enough to imply that \{{x}_{n}\} is ωCauchy since ω fails the triangle inequality.
3 Multivalued mappings in modular metric spaces
At this point we introduce some notation which will be used throughout the remainder of this work. For a subset M of modular metric space {X}_{\omega}. Set

(i)
CB(M)=\{C:C\text{is nonempty}\omega \text{closed and}\omega \text{bounded subset of}M\};

(ii)
K(M)=\{C:C\text{is nonempty}\omega \text{compact subset of}M\};

(iii)
the Hausdorff modular metric is defined on CB(M) by
{H}_{\omega}(A,B)=max\{\underset{x\in A}{sup}{\omega}_{1}(x,B),\underset{y\in B}{sup}{\omega}_{1}(y,A)\},
where {\omega}_{1}(x,B)={inf}_{y\in A}{\omega}_{1}(x,y).
Definition 3.1 Let (X,\omega ) be a modular metric space and M be a nonempty subset of {X}_{\omega}. A mapping T:M\to CB(M) is called a multivalued Lipschitzian mapping if there exists a constant k\ge 0 such that
A point x\in M is called a fixed point of T whenever x\in T(x). The set of fixed points of T will be denoted by Fix(T).
In [6], we defined Lipschitzian single valued maps. Our definition is more general than the one used by Chistyakov [1, 2]. Indeed, in the case of modular function spaces, it is proved in [15] that
for any \lambda >0 if and only if
Moreover, an example is given to show that
but T is not Lipschitzian with respect to {d}_{\omega} with constant 1.
Definition 3.2 Let (X,\omega ) be a modular metric space and M be a nonempty subset of {X}_{\omega}. A multivalued mapping T:M\to CB(M) is called

(i)
an ωcontraction if there exists a constant k\in [0,1) such that for any x,y\in M,
{H}_{\omega}(T(x),T(y))\le k{\omega}_{1}(x,y); 
(ii)
a (\epsilon ,k)ωuniformly locally contraction if there exists a constant k\in [0,1) such that for any x,y\in M,
{H}_{\omega}(T(x),T(y))\le k{\omega}_{1}(x,y),\phantom{\rule{1em}{0ex}}\text{whenever}{\omega}_{1}(x,y)\epsilon .
Before we state our results, we will need the following technical lemmas [7] in the setting of modular metric spaces.
Lemma 3.1 Let (X,\omega ) be a modular metric space and M be a nonempty subset of {X}_{\omega}. Let A,B\in CB(M), then for each \epsilon >0 and x\in A, there exists y\in B such that
Moreover, if B is ωcompact and ω satisfies the Fatou property, then for any x\in A, there exists y\in B such that
Proof The proof of the first part is easy. As for the second part, assume B is ωcompact and ω satisfies the Fatou property. Let x\in A. Then for any n\ge 1, there exists {y}_{n}\in B such that
Since B is ωcompact, we may assume that \{{y}_{n}\} ωconverges to y\in B. Since ω satisfies the Fatou property, we get
□
Lemma 3.2 Let (X,\omega ) be a modular metric space. Assume that ω satisfies {\mathrm{\Delta}}_{2}condition. Let M be a nonempty subset of {X}_{\omega}. Let {A}_{n} be a sequence of sets in CB(M), and suppose {lim}_{n\to \mathrm{\infty}}{H}_{\omega}({A}_{n},{A}_{0})=0 where {A}_{0}\in CB(M). Then if {x}_{n}\in {A}_{n} and {lim}_{n\to \mathrm{\infty}}{x}_{n}={x}_{0}, it follows that {x}_{0}\in {A}_{0}.
Proof Using Lemma 3.1, for every n\ge 1, there exists {y}_{n}\in {A}_{0} such that
Hence
which implies {lim}_{n\to \mathrm{\infty}}{\omega}_{2}({y}_{n},{x}_{0})=0. Since ω satisfies the {\mathrm{\Delta}}_{2}condition, we have {lim}_{n\to \mathrm{\infty}}{\omega}_{1}({y}_{n},{x}_{0})=0. Since {A}_{0} is ωclosed, we have {x}_{0}\in {A}_{0}. □
4 The main results
The statement of Nadler’s fixed point result [7] in modular metric spaces is as follows.
Theorem 4.1 Let (X,\omega ) be a modular metric space. Assume that ω is a convex regular modular which satisfies the {\mathrm{\Delta}}_{2}condition. Let M be a nonempty ωcomplete subset of {X}_{\omega}. Let T:M\to CB(M) be an ωcontraction map. Then T has a fixed point.
Proof Since T is an ωcontraction, there exists a constant k\in [0,1) such that for any x,y\in M, we have
Let {x}_{0} be an arbitrary but fixed element of M and {x}_{1}\in T({x}_{0}), then from Lemma 3.1, there exists {x}_{2}\in T({x}_{1}) such that
Similarly, there exists {x}_{3}\in T({x}_{2}) such that
By induction we build \{{x}_{n}\} such that
and {x}_{n+1}\in T({x}_{n}), for every n\ge 0. Since T is a contraction, we get
for every n\ge 0. Hence
for every n\ge 0. The technical Lemma 2.2 implies that \{{x}_{n}\} is ωCauchy. Since M is ωcomplete, \{{x}_{n}\} ωconverges to some point {x}_{0}\in M. Since
we conclude that {lim}_{n\to \mathrm{\infty}}{H}_{\omega}(T({x}_{n}),T({x}_{0}))=0. Since {x}_{n+1}\in T({x}_{n}), Lemma 3.2 implies {x}_{0}\in T({x}_{0}), i.e. {x}_{0} is fixed point of T. This completes the proof of Theorem 4.1. □
Edelstein [8] has extended the classical fixed point theorem for a contraction to the case when X is a complete εchainable metric space and the mapping T:X\to X is an (\epsilon ,k)uniformly locally contraction. This result was extended by Nadler [7] to multivalued mappings. Here we investigate Nadler’s result in modular metric spaces. First let us introduce the εchainable concept in modular metric spaces. Our definition is slightly different from the one used in the classical metric spaces since the modular functions fail in general the triangle inequality.
Definition 4.1 Let (X,\omega ) be a modular metric space. A nonempty subset M\subset {X}_{\omega} is said to be finitely εchainable (where \epsilon >0 is fixed) if and only if there exists N\ge 1 such that for any a,b\in M there is an N,\epsilonchain from a to b (that is, a finite set of points {x}_{0},{x}_{1},\dots ,{x}_{N}\in M such that {x}_{0}=a, {x}_{N}=b, and {\omega}_{1}({x}_{i},{x}_{i+1})<\epsilon, for all i=0,1,2,\dots ,N1).
We have the following result.
Theorem 4.2 Let (X,\omega ) be a modular metric space. Assume that ω is a convex regular modular which satisfies the {\mathrm{\Delta}}_{2}type condition and the Fatou property. Let M be a nonempty ωcomplete and ωbounded subset of {X}_{\omega}, which is finitely εchainable, for some fixed \epsilon >0. Let T:M\to K(M) be an (\epsilon ,k)ωuniformly locally contraction map. Then T has a fixed point in M.
Proof Since M is finitely εchainable, there exists N\ge 1 such that for any a,b\in M there is a finite set of points {x}_{0},{x}_{1},\dots ,{x}_{N}\in M such that {x}_{0}=a, {x}_{N}=b, and {\omega}_{1}({x}_{i},{x}_{i+1})<\epsilon, for all i=0,1,2,\dots ,N1. For any x,y\in M define
where the infimum is taken over all N,\epsilonchains {x}_{0},{x}_{1},\dots ,{x}_{N} from x to y. Our assumptions imply that {\omega}^{\ast}(x,y)<\mathrm{\infty}, for any x,y\in M. Using the basic properties of ω, we get
for any x,y\in M. Moreover, if {\omega}_{1}(x,y)<\epsilon, then we have {\omega}^{\ast}(x,y)\le {\omega}_{1}(x,y), for any x,y\in M. Fix x\in M. Set {z}_{0}=x. Choose {z}_{1}\in T({z}_{0}). Let {x}_{0},\dots ,{x}_{N} be an N,\epsilonchain from {z}_{0} to {z}_{1}. Such an N,\epsilonchain exists since M is finitely εchainable. Using Lemma 3.1, there exists {y}_{1}\in T({x}_{1}) such that {\omega}_{1}({z}_{0},{y}_{1})\le H(T({z}_{0}),T({x}_{1})). Repeating this process we find {y}_{2},\dots ,{y}_{N} such that {y}_{i}\in T({x}_{i}) and {\omega}_{1}({y}_{i},{y}_{i+1})\le H(T({x}_{i}),T({x}_{i+1})). It is easy to check that {z}_{0},{y}_{1},\dots ,{y}_{N} is an N,\epsilonchain from {z}_{0} to {y}_{N}\in T({z}_{1}). Set {y}_{N}={z}_{2}. Using the fact that T is an (\epsilon ,k)ωuniformly locally contraction map, we get
By induction, we construct a sequence \{{z}_{n}\}\in M such that
and {z}_{n+1}\in T({z}_{n}), for any n\ge 1. Obviously we have {\omega}^{\ast}({z}_{n},{z}_{n+1})\le {k}^{n}{\omega}^{\ast}({z}_{0},{z}_{1}), for any n\ge 1. Since ω satisfies the {\mathrm{\Delta}}_{2}type condition, there exists C>0 such that
for any n\ge 1. Lemma 2.2 implies that \{{z}_{n}\} is ωCauchy. Since M is ωcomplete, \{{z}_{n}\} ωconverges to some z\in M. We claim that z is a fixed point of T. Indeed using the ideas developed above, there exists {y}_{n}\in T(z) such that
for any n\ge 1. Since
for any n\ge 1, we conclude that \{{y}_{n}\} ωconverges to z. Since T(z) is ωclosed, we get z\in T(z). □
References
Chistyakov VV: Modular metric spaces, I: basic concepts. Nonlinear Anal. 2010, 72(1):1–14. 10.1016/j.na.2009.04.057
Chistyakov VV: Modular metric spaces, II: application to superposition operators. Nonlinear Anal. 2010, 72(1):15–30. 10.1016/j.na.2009.04.018
Nakano H: Modulared SemiOrdered Linear Spaces. Maruzen, Tokyo; 1950.
Musielak J Lecture Notes in Math. 1034. In Orlicz Spaces and Modular Spaces. Springer, Berlin; 1983.
Orlicz W: Collected Papers. Parts I, II. PWN, Warsaw; 1988.
Abdou AAN, Khamsi MA: Fixed point results of pointwise contractions in modular metric spaces. Fixed Point Theory Appl. 2013., 2013: Article ID 163
Nadler SB Jr.: Multivalued contraction mappings. Pac. J. Math. 1969, 30: 475–488. 10.2140/pjm.1969.30.475
Edelstein M: An extension of Banach’s contraction principle. Proc. Am. Math. Soc. 1961, 12: 7–10.
Mizoguchi N, Takahashi W: Fixed point theorems for multivalued mappings on complete metric spaces. J. Math. Anal. Appl. 1989, 141(1):177–188. 10.1016/0022247X(89)90214X
Reich S: Some fixed point problems. Atti Accad. Naz. Lincei, Rend. Cl. Sci. Fis. Mat. Nat. 1974, 57: 194–198.
Kutbi MA, Latif A: Fixed points of multivalued maps in modular function spaces. Fixed Point Theory Appl. 2009., 2009: Article ID 786357 10.1155/2009/786357
Khamsi MA, Kirk WA: An Introduction to Metric Spaces and Fixed Point Theory. Wiley, New York; 2001.
Kozlowski WM Series of Monographs and Textbooks in Pure and Applied Mathematics 122. In Modular Function Spaces. Dekker, New York; 1988.
Dominguez Benavides T, Khamsi MA, Samadi S: Uniformly Lipschitzian mappings in modular function spaces. Nonlinear Anal., Theory Methods Appl. 2001, 46(2):267–278. 10.1016/S0362546X(00)001176
Khamsi MA, Kozlowski WK, Reich S: Fixed point theory in modular function spaces. Nonlinear Anal. 1990, 14: 935–953. 10.1016/0362546X(90)90111S
Acknowledgements
The first author was supported by the Deanship of Scientific Research (DSR), King Abdoulaziz University. The author, therefore, acknowledge with thanks DSR.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors participated in the design of this work and performed equally. All authors read and approved the final manuscript.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Abdou, A.A., Khamsi, M.A. Fixed points of multivalued contraction mappings in modular metric spaces. Fixed Point Theory Appl 2014, 249 (2014). https://doi.org/10.1186/168718122014249
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/168718122014249