ETD system

Electronic theses and dissertations repository

 

Tesi etd-07192007-165722


Thesis type
Tesi di dottorato di ricerca
Author
Antonelli, Michela
email address
michela.antonelli@iet.unipi.it, miki@nonsolostudio.it
URN
etd-07192007-165722
Title
Computerized detection and diagnosis of lung lesions using CT
Settore scientifico disciplinare
ING-INF/05
Corso di studi
INGEGNERIA DELL'INFORMAZIONE
Commissione
Relatore Prof. Marcelloni, Francesco
Relatore Prof.ssa Lazzerini, Beatrice
Parole chiave
  • Image processing
  • Computer Aided
Data inizio appello
25/05/2007;
Consultabilità
parziale
Data di rilascio
25/05/2047
Riassunto analitico
The CAD system described in this thesis is able to automatically detect nodules in lung CT images and to <br>automatically perform the diagnosis of such nodules. The system has the following novel characteristics: <br>-it performs the automatic diagnosis without requiring any intervention by the radiologist in any of <br>the phases it consists of; <br>-it does not emulate a single radiologist, but a team of radiologists. <br>In particular, the first characteristic means that the system can replace the radiologist both for detecting <br>the presence of nodules (by distinguishing them from the other anatomical structures), and classifying <br>them into benign and malignant. <br>The second characteristic, on the other hand, means that the CAD system is made of a set of CAD <br>subsystems independent of each other, each able to perform a diagnosis. More precisely, we have <br>considered, within the radiologist’s activity, three distinct and subsequent phases: the search for ROIs, <br>their classification into nodules and non-nodules, and nodule diagnosis. <br>At the start we developed our project by implementing different techniques to perform the three phases <br>of CAD. Each technique, which can work “stand-alone&#34;, has been tested on a significant set of CT scans. <br>Subsequently we built a system that takes into consideration all the set of techniques to obtain a more <br>robust output for each phase. Our system is composed of three modules, one for each phase and each <br>module contains submodules to implement the techniques.
File