Cancer diagnostic method and system
View Patent ↗A method and system for classifying tissue suspected of being abnormal as being malignant or benign. The method includes generating a set of selection features, performing statistical applications to generate additional selection features, generating a feature vector for the abnormal tissue, feeding the feature vector into a neural network, and obtaining a result from the neural network as to whether the abnormal tissue is malignant or benign. The method and system may be used for determining the presence of cancers such as breast cancer.
1. A method for classifying tissue suspected of being abnormal in the body of a subject as being malignant or benign, comprising:
generating a set of selection features relating to data obtained about a subject which has the abnormal tissue;
performing at least three classification models using the set of selection features to generate further selection features;
using the further selection features in combination with the set of selection features to generate a feature vector for the abnormal tissue;
selecting features from the feature vector utilizing a first genetic algorithm;
generating an input vector with the features selected from the feature vector;
feeding the input vector into a neural network;
training the neural network with a second genetic algorithm; and
obtaining a result from the neural network as to whether the abnormal tissue is malignant or benign.
2. The method of claim 1 , wherein the data obtained about the subject is human interpreted data.
3. The method of claim 2 , wherein the human interpreted data includes mass shape of the abnormal tissue.
4. The method of claim 1 , wherein the data obtained about the subject is computer generated data relating to image processing features.
5. The method of claim 1 , further comprising normalising the selection features prior to feeding the neural network.
6. The method of claim 1 , wherein one of the classification models preformed is multivariate adaptive regression splines.
7. The method of claim 1 , wherein one of the classification models preformed is support vector machine.
8. The method of claim 1 , wherein the input vector is generated by multiplying a selection vector and a feature array.
9. The method of claim 8 , wherein the selection vector includes a set of binary numbers, and the feature array includes selection features.
10. The method of claim 8 , wherein the feature array includes a mixture of computer generated data relating to image processing features and human interpreted data.
11. The method of claim 1 , wherein the abnormal tissue is located in the breast of the subject.
12. A system for classifying tissue suspected of being abnormal in the body of a subject as being malignant or benign, comprising:
at least one input for receiving a set of selection features relating to information about a subject which has the abnormal tissue;
at least one processor coupled to said input, said processor configured to generate further selection features from at least three statistical calculations performed on said set of selection features; and
a neural network trained by a genetic algorithm, said neural network being configured to determine whether the abnormal tissue is malignant or benign utilizing said set of selection features and said further selection features.
13. The system of claim 12 , wherein the processor is programmed with a genetic algorithm to optimize the set of selection features.
14. The system of claim 12 , wherein one of the statistical calculations is multivariate adaptive regression splines.
15. The system of claim 12 , wherein one of the statistical calculations is support vector machine.
16. The system of claim 12 , wherein said selection features include computer generated data relating to image processing features.
17. The system of claim 12 , wherein said selection features include human interpreted data.