logo SBA

ETD

Digital archive of theses discussed at the University of Pisa

 

Thesis etd-01092019-111908


Thesis type
Tesi di laurea magistrale LM6
URN
etd-01092019-111908
Thesis title
Radiomic CT analysis in patients with non-small cell lung cancer: correlation with liquid biopsy
Department
RICERCA TRASLAZIONALE E DELLE NUOVE TECNOLOGIE IN MEDICINA E CHIRURGIA
Course of study
MEDICINA E CHIRURGIA
Supervisors
.
relatore Prof. Neri, Emanuele
Keywords
  • biopsia liquida
  • carcinoma polmonare non a piccole cellule
  • Ct
  • immunoterapia.
  • liquid biopsy
  • NSCLC
  • radiomica
  • radiomics
  • tc
Graduation session start date
29/01/2019
Availability
Withheld
Release date
29/01/2089
Abstract (Inglese)
Abstract (Italiano)
Radiomics is a process of extraction and analysis of quantitative features from diagnostic images. Liquid Biopsy is a test done on sample of blood to look for cells or pieces of tumourigenic DNA circulating in blood. Radiomic and liquid biopsy have a great potential in onclogy. They are minimally invasive,easy to perform and can be repeated in patient follow-up visits. In our study, we have analysed CT of 10 patients with NSCLC(3-4 stage) in immunotherapeutic treatmnet. Liquid biopsy was performed at basal and at first clinical reassessment. Analysed tc refer to the same time, they have been uploaded to QUIBIM software, and then ROI have been segmented and texture analysis has been started to extract radiomic features. Afterwards, statisctical analysis showed Cluster Prominance as the most predictive feature; the lower its values are, the more they seem to be predictive of non-response to treatment. From correlation analysis it was found that there is a positive correlation between Information measure of correlation 2 and TNF-A,and between Volume Max and INF-G. Instead between D2D value and INF-G and TNF-A there is negative correlation. Negative correlation between Information measure of correlation 1 and IFN-G and TNF-A. Our results showed that both radiomics and liquid biopsy could represent efficient diagnostic and predective methods.
File