deletions | additions       diff --git a/section_Review_of_Related_Literature__.tex b/section_Review_of_Related_Literature__.tex  index cf76ec3..4e88d26 100644  --- a/section_Review_of_Related_Literature__.tex  +++ b/section_Review_of_Related_Literature__.tex 
...
\subsection{Related Studies on Electromyography}  \subsection{Related Studies on EMG used in Computer Science}  In the game industry, EMG technology has been widely used to replace physical components, such as the traditional joysticks and keyboards with something virtual. In a study conducted by Wheeler and Jorgensen (2003), it has been understood that there are two forms of most used gesture-recognition systems for receiving inputs. First is through image processing with an external camera as the source of input. Second is through muscle sensors such as the wearable dry-electrode sleeve device they have developed to sense EMG signal as computer inputs. These EMG electrodes work by detecting skin currents with a very low-impendence connection with the skin. It receives the currents that travel in the muscle fiber from the innervation point to the end of the muscle. This device was tested using a virtual number pad and in their case, the participant had to be extra careful and precise with each movement because of the sensitivity and difficulty of distinguishing the keys that were hit.  As mentioned above, another gesture-recognition system is image processing through an external camera. In the study performed by Rautaray and Agrawal (2011), various image processing techniques were applied for hand tracking and gesture recognition in a virtual gaming environment such as Camshift, Lucas Kanade, Haar etc. The different gestures used for the game interactions were grab, punch and go.  Several studies have been conducted about the implementation of EMG based gestures on therapeutic games for rehabilitation purposes. An alternate interface has been developed in the study conducted by Armiger and Voggelstein (2008) for Guitar Hero® using surface EMG to train and assess the performance of upper-extremity amputees. Instead of using the guitar, EMG electrodes were used to record the myoelectric activity. After recording, the acquired data is processed in real-time using pattern recognition algorithms to classify the gestures and then use them to control the game. The scores obtained by the amputees were relatively lower than those of the non-amputees.  Another study about therapeutic games was performed by Viriyasaksathian et al. (2011). EMG gestures were applied to an augmented reality game for the upper-limb rehabilitation of stroke rehabilitation. The combination of music synchronization, biofeedback technology and augmented reality was employed to attract the attention of stroke patients since existing therapy methods are often boring thus results to lack of motivation.  \subsection{Related Studies on Machine Learning in processing EMG based gestures}  Several studies related to EMG pattern recognition have been conducted over the past decades. These techniques have been used to analyze EMG signals which have been complex to recognize due to large variations in signals. In a study conducted by Liu et al.\cite{Liu_2007}, a novel EMG classifier called cascaded kernel learning machine (CKLM) was proven to be effective, achieving a high recognition rate of 93.54\%. The study employed a cascaded architecture of kernel learning machines including the General Discriminant Analysis (GDA), and the support vector machine (SVM) which offers classification performance that matches or exceeds other classifiers and does so in a computationally efficient manner \cite{Oskoei_2008}.