Application of unsupervised learning in clinical oncology practice - exploring anxiety characteristics in chemotherapy-induced nausea and vomiting through principal variables
Yap, Kevin Yi-Lwern, Yak, Xin Ran, Shih, Vivianne, Chui, Wai Keung and Chan, Alexandre. (2010) Application of unsupervised learning in clinical oncology practice - exploring anxiety characteristics in chemotherapy-induced nausea and vomiting through principal variables. Journal of Computing, Vol.2 (Iss.7). pp. 163-171. ISSN 2151-9617Full text not available from this repository.
Official URL: http://www.journalofcomputing.org/volume-2-issue-7...
State anxiety is a risk factor for chemotherapy-induced nausea and vomiting (CINV). However, anxiety is a subjective symptom and is difficult to quantify in clinical practice. Clinicians need appropriate anxiety measures to assess patients’ risks of CINV. We illustrate how unsupervised learning techniques can be applied in oncology practice to explore anxiety characteristics that can be used for the clinical surveillance of cancer patients with CINV. A single-centre, prospective, observational study was done on 49 head-and-neck cancer patients on cisplatin-based chemotherapy and appropriate antiemetic therapy. CINV events and antiemetic use were recorded using a standardized diary, while patients’ anxiety characteristics were evaluated using the Beck Anxiety Inventory. Principal component analysis (PCA) was used as an exploratory technique for statistical analysis. Three anxiety characteristics (indigestion, faintness, numbness) were identified as potential clinical predictors of CINV through the use of principal variables. The potential of PCA as a technique for characterizing anxiety in patients with CINV is indeed appealing. Despite the need to address several key issues before PCA finds widespread applications in cancer supportive care, we hope this study shows the usefulness of PCA as a potential technique in analyzing clinical populations, so as to ultimately improve patients’ quality of life.
|Item Type:||Journal Article|
|Subjects:||H Social Sciences > HA Statistics
R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
|Divisions:||Faculty of Science > WMG (Formerly the Warwick Manufacturing Group)|
|Journal or Publication Title:||Journal of Computing|
|Publisher:||Journal of Computing|
|Official Date:||July 2010|
|Number of Pages:||9|
|Page Range:||pp. 163-171|
|Access rights to Published version:||Open Access|
|Funder:||Faculty of Science, National University of Singapore|
Actions (login required)