Diagnostic accuracy of 3D-based AI technology to identify the status of the predicted resection in margins in patients with locally advanced and recurrent rectal cancer.
EAES Academy. Pellino G. 07/05/22; 366543; P286
CLICK HERE TO LOGIN
REGULAR CONTENT
REGULAR CONTENT
Login now to access Regular content available to all registered users.
Abstract
Discussion Forum (0)
Rate & Comment (0)
Background
R0 or tumour-free surgical margins resection represents the strongest prognosticator of survival in locally advanced primary (LAPRC) and locally recurrent rectal cancer (LRRC). Magnetic resonance imaging (MRI) is the technique of choice to assess the infiltration of surrounding structures, currently representing the ideal tool for preoperative planning. In LAPRC, MRI negative predictive value of infiltration is reported to be 94%, however, a 54% positive predictive value has been estimated. In addition, diagnostic accuracy of MRI might not exceed 60% in LRRC.
Aim
The aim of this study was to assess the usefulness of a three-dimensional image processing and reconstruction (3D-IPR) model to achieve R0 resections and to compare the diagnostic accuracy between MRI and 3D-IPR regarding the infiltration of surrounding structures in LAPRC and LRRC.
Materials and Method
This is a prospective study performed at two referral centres for rectal cancer between January 2020 and January 2022. 3D-IPR was applied to MRI of patients with LAPRC or LRRC, before surgery. The MRI findings were compared with those of 3D-IPR, focusing on predicted surgical margins. The standard of reference was definitive pathology of the specimen
Results
Twelve patients were evaluated (7 LAPRC and 5 LRRC). A complete agreement between MRI and 3D-IPR was observed in 16% of cases.
One patient received anterior resection of the rectum, 4 en bloc rectal resection extended to surrounding structures, and 7 pelvic exenteration; 75% of specimens were classified as R0.
The diagnostic accuracy was 33% for MRI and 91% for 3D-IPR.
Conclusions
The 3D-IPR method can be useful to improve diagnostic accuracy of MRI scans in assessing the relationship with surrounding structures in patients with LAPRC and LRRC
R0 or tumour-free surgical margins resection represents the strongest prognosticator of survival in locally advanced primary (LAPRC) and locally recurrent rectal cancer (LRRC). Magnetic resonance imaging (MRI) is the technique of choice to assess the infiltration of surrounding structures, currently representing the ideal tool for preoperative planning. In LAPRC, MRI negative predictive value of infiltration is reported to be 94%, however, a 54% positive predictive value has been estimated. In addition, diagnostic accuracy of MRI might not exceed 60% in LRRC.
Aim
The aim of this study was to assess the usefulness of a three-dimensional image processing and reconstruction (3D-IPR) model to achieve R0 resections and to compare the diagnostic accuracy between MRI and 3D-IPR regarding the infiltration of surrounding structures in LAPRC and LRRC.
Materials and Method
This is a prospective study performed at two referral centres for rectal cancer between January 2020 and January 2022. 3D-IPR was applied to MRI of patients with LAPRC or LRRC, before surgery. The MRI findings were compared with those of 3D-IPR, focusing on predicted surgical margins. The standard of reference was definitive pathology of the specimen
Results
Twelve patients were evaluated (7 LAPRC and 5 LRRC). A complete agreement between MRI and 3D-IPR was observed in 16% of cases.
One patient received anterior resection of the rectum, 4 en bloc rectal resection extended to surrounding structures, and 7 pelvic exenteration; 75% of specimens were classified as R0.
The diagnostic accuracy was 33% for MRI and 91% for 3D-IPR.
Conclusions
The 3D-IPR method can be useful to improve diagnostic accuracy of MRI scans in assessing the relationship with surrounding structures in patients with LAPRC and LRRC
Background
R0 or tumour-free surgical margins resection represents the strongest prognosticator of survival in locally advanced primary (LAPRC) and locally recurrent rectal cancer (LRRC). Magnetic resonance imaging (MRI) is the technique of choice to assess the infiltration of surrounding structures, currently representing the ideal tool for preoperative planning. In LAPRC, MRI negative predictive value of infiltration is reported to be 94%, however, a 54% positive predictive value has been estimated. In addition, diagnostic accuracy of MRI might not exceed 60% in LRRC.
Aim
The aim of this study was to assess the usefulness of a three-dimensional image processing and reconstruction (3D-IPR) model to achieve R0 resections and to compare the diagnostic accuracy between MRI and 3D-IPR regarding the infiltration of surrounding structures in LAPRC and LRRC.
Materials and Method
This is a prospective study performed at two referral centres for rectal cancer between January 2020 and January 2022. 3D-IPR was applied to MRI of patients with LAPRC or LRRC, before surgery. The MRI findings were compared with those of 3D-IPR, focusing on predicted surgical margins. The standard of reference was definitive pathology of the specimen
Results
Twelve patients were evaluated (7 LAPRC and 5 LRRC). A complete agreement between MRI and 3D-IPR was observed in 16% of cases.
One patient received anterior resection of the rectum, 4 en bloc rectal resection extended to surrounding structures, and 7 pelvic exenteration; 75% of specimens were classified as R0.
The diagnostic accuracy was 33% for MRI and 91% for 3D-IPR.
Conclusions
The 3D-IPR method can be useful to improve diagnostic accuracy of MRI scans in assessing the relationship with surrounding structures in patients with LAPRC and LRRC
R0 or tumour-free surgical margins resection represents the strongest prognosticator of survival in locally advanced primary (LAPRC) and locally recurrent rectal cancer (LRRC). Magnetic resonance imaging (MRI) is the technique of choice to assess the infiltration of surrounding structures, currently representing the ideal tool for preoperative planning. In LAPRC, MRI negative predictive value of infiltration is reported to be 94%, however, a 54% positive predictive value has been estimated. In addition, diagnostic accuracy of MRI might not exceed 60% in LRRC.
Aim
The aim of this study was to assess the usefulness of a three-dimensional image processing and reconstruction (3D-IPR) model to achieve R0 resections and to compare the diagnostic accuracy between MRI and 3D-IPR regarding the infiltration of surrounding structures in LAPRC and LRRC.
Materials and Method
This is a prospective study performed at two referral centres for rectal cancer between January 2020 and January 2022. 3D-IPR was applied to MRI of patients with LAPRC or LRRC, before surgery. The MRI findings were compared with those of 3D-IPR, focusing on predicted surgical margins. The standard of reference was definitive pathology of the specimen
Results
Twelve patients were evaluated (7 LAPRC and 5 LRRC). A complete agreement between MRI and 3D-IPR was observed in 16% of cases.
One patient received anterior resection of the rectum, 4 en bloc rectal resection extended to surrounding structures, and 7 pelvic exenteration; 75% of specimens were classified as R0.
The diagnostic accuracy was 33% for MRI and 91% for 3D-IPR.
Conclusions
The 3D-IPR method can be useful to improve diagnostic accuracy of MRI scans in assessing the relationship with surrounding structures in patients with LAPRC and LRRC
Code of conduct/disclaimer available in General Terms & Conditions
{{ help_message }}
{{filter}}