Biomedical Engineering Reference
In-Depth Information
that is inspired by the way biological nervous systems, such as the brain,
process information. Simply speaking, it is software that is “trained” by
having its examples of input and the corresponding desired output pre-
sented to it.
Neural networks, with their remarkable ability to derive meaning from
complicated or imprecise data, can be used to extract patterns and de-
tect trends that are too complex to be noticed by either humans or other
computer techniques. A trained neural network can be thought of as an
“expert” in the category of information it has been required to analyze.
Thetypicalstructureofneural network consists of a layer of d (the
dimension of the futures) input units, a layer of output units, and a vari-
able number of hidden layers of units, as shown in Figure 1. Generally
more layers result in higher accuracy, but also are more time-consuming on
computation.
The construction of the NN for this study and test results will be shown
in the next section.
Fig. 1. Typical Structure of Neural Network.
4. Application to Lung Cancer Data
4.1. Data Structure
S INT
A data set records the survival times (
, in months) of the patients
seen at Vanderbilt University School of Medicine Hospital. The data set
also records patients' hospital condition including
PT : patient term, ranges from T 1to T 4
Search WWH ::




Custom Search