Abstract | ||
---|---|---|
In this study, we propose a novel method of lung lesion detection in FDG-PET/CT volumes without labeling lesions. In our method,the probability distribution over normal standardized uptake values (SUVs) is estimated from the features extracted from the corresponding volume of interest (VOI) in the CT volume, which include gradient-based and texture-based features. To estimate the distribution, we use Gaussian process regression with an automatic relevance determination kernel, which provides the relevance of feature values to estimation. Our SUV is judged as normal or abnormal by comparison with the estimated SUV distribution. According to the validation using 28 FDG-PET/CT volumes with 34 lung lesions, the sensitivity of proposed method at 5.0 false positives per case was 81.9%. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1117/12.2255588 | Proceedings of SPIE |
Keywords | Field | DocType |
FDG-PET/CT,computer-assisted detection,distribution estimation,Gaussian process regression,anomaly detection | Nuclear medicine,Kriging,Computer vision,Anomaly detection,Volume of interest,Lung lesion,PET-CT,Lesion,Probability distribution,Artificial intelligence,Physics,False positive paradox | Conference |
Volume | ISSN | Citations |
10134 | 0277-786X | 0 |
PageRank | References | Authors |
0.34 | 4 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ryosuke Kamesawa | 1 | 0 | 0.34 |
Issei Sato | 2 | 331 | 41.59 |
Shouhei Hanaoka | 3 | 26 | 7.56 |
Nomura, Y. | 4 | 31 | 9.51 |
Mitsutaka Nemoto | 5 | 46 | 8.42 |
Naoto Hayashi | 6 | 20 | 6.38 |
Masashi Sugiyama | 7 | 3353 | 264.24 |