Keywords:
CAD, Physics, CT, Computer applications, Computer Applications-Detection, diagnosis, Tissue characterisation
Authors:
G. Ficarra1, E. Barabino2, M. Verda3, S. Casella4, S. Caprioli5, G. Cittadini1; 1Genoa/IT, 2Genova/IT, 3Imperia/IT, 4Savona/IT, 5Arenzano/IT
DOI:
10.26044/ecr2019/C-2626
Conclusion
Radiomics is an expanding field that aims at identify biomarkers and to unveil information of tumor heterogeneity hidden in medical images,
which could improve diagnosis15,16,
predict clinical outcomes like response to treatment 17,18 and even complications19,20.
Radiomics involves several steps (image acquisition,
segmentation,
feature extraction,
and feature selection) with issues that have to be faced; different image acquisition protocols,
segmentation techniques and processing procedures have all been used in previous studies.
This could yield divergent results because of a lack of reproducibility not only between centers,
but also within the same center21,22,23.
CTTA derived features and in particular parameters derived from larger co-occurence matrices have proven to be reliable,
both in intra-CT and in inter-CT evaluation. Among acquisition parameters those that mainly influenced variability resulted to be milliamperage and section thickness.
The lowest variability can be achieved by keeping section thickness between 2.5mm and 5mm,
milliamperage between 150mA and 350mA and voltage around 140kV.