CLASSIFICATION OF BREAST CANCER MALIGNANCY USINGCYTOLOGICAL IMAGES OF FINE NEEDLE

ASPIRATION BIOPSIES

• Breast cancer is the most often diagnosed cancer among women aged 40 to 60.

• According to the World Health Organization there are 7.6 million deaths worldwide due to cancer each year, out of which 502,000 are caused by breast cancer alone.

• Cancers in their early stages are vulnerable to treatment while cancers in their most advanced stages are usually almost impossible to treat.

• The most common diagnostic tools are mammography and a fine needle aspiration biopsy (FNA).

• Mammography, which is a non-invasive method, is most often used for screening purposes rather than for precise diagnosis.

• It allows a physician to find possible locations of

microcalcifications and other indicators in breast tissue.

• When a suspicious region is found, the patient is sent to

a pathologist for a more precise diagnosis. This is when

the FNA is taken.

• A fine needle aspiration biopsy is an invasive method to extract a small sample of the questionable breast tissue that allows the pathologist to describe the type of the cancer in detail.

• Using this method pathologists can very adequately describe not only the type of the cancer but also its genealogy and malignancy.

• The stage of cancer depends on the malignancy factor that is assigned during an FNA examination. The determination of malignancy is essential when predicting the progression of cancer.

Block diagram of the cell classification system

Segmentation

Feature Extraction

Classification

Cell images

Neuro fuzzy system

• In the field of artificial intelligence, neuro-fuzzy refers to combinations of artificial neural networks and fuzzy logic.

• Neuro-fuzzy was proposed by J. S. R. Jang

• Neuro-fuzzy hybridization results in a hybrid intelligent systemthat synergizes these two techniques by combining the human-like reasoning style of fuzzy systems with the learning and connectionist structure of neural networks.

• Deriving fuzzy rules from trained RBF networks.

• Fuzzy logic based tuning of neural network training parameters.

• Fuzzy logic criteria for increasing a network size.

• Realising fuzzy membership function through clustering algorithms in unsupervised learning in SOMs and neural networks.

• Representing fuzzification, fuzzy inference and defuzzification through multi-layers feed-forward connectionist networks.

Neuro Fuzzy Reasoner for Student Modeling

• The steps to take in order to apply the NFR model thestudent modeling problem in this case are as follows:

• 1. Defining input and output values;• 2. Defining fuzzy sets for input values;• 3. Defining fuzzy rules;• 4. Creating and training the neural network

• 1. Input and output valuesInput values:

Test score [0..100]The time needed to complete the test [0..120]Output values:Classes of students: {Bad, Good, Very good, Excellent}

2. Fuzzy sets

• The input space is partitioned by the following fuzzy sets:

• - Test score: Bad, Low, Mid, High

• - The time needed to complete the test,

interpreted as speed: Slow, Moderate,

Fast

• 3. Fuzzy rules

The rules for student classification are taken from the

human teacher. Twelve such rules are shown in Table 1.

The values of linguistic variables in the premises are

interpreted as the previously defined fuzzy sets, and the

rules bellow are interpreted like:

IF ( (TEST_SCORE IS HIGH) AND

(STUDENT_SPEED IS FAST) )

THEN STUDENT_CLASS IS EXCELLENT

4.Creating and training the neural network

• When the fuzzy model is defined, the construction of the corresponding NFR model is straightforward.

• The NFR model that corresponds to the previously defined fuzzy model is shown in fig. 4.

• The network is constructed using the following principles:1. The number of cells in the input layer L1 is equal to the number of inputs;2. The number of cells in the fuzzyfication layer L2 is equal to the number of fuzzy sets;

3. The number of cells in the premise layer is equal to the number of rules;4. The number of cells in the output layer is equal to the the number of classification classes;

5. The connection pattern is the same for all NFR models and it is shown in fig. 4.