A Critical Analysis of the Robustness of Radiomics to Variations in Segmentation Methods in 18 F-PSMA-1007 PET Images of Patients Affected by Prostate Cancer.
Giovanni PasiniGiorgio Ivan RussoCristina MantarroFabiano BiniSelene RichiusaLucrezia MorganteAlbert ComelliGiorgio Ivan RussoMaria Gabriella SabiniSebastiano CosentinoFranco MarinozziMassimo IppolitoAlessandro StefanoPublished in: Diagnostics (Basel, Switzerland) (2023)
Our study showed that segmentation methods influence radiomics features and that Shape features were the least reproducible (average ICC: 0.27), while GLCM features the most reproducible. Moreover, feature reproducibility changed depending on segmentation type, resulting in 51.18% of LoG features exhibiting excellent reproducibility (range average ICC: 0.68-0.87) and 47.85% of wavelet features exhibiting poor reproducibility that varied between wavelet sub-bands (range average ICC: 0.34-0.80) and resulted in the LLL band showing the highest average ICC (0.80). Finally, model performance showed that region growing led to the highest accuracy (74.49%), improved sensitivity (84.38%) and AUC (79.20%) in contrast with manual segmentation.
Keyphrases
- convolutional neural network
- deep learning
- prostate cancer
- machine learning
- end stage renal disease
- magnetic resonance
- squamous cell carcinoma
- ejection fraction
- computed tomography
- pet imaging
- magnetic resonance imaging
- peritoneal dialysis
- radical prostatectomy
- positron emission tomography
- patient reported outcomes
- patient reported