Biomedical subjects

Daoying Geng

Publications and source records attributed to Daoying Geng.

5 recordsLinked to original sources

PubMed2006

[Automatic diagnosis of malignant degree of brain glioma based on Bayesian network].

Bayesian network connects graph theory with statistics, being an important research direction in data mining. Compared with other approaches used for data mining, Bayesian network can combine prior knowledge with observed data. Besides that, it can handle incomplete data sets. This paper applies Bayesian network to predict the malignant degree of brain glioma. Totally 280 cases are collected, and some of them contain missing values. Preprocessing is taken to make them applicable to the algorithms. Unlike MLP network, both Bayesian network and decision tree use attribute-value pairs to represent diagnostic knowledge derived from treated cases. These could improve both the understandability and applicability of their results. Results of all these algorithms can achieve accuracy rate over 80%, which satisfies the requirement of neuroradiologists.

Bayes Theorem↗

PubMed2006

[Rule induction algorithm for brain glioma using support vector machine].

A new proposed data mining technique, support vector machine (SVM), is used to predict the degree of malignancy in brain glioma. Based on statistical learning theory, SVM realizes the principle of data dependent structure risk minimization, so it can depress the overfitting with better generalization performance, since the prediction in medical diagnosis often deals with a small sample. SVM based rule induction algorithm is implemented in comparison with other data mining techniques such as artificial neural networks, rule induction algorithm and fuzzy rule extraction algorithm based on fuzzy max-min neural networks (FRE-FMMNN) proposed recently. Computation results by 10 fold cross validation method show that SVM can get higher prediction accuracy than artificial neural networks and FRE-FMMNN, which implies SVM can get higher accuracy and more reliability. On the whole data sets, SVM gets one rule with the classification accuracy of 89.29%, while FRE-FMMNN gets two rules of 84. 64%, in which the rule got by SVM is of quantity relation and contains more information than the two rules by FRE-FMMNN. All the above show SVM is a potential algorithm for the medical diagnosis such as the prediction of the degree of malignancy in brain glioma.

Algorithms↗

PubMed2002

[Computer aided intracranial aneurysm embolization with GDC].

OBJECTIVE: To establish an expert system that automatically generates optimal GDC selection program for the embolization of intracranial aneurysm. METHODS: Twenty highly cost-effective cases of intracranial aneurysm embolized with GDC dense packing were collected. Each of them contains information including aneurysm's volume measured by three-dimension digital subtraction angiography (3D DSA), aneurysm's location, maximum transverse diameter, maximum length diameter, and GDC selection program. An expert system made up of a case base, a mathematical model simulating experts' experience (established with the help of data mining techniques combining multi-layer perceptron network with polyhedrons in high dimensional space), and data envelopment analysis (DEA), was implemented. RESULTS: When the user inputted four required parameters (volume, location, maximum transverse diameter, and maximum length diameter) into the expert system and clicked the "program design" button, candidate GDC selection program(s) would be presented in the result box. CONCLUSION: Case base, data mining techniques, and DEA can be used to establish the expert system that automatically generates optimal GDC selection program for the embolization of intracranial aneurysm. Its clinical value needs to be further evaluated.

Adult↗

PubMed2002

[Three dimensional digital subtraction angiography in volume embolization ratio measurement of densely packing experimental aneurysms].

OBJECTIVE: To calculate the volume embolization ratio of densely packing experimental aneurysms by three dimensional digital subtraction angiography (3D-DSA). METHODS: Six experimental crotch aneurysms were created microsurgically in the common carotid artery of white rabbits. Two weeks later, each aneurysm's volume was measured with 3D-DSA surface shaded display(SSD) and the correction of lacteprene balloon calibration method. In the same time, the aneurysms were densely packed with electric detachable coils. The volume of coils that were used in each aneurysm was calculated separately. The ratio of coils volume and aneurysm volume was the volume embolization ratio (VER). RESULTS: The aneurysms volumes measured by 3D DSA SSD ranged from 0.037 to 0.087 ml. The VER ranged from 23.5% to 32.5% (average 27.4%). CONCLUSION: The minimum VER of densely packing experimental crotch aneurysms with electronic detachable coils was 23.5%.

Angiography, Digital Subtraction↗

PubMed2002

[Data mining in diagnostic knowledge acquisition from patients with brain glioma].

In order to correctly predict the malignant degree of brain glioma, three data mining algorithms: multi-layer perceptron network(MLP), decision tree, and rule induction are adopted to acquire diagnostic knowledge from patients with brain glioma cases. Totally 280 cases are collected, and some of them contain missing values. Preprocessing is taken to make them applicable to all three algorithms. Performance comparisons are carried out with a 10-fold cross validation test. Although the result of MLP is hard to be understood and cannot be applied directly, its reliability and accuracy are the highest when only a few hidden nodes are involved. Unlike MLP, both decision tree and rule induction use attribute-value pairs to represent diagnostic knowledge derived from treated cases. These could improve both the understandability and applicability of their results. When compared with rule induction, the inherent restriction in structure makes decision tree more efficient in decision-making but meanwhile hurts its simplicity, accuracy, and reliability. For testing samples, results of all these algorithms can achieve accuracy rate over 80%, which satisfies the basic requirement of neuroradiologists. If diagnostic accuracy rate is the main factor to be considered, MLP with only a few hidden nodes is the best. If the result is expected to be further checked or evaluated, rule induction will be the best algorithm. This work proves that data mining techniques can be used to obtain valid diagnostic knowledge from brain glioma cases and make computer aided diagnosis system in this field feasible.

Algorithms↗