DocumentCode
3155657
Title
Detection of Pulmonary Nodules among Monochrome LCDs with Different Resolutions - An Observer Performance Study
Author
Yin, Jiandong ; Guo, Qiyong ; Sha, Xianzheng
Author_Institution
Dept. of Radiol., Shengjing Hosp. of China Med. Univ., Shenyang, China
fYear
2010
fDate
18-20 June 2010
Firstpage
1
Lastpage
4
Abstract
Objective To evaluate the influence of monochrome LCDs with different resolutions on the detection performance of pulmonary nodules. Materials and Methods 93 chest images were selected, including 55 normal images and 38 positive images. Three high-, three mid-, and three low-experienced radiologists interpreted these images using three types of displays. Regarding the presence of nodule, five-point confidence level rating scale was used: definite absence, probable absence, indetermination, probable presence and definite presence. Observer marked their confidence level of every image according to the presence of pulmonary nodule on the given display. Software SPSS 13.0 was used for statistical analysis. Results The areas under ROC curves for high-experienced radiologists were 0.774, 0.784 and 0.816 on 2MP, 3MP and 5MP display, respectively; those for mid-experienced radiologists were 0.754, 0.764 and 0.768, respectively; and those for low-experienced radiologists were 0.695, 0.754 and 0.774, respectively. There were no substantial differences among different types of displays or radiologists of different aptitudes. Conclusion It´ s equivalent of the detection performance of pulmonary nodules among on 2MP, 3MP and 5MP monochrome LCDs when there is no restriction on the use of image post-processing tools.
Keywords
computerised tomography; liquid crystal displays; lung; medical image processing; statistical analysis; tumours; CT image; ROC curves; chest images; confidence level; definite absence; definite presence; five-point confidence level rating scale; image post-processing tools; indetermination; monochrome LCDs; normal images; positive images; probable absence; probable presence; pulmonary nodule detection; software SPSS 13.0; statistical analysis; Biomedical engineering; Biomedical imaging; Cathode ray tubes; Computer displays; Hospitals; Human factors; Liquid crystal displays; Medical diagnostic imaging; Radiology; Statistical analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Bioinformatics and Biomedical Engineering (iCBBE), 2010 4th International Conference on
Conference_Location
Chengdu
ISSN
2151-7614
Print_ISBN
978-1-4244-4712-1
Electronic_ISBN
2151-7614
Type
conf
DOI
10.1109/ICBBE.2010.5518211
Filename
5518211
Link To Document