The aim of this study is to present a fully automatic deep learning algorithm to segment liver Colorectal cancer metastases (lmCRC) on CT images, based on a U-Net structure, comparing nets with and without the transfer learning approach. This is a bi-centric study, enrolling patients who underwent CT exam before (baseline) and after first-line therapy (TP1). Patients were divided into training (using a portion of baseline sequences from both centers) to train the DL model, and two validation sets: one with baseline (valB), and one with TP1 (valTP1) sequences. The reference standard for the automatic segmentations was defined by the manual segmentations performed by an experienced radiologist on the portal phase of the baseline and TP1 CT exam. The best performing model obtained Dice Similarity Coefficient (DSC) of 0.68±0.24, Precision (Pr) of 0.74±0.27, Recall (Re) of 0.73±0.26, Detection Rate (DR) of 93% on the valB, and DSC of 0.61±0.28, Pr of 0.68±0.31, Re of 0.65±0.29 and DR of 88% on the valTP1. These encouraging results, if confirmed on larger dataset, might provide a reliable and robust tool that can be used as first step of future radiomics analyses aimed at predicting response to therapy, improving the management of lmCRC patients.

A Deep Learning model to segment liver metastases on CT images acquired at different time-points during chemotherapy / Defeudis, Arianna; Panic, Jovana; Guzzinati, Walter; Pusceddu, Laura; Vassallo, Lorenzo; Regge, Daniele; Giannini, Valentina. - ELETTRONICO. - (2022), pp. 1-5. ((Intervento presentato al convegno 17th IEEE International Symposium on Medical Measurements and Applications (IEEE MeMeA 2022) tenutosi a Giardini Naxos - Taormina, Messina (Italy) nel 22 June - 24 June 2022 [10.1109/MeMeA54994.2022.9856589].

A Deep Learning model to segment liver metastases on CT images acquired at different time-points during chemotherapy

Jovana Panic;Valentina Giannini
2022

Abstract

The aim of this study is to present a fully automatic deep learning algorithm to segment liver Colorectal cancer metastases (lmCRC) on CT images, based on a U-Net structure, comparing nets with and without the transfer learning approach. This is a bi-centric study, enrolling patients who underwent CT exam before (baseline) and after first-line therapy (TP1). Patients were divided into training (using a portion of baseline sequences from both centers) to train the DL model, and two validation sets: one with baseline (valB), and one with TP1 (valTP1) sequences. The reference standard for the automatic segmentations was defined by the manual segmentations performed by an experienced radiologist on the portal phase of the baseline and TP1 CT exam. The best performing model obtained Dice Similarity Coefficient (DSC) of 0.68±0.24, Precision (Pr) of 0.74±0.27, Recall (Re) of 0.73±0.26, Detection Rate (DR) of 93% on the valB, and DSC of 0.61±0.28, Pr of 0.68±0.31, Re of 0.65±0.29 and DR of 88% on the valTP1. These encouraging results, if confirmed on larger dataset, might provide a reliable and robust tool that can be used as first step of future radiomics analyses aimed at predicting response to therapy, improving the management of lmCRC patients.
File in questo prodotto:
File Dimensione Formato  
A_Deep_Learning_model_to_segment_liver_metastases_on_CT_images_acquired_at_different_time-points_during_chemotherapy.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 649.94 kB
Formato Adobe PDF
649.94 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
ManuscriptFinal.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 5.43 MB
Formato Adobe PDF
5.43 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2971148