Manual and semi-automated delineation of locally advanced rectal cancer subvolumes with diffusion-weighted MRI

Abstract

Objectives To evaluate interobserver agreement for T2 weighted (T2W) and diffusion-weighted MRI (DW-MRI) contours of locally advanced rectal cancer (LARC); and to evaluate manual and semi-automated delineations of restricted diffusion tumour subvolumes. Methods 20 cases of LARC were reviewed by 2 radiation oncologists and 2 radiologists. Contours of gross tumour volume (GTV) on T2W, DW-MRI and co-registered T2W/DW-MRI were independently delineated and compared using Dice Similarity Coefficient (DSC), mean distance to agreement (MDA) and other metrics of interobserver agreement. Restricted diffusion subvolumes within GTVs were manually delineated and compared to semi-automatically generated contours corresponding to intratumoral apparent diffusion coefficient (ADC) centile values. Results Observers were able to delineate subvolumes of restricted diffusion with moderate agreement (DSC 0.666, MDA 1.92 mm). Semi-automated segmentation based on the 40th centile intratumoral ADC value demonstrated moderate average agreement with consensus delineations (DSC 0.581, MDA 2.44 mm), with errors noted in image registration and luminal variation between acquisitions. A small validation set of four cases with optimised planning MRI demonstrated improvement (DSC 0.669, MDA 1.91 mm). Conclusion Contours based on co-registered T2W and DW-MRI could be used for delineation of biologically relevant tumour subvolumes. Semi-automated delineation based on patient-specific intratumoral ADC thresholds may standardise subvolume delineation if registration between acquisitions is sufficiently accurate. Advances in knowledge This is the first study to evaluate the feasibility of semi-automated diffusion-based subvolume delineation in LARC. This approach could be applied to dose escalation or ‘dose painting’ protocols to improve delineation reproducibility.

Semi-automatic tumor segmentation of rectal cancer based on functional magnetic resonance imaging

“…The interobserver agreement in the underlying dataset of our study was determined as DICE of 0.82 [0.07] with an MSD of 1.2 [0.4] mm. This was in good agreement with variations reported in previous studies [4] , [5] , [6] . Comparing the interobserver DICE to the DICE of our best performing model (0.72) suggested that the model may not be sufficient on its own.…”

“…Using the union instead of an individual contour represented a conservative approach that aimed at including all suspicious tissue in the ground truth. For manual delineations, the delineated volume depends not only on the observer but also on the available MR images, as noted by Hearn et al [6] . The interobserver agreement in the underlying dataset of our study was determined as DICE of 0.82 [0.07] with an MSD of 1.2 [0.4] mm.…”

“…It has also been entitled the weakest link in radiotherapy planning [3] , in part due to inter- and intraobserver variations. High interobserver variations have been reported for several cancer types including rectal cancer [4] , [5] , [6] , which was the 8th most common cancer type in 2020 and contributed 3.8% of all new reported cases globally [7] .…”

DWI-MR and PET-CT Functional Imaging for Boost Tumor Volume Delineation in Neoadjuvant Rectal Cancer Treatment

“…The same moderate agreement (DICE of 0.581) was observed regarding semi-automated diffusion-based volume delineation. Also, semi-automated delineation on specific ADC thresholds seemed to be able to standardize rectal contouring in case of accurate coregistration, applying this method in dose escalation or "dose painting" protocols (32).…”

Magnetic resonance imaging (MRI) guided proton therapy: A review of the clinical challenges, potential benefits and pathway to implementation

“…For example diffusion weighted imaging is correlated with cellularity [72][73][74], R2* imaging with hypoxia [75], and dynamic contrast-enhanced imaging with perfusion [74]. These techniques provide insight into tumour physiologic heterogeneity and factors of radio-resistance, and could be used to predict treatment response, personalise treatment and guide dose escalation [74,76]. MRI-guidance challenges the current paradigm that the tumour is treated as a uniform static whole, rather than as a heterogeneous dynamic tumour microenvironment.…”