Page 260 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications
P. 260
Chapter 9 Applications of deep learning in biomedical engineering 251
The significant application of CNN in the biomedical field
includes various problems such as follows:
• Diagnosing breast cancer
• Tissue extraction
• Detection and classification of tumors
• Assessment of bone age in X-ray images
• EEG classification
• Detection of arrhythmia using electrocardiogram (ECG)
signals [10]
Biomedical scientists are interested in advancement of disease
detection using time series data. For instance, the doctors can
analyze the heart functions by monitoring ECG signals. Similarly,
brain's functions and its related disorders such as seizure can be
diagnosed using EEG signals.
10. Recurrent neural network
An RNN is an ANN, which is used in the applications such as
speech recognition and NLP. RNNs are designed to identify data's
sequential characteristics and use patterns to predict the next
layout.
RNNs are utilized in DL and in the advancement of models
that reproduce the movement of neurons in the human brain.
They are particularly powerful in use cases in which context is
critical to predict an outcome and are distinct from other types
of ANN. As they use feedback loops to process a sequence of
data that informs the final output, which can also be a sequence
of data, these feedback loops permit data to persist; the impact is
frequently depicted as memory.
ANNs are made with interconnected components that are
loosely intended to work like the human brain. They are made
out of layers of artificial neurons that have the ability to process
input and forward output to other nodes in the network. The
nodes are associated by edges or weights that impact a signal's
strength and the network's yield [11]. The architecture of RNN
is shown in Fig. 9.5.
11. Applications of recurrent neural network
in biomedicine
RNN is a type of neural network that can classify sequential data
by capturing dynamic information. It is having the ability of storing
past information. The foremost benefit of RNN is the capability of
