José Félix Mañes Ferrer

Segmentation method for breast tumor diagnosis based on Artificial Neural Network algorithm applied to dynamic 18F-FDG PET images

To establish an accurate and automatic image segmentation method for extracting the tumor from the healthy tissue using the dedicated 18F-FDG Mammography with Molecular Imaging (MAMMI) Positron Emission Tomography (PET) dynamic images of breast in vivo, the paper presents a novel method using Artificial Neural Network (ANN) combined with time activity curves (TAC) of each voxel as input vector. TACs voxel by voxel were obtained from PET images with average filtering and least-squares fitting algorithm to improve the signal noise ratio (SNR). The data was then normalized and constructed as input feature vectors of the ANN network to train or segment the tumor regions. The initial pilot validation with 2 patients data of the proposed method using FDG has shown promising results.

SU‐E‐I‐83: Detectability Limits of a New Positron Emission Mammography in Relation to Tumour‐Size, Tumour‐To‐Background Ratio and Activity Concentration

PURPOSE Positron emission mammography (PEM) improves spatial resolution and sensitivity, making it suitable for early breast tumours detection. The aim of this study is to evaluate the limits of a dedicated breast PET in terms of tumour-size, tumour-to-background activity concentration ratio (TBR) and activity concentration. METHODS A dedicated PEM is evaluated. To characterize the device, we use a phantom of 15 cm of diameter containing 6 inserts of inner diameters 18, 15, 11, 8, 5 and 3 mm. To evaluate the detectability limits images are acquired by varying the TBR from 10 to 2 and without background activity concentration. In all the studies the activity concentration for the 6 inserts is the same (3.7 kBq/ml). To asses the activity concentration limit, the SBR is maintained fixed and acquisitions at different times are performed. RESULTS When there is not background, all the inserts are visible in the PEM. Increasing the background decreases the detectability. With a SBR of 10, the smaller insert is still visible. For TBR between 10, inserts >= 5 mm can be seen With a TBR of 2, only inserts with a diameter higher than 5 mm are visible. When the TBR is fixed, decreasing the activity concentration, decreases the capability of detectability. CONCLUSIONS The results suggest that PEM can be used for diagnosis of small lesions when TBR is higher than 2. Further clinical studies need to be carried out in order to validate these results.