Page 70 - 22-0424

The International Journal of the Royal Society of Thailand
Volume XI - 2019

often needs large storage sever requiring high velocity of data transfer. Data may
be presented with variety in forms such as structured or non-structured data.
Some data may change consistency overtime (variability) and some may need to

be verified before use (veracity) (Archenaa and Anita, 2015). The origins of big
data were summarized in Figure 5.

Figure 5 The source of big data. Tissue samples from patients or volunteers were stored in
biobank and converted into interested genetic code by various genetic methods. Another source
was from electronic medical records containing text data, picture data and audio data. Specific
real-world data, such as insurance document, can also be used as a source of big data.

Artificial intelligence (AI) is the form of computations that possible to
perceive, reason, and act (Niel and Bastard, 2019). Algorithms use in AI are random
forest, support vector machine and artificial neural networks (ANN). ANN consist
of 3 layers including input layer, hidden layer and output layer. Each artificial

neuron is interconnected to all other neurons in the next layer. They received
multiple weighted inputs to create outputs. AI can perform various functions
after trained by using adequately large data sets. These data sets are made up of
training batches and testing batches, are used to verify performance of trained AI.

Deep learning is defined as a form of machine learning algorithm based
on neural networks containing multiple layers of nonlinear processing units for
feature extraction and transformation. It is often necessary in processing of complex
data such as sounds and images. With deep learning, AI can perform more complex
tasks with lower computational resource requirement. These can be achieved
by adding more hidden layers of neuron into ANN. Examples of deep learning

64 Precision Medicine in Acute kidney injury

11/7/2565 BE 13:28
_22-0424(055-076)7.indd 64 11/7/2565 BE 13:28
_22-0424(055-076)7.indd 64