This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
This chapter focuses on the background information from previous researcher that is considered significant in development of this research, approach that have used for then Artificial Neural Network especially in Kohonen Neural Network and Backpropagation Neural Network. They are many sources for literature review which are journals, articles, past student research paper or thesis, books, encyclopaedia, online source such as at Science Direct, IEEE and others. The explanation will be done sequentially where it will help the development process in this research.
Southeast Asia is home to a wide diversity of snake species occurring in Malaysia, Singapore, Thailand, Borneo, Indonesia, Philippines and everywhere else in this region. Peninsular Malaysia and Borneo alone are home to possibly more than 200 different species of snakes that occupy all levels of tropical rainforest. These snakes come in a kaleidoscope of colors' with a wide range of shapes and sizes. Snakes also play an important ecological role in preserving the natural environment through their predatory role of feeding on frogs, lizards, mammals, fish, insects and even other snakes.
In Malaysia, and around coastal waters of the region, according to Tan N.H(n.d) there are at least 18 different species venomous front fanged land snakes and more than 22 different species of sea snakes. These venomous are divided to other 5 subfamilies. They are Crotaline, Elapinae, Laticaudinae, Hydrophiini and Ephalipiini. Crotalinae and Elapinae are land snake types and Laticaudinae, Hydrophiini and Ephalopiini are sea snake types (Tan N.H, 2000).
TheÂ Crotalinae, or crotalines, are aÂ subfamilyÂ ofÂ venomousÂ vipersÂ found inÂ AsiaÂ and theÂ Americas. They are distinguished by the presence of a heat-sensing pit organ located between the eye and the nostril on either side of the head. These snakes range in size from the diminutive hump-nosed viper, hey all share a common characteristic: a deep pit, or fossa, in the loreal area between the eye and the nostril on either side of the head. TheseÂ loreal pitsÂ are the external openings to a pair of extremely sensitiveÂ infraredÂ detecting organs, which in effect give the snakes a sixth sense that helps them to find and perhaps even judge the size of the small warm-blooded prey on which they feed (Campbell & Lamar, 2004)
Elapinae And Hydrophiini
ElapidaeÂ or in GreekÂ called éllopsÂ which meaning to sea-fishÂ is aÂ familyÂ ofÂ venomousÂ snakesÂ found inÂ tropicalÂ andÂ subtropicalÂ regions around the world, terrestrially inÂ AsiaÂ andÂ North AmericaÂ and aquatically in theÂ PacificÂ andÂ Indian Oceans.Â All elapids have a pair ofÂ proteroglyphousÂ fangs that are used to injectÂ venomÂ from glands located towards the rear of the upper jaws. In outward appearance terrestrialÂ elapids look similar to theÂ colubridae: almost all have long and slender bodies with smooth scales, a head that is covered with large shields and not always distinct from the neck, and eyes with round pupils. In addition, their behavior is usually quite active and most areÂ oviparous. The fangs are the first two teeth on eachÂ maxillaryÂ bone, which are enlarged and hollow, and usually only one is in place on each side at any time. The maxilla is intermediate in length and mobility between typical colubrids (long, less mobile) and viperids (very short, highly mobile).Â
TheÂ Laticauda is the least adapted to sea life of all the members of Hydrophiidae; it retains the wide ventral scales typical of terrestrial snakes and has only a poorly developed tail fin.Â LaticaudaÂ are adapted to living on land and in shallow seas.. Litacauda often active at night, which is when they prefer to hunt. Even though they contain one of the most toxic venoms in the world (their bite is ten times more toxic than that of theÂ King Cobra),Â LaticaudaÂ are usually not aggressive towards humans, and inÂ New Caledonia, where they are calledÂ tricot rayéÂ ("stripeyÂ sweater"), children play with them. Bites are extremely rare, but must be treated immediately.
Snakebite Cases in Malaysia
Most of the cases of snake bites in Malaysia are due to the Malayan pit viper (Agkistrodon rhodostoma) (Reid et al, 1963; Lim and Abu Bakar, 1970; Muthusamy, 1988; Lim, 1990). According to Tan(2000) snakebite is s serious medical problem in Malaysia. He add that from 1958 to 1980, there were 50 and half thousand cases of recorded in the hospitals in Malaysia. (Tan N.H, 2000).
Table 1.1: The Management of Snakebites in Malaysia
Malayan pit viper (Calloselasma rhodostoma)
Asian common cobra (Naja naja)
Asian lance-headed viper (Trimeresurus)
King cobra (Ophiophagus hannah)
Krait (Bun garus)
(Tan N.H, 2000)
From that statistic we can conclude that most of the cases of snakebite in Malaysia in unidentified cases. This is serious problem because the initial step to helping snakebite victim is to identify snake type for further medical treatment. Even doctor will give polyvalent to unidentified cases; this doesn't mean that it is good solution.
Snakebite symptoms' can divide to several categories by what type of venom poisoning victim showing a signs. We do research on Elapid Venom Poisoning, Pit Viper Venom Poisoning and Sea Snake Venom Poisoning. Snakebite symptoms' can divide to several categories by what type of venom poisoning victim showing a signs. There are Elapid Poisoning Venom, Pit Viper Venom Poisoning and Sea Snake Venom Poisoning.
Elapid Venom Poisoning
All elapids are venomous and many are potentially deadly. The venoms are mostlyÂ neurotoxicÂ and are considered more dangerous than the mainlyÂ proteolytic venomsÂ ofÂ vipers. According to NSW Health (2007) most victim from Elapid Venom Poisoning will show paralytic sign.
Ptosis : Droopiness of a body part especially at eyelids.
Opthalmoplegia: Paralysis of the one or moreÂ extraocular muscles which are responsible forÂ eye movement.
Fixed dilated pupils: SuddenÂ pupillaryÂ dilation and loss of ability to constrict in response to light.
Dysarthria: MotorÂ speech disorderÂ resulting fromÂ neurological injury,characterized by poor articulation.
Dysphalgia : Symptom of difficulty in swallowing.
Tongue protrusion : Also called "reverse" or "immature" swallow is the common name given to or facial muscular imbalance, a human behavioral pattern in which the tongue protrudes through the anterior incisorsÂ duringÂ swallowing,Â speechÂ and whiles the tongue is at rest
Limb weakness, Respiratory weakness and Peak flow rate
Pit Viper Venom Poisoning
Viperid venoms typically contain an abundance ofÂ protein-degrading enzymes, calledÂ proteases that produce symptoms such as pain, strong local swelling andÂ necrosis, blood loss from cardiovascular damage complicated byÂ coagulopathy, and disruption of the blood clotting system. Death is usually caused by collapse in blood pressure. Proteolytic venom is also dual-purpose: it is used for defense and to immobilize prey, as with neurotoxic venoms, and also many of the enzymes have a digestive function, breaking down molecules in prey items, such asÂ lipids,Â nucleic acids, and proteins. (Slowinski, 2000).Â This is important, as many vipers have weak digestive systems. According to NSW Health (2007) most victim from Pit Viper Venom Poisoning will show coagulopathy signs.
Persistant blood ooze.
Haematuria: the presence ofÂ red blood cellsÂ (erythrocytes) in theÂ urine. It may beÂ idiopathicÂ and/orÂ benign, or it can be aÂ sign that there is aÂ kidney stoneÂ or aÂ tumorÂ in theÂ urinary tractÂ (kidneys,Â ureters,Â urinary bladder,Â prostate, andÂ urethra), ranging from trivial to lethal. IfÂ white blood cellsÂ are found in addition to red blood cells, then it is a signal of urinary tract infection.
And Active bleeding.
Sea Snake Venom Poisoning
Same like Elapidae family, the majority of sea snakes are highly venomous; however, when bites occur, it is rare for much venom to be injected, so that envenomation symptoms usually seem non-existent or trivial (Â Food and Agriculture Organization of the United Nations [FAO], 2007). Â For example,Â Palmaris platurusÂ has venom more potent than any other terrestrial snake species inÂ Costa Rica, but despite its abundance in the waters off its western coast, few human fatalities have been reported (Campbell & Lamar, 2004). Nevertheless, all sea snakes should be handled with great caution (FAO, 2007)
Myoglobinuria: Presence ofÂ myoglobinÂ in theÂ urine, usually associated withÂ rhabdomyolysisÂ or muscle destruction. Myoglobin is present inÂ muscle cells as a reserve ofÂ oxygen.
Snake Fang Mark
According to Nishioka, Silveira & Bauab (1995), bites marks are useful for the differential diagnosis of snakebite. Different of diagnosis between venomous and non-venomous snakebite is very important cause victims of the former can benefit from treatment with specific anti-venin. This will help preventing a doctor to use polyvalent and also monovalent(for non venomous snake case) in meaning of preventing a long term side effect of using anti-venom.
ARTIFICIAL NEURAL NETWORK (ANN)
AnÂ artificial neural network (ANN), usually called "neural network" (NN), is aÂ mathematical modelÂ orÂ computational modelÂ that tries to simulate the structure and/or functional aspects ofÂ biological neural networks. The human brain incorporates nearly 10 billion neurons and 60 trillion connections, synapse, between them. (Sherpherd and Koch,1990). It consists of an interconnected group ofÂ artificial neuronsÂ and processes information using aÂ connectionistÂ approach toÂ computation. In most cases an ANN is anÂ adaptive systemÂ that changes its structure based on external or internal information that flows through the network during the learning phase. Modern neural networks areÂ non-linearÂ statisticalÂ data modelingÂ tools. They are usually used to model complex relationships between inputs and outputs or toÂ find patternsÂ in data.
What has attracted the most interest in neural networks is the possibility ofÂ learning. There are three major learning paradigms, each corresponding to a particular abstract learning task. These areÂ supervised learning,Â unsupervised learningÂ andÂ reinforcement learning. Usually any given type of network architecture can be employed in any of those tasks.
InÂ machine learning,Â unsupervised learningÂ is a class of problems in which one seeks to determine how the data are organized. It is distinguished fromÂ supervised learningÂ (andÂ reinforcement learning) in that the learner is given only unlabeled examples. Unsupervised learning is closely related to the problem ofÂ density estimationÂ inÂ statistics. However unsupervised learning also encompasses many other techniques that seek to summarize and explain key features of the data.
One form of unsupervised learning isÂ clustering. Another example is blind source separation based onÂ Independent Component AnalysisÂ (ICA). AmongÂ neural networkÂ models, theÂ Self-organizing mapÂ (SOM) andÂ Adaptive resonance theoryÂ (ART) are commonly used unsupervised learning algorithms. The SOM is a topographic organization in which nearby locations in the map represent inputs with similar properties. The ART model allows the number of clusters to vary with problem size and lets the user control the degree of similarity between members of the same clusters by means of a user-defined constant called theÂ vigilance parameter. ART networks are also used for many pattern recognition tasks, such asÂ automatic target recognitionÂ and seismic signal processing. The first version of ART was "ART1", developed by Carpenter and Grossberg(1988).
Self-organizing Map (SOM)
AÂ self-organizing map (SOM)Â orÂ self-organizing feature map (SOFM)Â is a type ofÂ artificial neural networkÂ that is trained usingÂ unsupervised learningÂ to produce a low-dimensional (typically two-dimensional), discretized representation of the input space of the training samples, called aÂ map. Self-organizing maps are different from other artificial neural networks in the sense that they use a neighborhood function to preserve theÂ topologicalÂ properties of the input space. SOM useful forÂ visualizingÂ low-dimensional views of high-dimensional data, akin toÂ multidimensional scaling. The model was first described as an artificial neural network by theÂ FinnishÂ professorÂ Teuvo Kohonen, and is sometimes called aÂ Kohonen map. (Â Kohonen & Honkela, 2007)
Operate in two modes: training and mapping. Training builds the map using input examples. It is a competitive process, also calledÂ vector quantization. Mapping automatically classifies a new input vector.
Consist of components called nodes or neurons. Associated with each node is a weight vector of the same dimension as the input data vectors and a position in the map space. The usual arrangement of nodes is a regular spacing in a hexagonal or rectangular grid. The self-organizing map describes a mapping from a higher dimensional input space to a lower dimensional map space. The procedure for placing a vector from data space onto the map is to find the node with the closest weight vector to the vector taken from data space and to assign the map coordinates of this node to our vector.