ESTRO 2024 - Abstract Book

S4966

Physics - Radiomics, functional and biological imaging and outcome prediction

ESTRO 2024

[1] Mahvi DA, Liu R, Grinstaff MW, Colson YL, Raut CP. Local Cancer Recurrence: The Realities, Challenges, and Opportunities for New Therapies. CA Cancer J Clin. 2018 Nov; 68 (6): 488-505.

[2] Ling CC, Humm J, Larson S, Amols H, Fuks Z, Leibel S, Koutcher JA. Towards multidimensional radiotherapy (MD CRT): biological imaging and biological conformality. Int J Radiat Oncol Biol Phys. 2000 Jun 1; 47 (3): 551-60.

[3] Thorwarth D. Biologically adapted radiation therapy. Z Med Phys. 2018 Aug; 28 (3): 177-183.

[4] Gurney-Champion OJ, Mahmood F, van Schie M, Julian R, George B, Philippens MEP, van der Heide U, Thowarth D, Redalen KR. Quantitative imaging for radiotherapy purposes. Radiother Oncol. 2020 May; 146: 66-75. [5] Reynaud, O. Time-dependent diffusion MRI in cancer: tissue modeling and applications. Frontiers in Physics. 2017, 5:58.

[5] Reynaud, O. Time-dependent diffusion MRI in cancer: tissue modeling and applications. Frontiers in Physics. 2017, 5:58.

[6] Topgaard D. Advanced diffusion encoding methods in MRI. vol 24. Royal Society of Chemistry. 2020.

323

Digital Poster

A comparison of 3D CNN architectures for monitoring ovarian cancer patients using PET/CT scans

Mohammad Hossein Sadeghi 1 , Sedigheh Sina 1,2 , Mehrosadat Alavi 3 , Zahra Nasiri Feshani 4 , Amir Hossein Farshchitabrizi 5,2 1 Shiraz University, School of Mechanical Engineering, Shiraz, Iran, Islamic Republic of. 2 Shiraz University, Radiation Research Center, Shiraz, Iran, Islamic Republic of. 3 Shiraz University of Medical Sciences, Department of Nuclear Medicine, Shiraz, Iran, Islamic Republic of. 4 Kowsar Hospital, Nuclear Medicine Department, PET/CT Center, Shiraz, Iran, Islamic Republic of. 5 Shiraz University of Medical Sciences, Namazi Hospital, Shiraz, Iran, Islamic Republic of

Purpose/Objective:

Ovarian Cancer (OC) is a cause of cancer-related deaths, in women. It is often detected at stages (Stage III and IV) which makes staging crucial for planning effective treatments. This study aims to compare the performance of 3D Convolutional Neural Network (CNN) architectures, such, as VGG 16, GoogLeNet, UNet, AlexNet, DenseNet, and ResNet34. We also evaluate our state-of-the-art deep learning model to classify OC stages accurately using PET/CT scans.

Material/Methods:

The dataset comprises PET-CT scans from 37 OC patients, resulting in a total of 1224 images. Data augmentation techniques are employed to augment the dataset while ensuring diversity and reducing the risk of overfitting. The CT scans are primarily used for attenuation correction during image reconstruction. PET scans are the primary modality