TY - GEN
T1 - Computer-based image analysis of liver steatosis with large-scale microscopy imagery and correlation with magnetic resonance imaging lipid analysis
AU - Kong, Jun
AU - Lee, Michael J.
AU - Bagci, Pelin
AU - Sharma, Puneet
AU - Martin, Diego
AU - Adsay, N. Volkan
AU - Saltz, Joel H.
AU - Farris, Alton B.
PY - 2011
Y1 - 2011
N2 - Most pathology analyses and measurements are prevalently carried out by trained reviewers in both clinical and research settings. Therefore, the resulting outputs are inexorably biased by interpreters and degraded with poor reproducibility. In this paper, we propose a computerized image analysis paradigm enabling quantitative characterizations of steatosis areas in microscopy images of pediatric liver biopsies. With the same set of patients, we also acquired the lipid measurements from magnetic resonance imaging data analysis for correlation investigation. Our preliminary results suggest a high correlation between the steatosis areas quantized with microscopy images and the lipid percentages calculated from radiology imaging data. Additionally, we compared the performance of the proposed analysis method with those of three certified pathologists and a popular commercial algorithm. The results suggest the superiority of our method to both human reviewers and the commercial method in terms of the steatosis-lipid correlation strength. This demonstrates that the developed method is promising for generating quantitative and reliable analysis results to better support further liver disease study.
AB - Most pathology analyses and measurements are prevalently carried out by trained reviewers in both clinical and research settings. Therefore, the resulting outputs are inexorably biased by interpreters and degraded with poor reproducibility. In this paper, we propose a computerized image analysis paradigm enabling quantitative characterizations of steatosis areas in microscopy images of pediatric liver biopsies. With the same set of patients, we also acquired the lipid measurements from magnetic resonance imaging data analysis for correlation investigation. Our preliminary results suggest a high correlation between the steatosis areas quantized with microscopy images and the lipid percentages calculated from radiology imaging data. Additionally, we compared the performance of the proposed analysis method with those of three certified pathologists and a popular commercial algorithm. The results suggest the superiority of our method to both human reviewers and the commercial method in terms of the steatosis-lipid correlation strength. This demonstrates that the developed method is promising for generating quantitative and reliable analysis results to better support further liver disease study.
KW - Data correlation
KW - Large-scale microscopy image analysis
KW - Liver steatosis quantification
KW - Parallel computation
KW - Tissue representation
UR - http://www.scopus.com/inward/record.url?scp=84862960386&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84862960386&partnerID=8YFLogxK
U2 - 10.1109/BIBM.2011.37
DO - 10.1109/BIBM.2011.37
M3 - Conference contribution
SN - 9780769545745
T3 - Proceedings - 2011 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2011
SP - 333
EP - 338
BT - Proceedings - 2011 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2011
T2 - 2011 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2011
Y2 - 12 November 2011 through 15 November 2011
ER -