Atomatic Detection and Diagnosis of Severe Viral Pneumonia CT Images Based on LDA-SVM
In: IEEE Sensors Journal, Jg. 20 (2020-10-15), S. 11927-11934
Online
unknown
Zugriff:
The identification of pneumonia types mainly depends on the experience of doctors, but some CT images of pneumonia are very similar, even experienced doctors are prone to misdiagnosis. In order to solve the problems of inefficiency, coarse granularity and poor accuracy under the background of large data, LDA-SVM (Linear Discriminate Analysis - support vector machine) classification algorithm in machine learning field is introduced. LDA is used to extract features from images, and SVM classifier is used to classify the sub-datasets with strong fusion features. On this basis, fusion index and intermediary centrality index are selected to measure the fusion degree of patent sub-centralization technology and identify the key technologies in the fusion process, Because of the fusion of several algorithms, the algorithm needs many iteration training, and the computation time is too long. And simulation results show that our proposed method has significant improvement on identification accuracy rate.
Titel: |
Atomatic Detection and Diagnosis of Severe Viral Pneumonia CT Images Based on LDA-SVM
|
---|---|
Autor/in / Beteiligte Person: | Cao, Congcong ; Ling, Gengfei |
Link: | |
Zeitschrift: | IEEE Sensors Journal, Jg. 20 (2020-10-15), S. 11927-11934 |
Veröffentlichung: | Institute of Electrical and Electronics Engineers (IEEE), 2020 |
Medientyp: | unknown |
ISSN: | 2379-9153 (print) ; 1530-437X (print) |
DOI: | 10.1109/jsen.2019.2959617 |
Schlagwort: |
|
Sonstiges: |
|