Gene selection using pyramid gravitational search algorithm

Genetics play a prominent role in the development and progression of malignant neoplasms. Identification of the relevant genes is a high-dimensional data processing problem. Pyramid gravitational search algorithm (PGSA), a hybrid method in which the number of genes is cyclically reduced is proposed to conquer the curse of dimensionality. PGSA consists of two elements, a filter and a wrapper method (inspired by the gravitational search algorithm) which iterates through cycles. The genes selected in each cycle are passed on to the subsequent cycles to further reduce the dimension. PGSA tries to maximize the classification accuracy using the most informative genes while reducing the number of genes. Results are reported on a multi-class microarray gene expression dataset for breast cancer. Several feature selection algorithms have been implemented to have a fair comparison. The PGSA ranked first in terms of accuracy (84.5%) with 73 genes. To check if the selected genes are meaningful in terms of patient’s survival and response to therapy, protein-protein interaction network analysis has been applied on the genes. An interesting pattern was emerged when examining the genetic network. HSP90AA1, PTK2 and SRC genes were amongst the top-rated bottleneck genes, and DNA damage, cell adhesion and migration pathways are highly enriched in the network.

A novel diagnostic and prognostic approach for unresponsive patients with anthroponotic cutaneous leishmaniasis using artificial neural networks

Cutaneous leishmaniasis (CL) imposes a major health burden throughout the tropical and subtropical regions of the globe. Unresponsive cases are common phenomena occurred upon exposure to the standard drugs. Therefore, rapid detection, prognosis and classification of the disease are crucial for selecting the proper treatment modality. Using machine learning (ML) techniques, this study aimed to detect unresponsive cases of ACL, caused by Leishmania tropica, which will consequently be used for a more effective treatment modality. This study was conducted as a case-control setting. Patients were selected in a major ACL focus from both unresponsive and responsive cases. Nine unique and relevant features of patients with ACL were selected. To categorize the patients, different classifier models such as k-nearest neighbors (KNN), support vector machines (SVM), multilayer perceptron (MLP), learning vector quantization (LVQ) and multipass LVQ were applied and compared for this supervised learning task. Comparison of the receiver operating characteristic graphs (ROC) and confusion plots for the above models represented that MLP was a fairly accurate prediction model to solve this problem. The overall accuracy in terms of sensitivity, specificity and area under ROC curve (AUC) of MLP classifier were 87.8%, 90.3%, 86% and 0.88%, respectively. Moreover, the duration of the skin lesion was the most influential feature in MLP classifier, while gender was the least. The present investigation demonstrated that MLP model could be utilized for rapid detection, accurate prognosis and effective treatment of unresponsive patients with ACL. The results showed that the major feature affecting the responsiveness to treatments is the duration of the lesion. This novel approach is unique and can be beneficial in developing diagnostic, prophylactic and therapeutic measures against the disease. This attempt could be a preliminary step towards the expansion of ML application in future directions.

The effect of parameters of equilibrium-based 3-D biomechanical models on extracted muscle synergies during isometric lumbar exertion

A hallmark of more advanced models is their higher details of trunk muscles represented by a larger number of muscles. The question is if in reality we control these muscles individually as independent agents or we control groups of them called "synergy". To address this, we employed a 3-D biomechanical model of the spine with 18 trunk muscles that satisfied equilibrium conditions at L4/5, with different cost functions. The solutions of several 2-D and 3-D tasks were arranged in a data matrix and the synergies were computed by using non-negative matrix factorization (NMF) algorithms. Variance accounted for (VAF) was used to evaluate the number of synergies that emerged by the analysis, which were used to reconstruct the original muscle activations. It was showed that four and six muscle synergies were adequate to reconstruct the input data of 2-D and 3-D torque space analysis. The synergies were different by choosing alternative cost functions as expected. The constraints affected the extracted muscle synergies, particularly muscles that participated in more than one functional tasks were influenced substantially. The compositions of extracted muscle synergies were in agreement with experimental studies on healthy participants. The following computational methods show that the synergies can reduce the complexity of load distributions and allow reduced dimensional space to be used in clinical settings.

Design and development of a hand robotic rehabilitation device for post stroke patients

Robot-mediated rehabilitation is a rapidly advancing discipline that seeks to develop improved treatment procedures using new technologies, e.g., robotics, coupled with modern theories in neuroscience and rehabilitation. A robotic device was designed and developed for rehabilitation of upper limbs of post stroke patients. A novel force feedback bimanual working mode provided real-time dynamic sensation of the paretic hand. Results of the preliminary clinical tests revealed a quantitative evaluation of the patient's level of paresis and disability.

How does the central nervous system address the kinetic redundancy in the lumbar spine? Three-dimensional isometric exertions with 18 Hill-model-based muscle fascicles at the L4—L5 level

The human motor system is organized for execution of various motor tasks in a different and flexible manner. The kinetic redundancy in the human musculoskeletal system is a significant property by which the central nervous system achieves many complementary goals. An equilibrium-based biomechanical model of isometric three-dimensional exertions of trunk muscles has been developed. Following the definition and role of the uncontrolled manifold, the kinetic redundancy concept is explored in mathematical terms. The null space of the kinetically redundant system when a certain joint moment and/or stiffness are needed is derived and discussed. The aforementioned concepts have been illustrated, using a three-dimensional three-degrees-of-freedom biomechanical model of the spine with 18 anatomically oriented Hill-type-model muscle fascicles. The considerations of stability and its consequence on the internal loading of the spine and coactivation consequences are discussed in both general and specific cases. The results can shed light on the interaction mechanisms in muscle activation patterns seen in various tasks and exertions and can provide a significant understanding for future research studies and clinical practices related to low-back disorders. Alteration of recruitment patterns in low-back-pain patients has been explained on the basis of this biomechanical analysis. The higher coactivation results in higher internal loading while providing higher joint stiffness that enhances spinal stability, which guards against spinal deformation in the presence of any perturbations.