Milan Milivojevic

Mr. Milan Milivojevic

orcid

Research Focus & Metrics

Representative Publications

View All Publications
5 September 2019 in Journal of Biological Engineering
Verified
Journal Article
Open access
PDF file
IPIvana Pajic-Lijakovic
Ivana Pajic-Lijakovic
MMMilan Milivojevic
Milan Milivojevic
Jamming state transition has been used in literature to describe migrating-to-resting cell state transition during collective cell migration without proper rheological confirmation. Yield stress often has been used as an indicator of a jamming state. Yield stress points to the liquid-to-solid state transition, but not a priori to jamming state transition. Various solid states such as elastic solid and viscoelastic solids can be considered in the context of their ability to relax. The relaxation time for (1) an elastic solid tends to zero, (2) Kelvin-Voigt viscoelastic solid is finite, and (3) jamming state tends to infinity. In order to clarify the meaning of jamming state from the rheological standpoint we formulated the constitutive model of this state based on following conditions (1) migration of the system constituents is much damped such that the diffusion coefficient tends to zero, (2) relaxation time tends to infinity, (3) storage and loss moduli satisfy the condition G′(ω)/G"(ω) = const > 1. Jamming state represents the non-linear viscoelastic solid state. The main characteristic of this state is that the system cannot relax. Jamming state transition of multicellular systems caused by collective cell migration is discussed on a model system such as cell aggregate rounding after uni-axial compression between parallel plates based on the data from the literature. Cell aggregate rounding occurs via successive relaxation cycles. Every cycle corresponds to a different scenario of cell migration. Three scenarios were established depending on the magnitude of mechanical and biochemical perturbations (1) ordered scenario with reduced perturbations corresponds to the case that most of the cells migrate, (2) disordered scenario corresponds to the case that some cell groups migrate while the others (at the same time) stay in resting state (corresponds to medium perturbations), and (3) highly suppressed cell migration under large perturbations corresponds to the viscoelastic solid under jamming state. If cells reach the jamming state in one cycle, they are able to overcome this undesirable state and start migrating again in the next cycle by achieving the first or second scenarios again.

1 January 2016 in Telfor Journal
Verified
Journal Article
Closed access
PDF file
MSMarijeta Slavkovic-Ilic
Marijeta Slavkovic-Ilic
AGAna Gavrovska
Ana Gavrovska
MMMilan Milivojevic
Milan Milivojevic
IRIrini Reljin
Irini Reljin
BRBranimir Reljin
Branimir Reljin
The first step in most computer aided diagnosis systems is an accurate segmentation of breast region, which affects not only the accuracy but also the speed of the analysis because it significantly reduces the area of the image to be examined. The second step usually includes removal of pectoral muscle region, which is seen in mediolateral oblique view mammograms. This is primarily done to reduce the number of false positive breast cancer detections. In this paper, a method for the segmentation of breast region based on contrast enhancement and k-means algorithm is proposed. To extract pectoral muscle, a region of interest is found, its contrast is enhanced and the pectoral muscle is identified using k-means algorithm. Cubic polynomial fitting is used for the estimation of muscle's boundary. The method is validated with mammograms from miniMIAS database.

19 May 2019 in Applied Bionics and Biomechanics
Verified
Journal Article
Open access
PDF file
IPIvana Pajic-Lijakovic
Ivana Pajic-Lijakovic
MMMilan Milivojevic
Milan Milivojevic
Mathematical modeling is often used in tissue engineering in order to overcome one of its major challenges: transformation of complex biological and rheological behaviors of cells and tissue in a mathematically predictive and physically manipulative engineering process. The successive accomplishment of this task will greatly help in quantifying and optimizing clinical application of the tissue engineering products. One of the problems emerging in this area is the relation between resting and migrating cell groups, as well as between different configurations of migrating cells and viscoelasticity. A deeper comprehension of the relation between various configurations of migrating cells and viscoelasticity at the supracellular level represents the prerequisite for optimization of the performance of the artificial epithelium. Since resting and migrating cell groups have a considerable difference in stiffness, a change in their mutual volume ratio and distribution may affect the viscoelasticity of multicellular surfaces. If those cell groups are treated as different phases, then an analogous model may be applied to represent such systems. In this work, a two-step Eyring model is developed in order to demonstrate the main mechanical and biochemical factors that influence configurations of migrating cells. This model could be also used for considering the long-time cell rearrangement under various types of applied stress. The results of this theoretical analysis point out the cause-consequence relationship between the configuration of migrating cells and rheological behavior of multicellular surfaces. Configuration of migrating cells is influenced by mechanical and biochemical perturbations, difficult to measure experimentally, which lead to uncorrelated motility. Uncorrelated motility results in (1) decrease of the volume fraction of migrating cells, (2) change of their configuration, and (3) softening of multicellular surfaces.

1 January 2021 in Telfor Journal
Verified
Journal Article
Closed access
PDF file
MMMilan S. Milivojević
Milan S. Milivojević
AGAna Gavrovska
Ana Gavrovska
Entire world has been dealing with the number of new Coronavirus 2 or COVID-19 cases. The spread of this severe acute respiratory syndrome has produced many concerns worldwide. Having data related to coronavirus available for tests, novel models for forecasting the number of new cases can be developed. In this paper, a long short-term memory (LSTM) based methodology is applied for such prediction. Here, experimental analysis is performed with the parameters, such as the number of layers and units of the network. The root mean squared error is calculated for data corresponding to the Republic of Serbia, as well as per different continents. The results show that LSTM model can be useful for further analysis and time series prediction.

1 July 2013
Verified
Conference Paper
Closed access
MSM. Slavkovic
M. Slavkovic
BRB. Reljin
B. Reljin
AGA. Gavrovska
A. Gavrovska
MMM. Milivojevic
M. Milivojevic
This paper proposes a face recognition system based on Gabor image representation, principal component analysis and neural networks. Gabor filter bank at 5 scales and 8 orientations is used for face representation. The dimensionality of Gabor representation of face images is reduced by the Principal component analysis. The images are classified using the Feedforward neural network with backpropagation learning algorithm. The ORL database of faces is used for testing.

1 January 2020 in Telfor Journal
Verified
Journal Article
Open access
PDF file
MĐMiljan Đorđević
Miljan Đorđević
MMMilan Milivojević
Milan Milivojević
AGAna Gavrovska
Ana Gavrovska
Nowadays advantages in face-based modification using DeepFake algorithms made it possible to replace a face of one person with a face of another person. Thus, it is possible to make not only copy-move modifications, but to implement artificial intelligence and deep learning for replacing face movements from one person to another. Still images can be converted into video sequences. Consequently, the contemporaries, historical figures or even animated characters can be lively presented. Deepfakes are becoming more and more successful and it is difficult to detect them in some cases. In this paper we explain the video sequences we produced (e.g. using X2Face method, and First Order Motion Model for Image Animation) and perform deepfake video analysis using SIFT (Scale Invariant Feature Transform) based approach. The experiments show the simplicity in video forgery production, as well as the possible role of SIFT keypoints detection in differentiation between the deeply forged and original video content.