Study on Text Mining Algorithm for Ultrasound Examination of Chronic Liver Diseases Based on Spectral Clustering
Bingguo Chang, Xiaofei Chen
Available Online May 2018.
- https://doi.org/10.2991/amcce-18.2018.78How to use a DOI?
- component; Text mining; Ultrasonic examination report; Spectral clustering learning algorithm; Chronic liver disease
- Ultrasonography is an important examination for the diagnosis of chronic liver disease. The doctor gives the liver indicators and suggests the patient’s condition according to the description of ultrasound report. With the rapid increase in the amount of data of ultrasound report, the workload of professional physician to manually distinguish ultrasound results significantly increases. In this paper, we use the spectral clustering method to cluster analysis of the description of the ultrasound report, and automatically generate the ultrasonic diagnostic diagnosis by machine learning. 110 groups ultrasound examination report of chronic liver disease were selected as test samples in this experiment, and the results were validated by spectral clustering and compared with k-means clustering algorithm. The results show that the accuracy of spectral clustering is 92.73%, which is higher than that of k-means clustering algorithm, which provides a powerful ultrasound-assisted diagnosis for patients with chronic liver disease.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Bingguo Chang AU - Xiaofei Chen PY - 2018/05 DA - 2018/05 TI - Study on Text Mining Algorithm for Ultrasound Examination of Chronic Liver Diseases Based on Spectral Clustering BT - 2018 3rd International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2018) PB - Atlantis Press SP - 454 EP - 459 SN - 2352-5401 UR - https://doi.org/10.2991/amcce-18.2018.78 DO - https://doi.org/10.2991/amcce-18.2018.78 ID - Chang2018/05 ER -