Variability In Delineation Research Articles

Factors of interobserver variability in prostate tumor MRI delineation: impact of PI-QUAL score.

Prostate cancer ranks as the second most common cancer in men worldwide. Dose escalation to the tumor and/or the prostate improves biochemical recurrence-free survival. However, interobserver variability in lesion contouring poses a significant limitation to such therapeutic approaches. Therefore, a study of factors influencing this variability is necessary. Three independent readers delineated the index prostate lesion (DIL) using T2w and ADC sequences for each patient. Clinical data were retrospectively collected for all participants. Logistic regression analysis was employed to examine the correlation between clinical features and a mean DICE coefficient > 0.7. Features exhibiting a p value < 0.05 on univariate analysis were subjected to multivariate analysis. The study comprised 68 patients, with a median DICE coefficient of 0.69 (95% CI 0.65-0.71), wherein 42.6% (29/68) attained a mean DICE > 0.7. Univariate analysis identified the PI-QUAL score, maximum diameter of DIL, and mean DIL volume as significant (p < 0.05) predictors. In multivariate analysis, only the PI-QUAL score (p = 0.008) remained statistically associated with a DICE coefficient > 0.7. The PI-QUAL score emerges as the primary predictive factor for minimizing inter-reader variability in intraprostatic dominant lesion segmentation. These findings underscore the importance of considering PI-QUAL scores when devising focal treatment plans. Adoption of a multi-reader approach involving diverse medical specialists (radiologists, radiotherapists, urologists) is advocated, particularly for MRIs with low PI-QUAL scores. Radiotherapy is a major treatment for patients with localized prostate cancer. Dose escalation to the tumor leads to improved cancer control. Precise delineation of the dominant intraprostatic lesion (DIL) remains a limitation to focal treatments. Features influencing inter-reader variability were never evaluated. In this study, we identified that the PI-QUAL score was the sole predictor of the inter-reader delineation variability of the DIL.

Interobserver variability of clinical target volume delineation in patients undergoing breast-conserving surgery without surgical clips: a pilot study on preoperative magnetic resonance simulation

BackgroundIn patients undergoing breast-conserving therapy without surgical clip implantation, the accuracy of tumor bed identification and the consistency of clinical target volume (CTV) delineation under computed tomography (CT) simulation remain suboptimal. This study aimed to investigate the feasibility of implementing preoperative magnetic resonance (MR) simulation on delineations by assessing interobserver variability (IOV).MethodsPreoperative MR and postoperative CT simulations were performed in patients who underwent breast-conserving surgery with no surgical clips implanted. Custom immobilization pads were used to ensure the same supine position. Three radiation oncologists independently delineated the CTV of tumor bed on the images acquired from MR and CT simulation registration and CT simulation alone. Cavity visualization score (CVS) was assigned to each patient based on the clarity of the tumor bed on CT simulation images. IOV was indicated by generalized conformity index (CIgen), denoted as CIgen−CT and CIgen−MR/CT, and the distance between the centroid of mass (dCOM), denoted as dCOMCT and dCOMMR/CT. The variation of IOV in different CVS subgroups was analyzed.ResultsA total of 10 patients were enrolled in this study. The median and interquartile range (IQR) of maximum pathological diameter of the tumors in all patients were 1.55 (0.80–1.92) cm. No statistical significance was found between the volumes of CTVs on CT simulation and on MR/CT simulation registration images (p = 0.387). CIgen−MR/CT was significantly larger than CIgen−CT (p = 0.005). dCOMMR/CT was significantly smaller than dCOMCT (p = 0.037). The median and IQR of CVS in all patients were 2.34 (2.00–3.08). The difference of CIgen between CIgen−MR/CT and CIgen−CT was larger in the low CVS group (p = 0.016). The difference of dCOM showed a decreasing trend when CVS was lower, although it did not reach statistical significance (p = 0.095).ConclusionsFor patients who underwent breast-conserving surgery without surgical clip implantation, the use of preoperative MR simulation in delineating the CTV of tumor bed decreased the IOV among observers. The consistency of tumor bed identification was improved especially in cases where the margins of tumor bed were challenging to visualize on CT simulation images. The study findings offer potential benefits in reducing local recurrence and minimizing tissue irritation in the surrounding areas. Future investigation in a larger patient cohort to validate our results is warranted.

Inter-observer variation in gross tumour volume delineation of oesophageal cancer on MR, CT and PET/CT.

The aim of our study was to assess the inter-observer variability in delineation of the gross tumour volume (GTV) of oesophageal cancer on magnetic resonance (MR) in comparison to computed tomography (CT) and positron emission tomography and CT (PET/CT). Twenty-three consecutive patients with oesophageal cancer treated with chemo-radiotherapy were enrolled. All patients had PET/CT and MR imaging in treatment position. Five observers independently delineated the GTV on CT alone, MR alone, CT with co-registered MR, PET/CT alone and MR with co-registered PET/CT. Volumes of GTV were measured per patient and imaging modality. Inter-observer agreement, expressed in generalized conformity index (CIgen), volumetric conformity index (VCI), planar conformity index (PCI) and inter-delineation distance (IDD) were calculated per patient and imaging modality. Linear mixed models were used for statistical analysis. GTV volume was significantly lower on MR (33.03 cm3) compared to CT (37.1 cm3; p = 0.002) and on PET/CT MR (35.2 cm3; p = 0.018) compared to PET/CT (39.1 cm3). The CIgen was lowest on CT (0.56) and highest on PET/CT MR (0.67). The difference in CIgen between MR (0.61) and CT was borderline significant (p = 0.048). The VCI was significantly higher on MR (0.71; p = 0.007) and on CT MR (0.71; p = 0.004) compared to CT (0.67). The PCI was significantly higher on CT MR (0.67; p = 0.031) compared to CT (0.64). The largest differences were observed in the cranio-caudal direction. The highest inter-observer agreement was found for PET/CT MR and the lowest for CT. MR could reduce the difference in delineation between observers and provide additional information about the local extent of the tumour.

On the trail of CBCT-guided adaptive rectal boost radiotherapy, does daily delineation require a radiation oncologist?

IntroductionDose-escalation radiotherapy for rectal tumours is increasingly considered as a non-operative approach, with online-adaptive radiotherapy (oART) supporting this approach by correcting inter-fraction tumour position errors. However, using cone-beam computed tomography (CBCT)-guided oART requires daily target volume delineation by different operators, leading to inter-operator delineation variability and potential dosimetric issues. This study aims to compare and quantify the inter-operator and inter-professional delineation variability of the rectal boost volume on CBCT, including volumes by an automatically delineated oART treatment planning system. Materials and methodsA rectal boost volume, defined as the primary tumour extended to the entire adjacent rectal wall, was delineated on 10 CBCTs from 5 patients by 15 operators: 4 expert radiation oncologists (ROs), 4 radiation therapists (RTTs) and 7 non-expert ROs. These contours were compared between the different professional groups. A comparison to the average volume of the group (ROs, RTTs, or non-expert ROs) with the lowest delineation variability was also performed for each individual volume including the volume automatically generated by an oART treatment planning system. ResultsDelineation variability was the highest in the superior (range: 2.3–6.0 mm), and inferior (2.3–12.4 mm) directions, compared to the left (0.2–4.4 mm), right (0.3–2.0 mm), anterior (0.1–2.9 mm), and posterior (0.5–4.0 mm) directions. Non-expert ROs, RTTs, and automatic oART volume showed similar ranges of delineation errors when compared to the expert ROs’ volume, which was chosen as reference volume since this professional group showed the lowest variability. DiscussionExpert ROs showed consistent results. Other professional groups exhibit similar variability, comparable to the automatic oART volume. Therefore, RTTs could safely perform the rectal boost delineation without non-expert ROs supervision in the absence of expert ROs during CBCT-based oART. Moreover, these findings provide quantitative data to compute accurate margins for the rectal boost planning target volume in a CBCT-guided oART workflow.

Surface water potential zones delineation and spatiotemporal variation characteristics of water pollution and the cause of pollution formation in Brahmani River Basin, Odisha

Waste water reuse is an effective method for reducing the quantity of wastewater that enters the environment. As a result, the appropriateness was assessed in accordance with its composition and the global criteria for drinking water quality. This study aims to evaluate the quality of surface water of Brahmani River, Odisha, by adapting multivariate statistical approaches, namely Cluster Analysis (CA) and multi-criteria decision-making (MCDM) methods such as Simple Additive Weighting (SAW). Again, Geographical Information System (GIS) is employed to assess the drinking water's quality in relation to its application. Around 15 physicochemical parameters, gathered from seven distinct sites, following a period of 2022 till 2024. The results showed high values of EC, SO42−, NO3−, PO43−, F− and Ca2+, in three stations, while other studied parameters are within the permissible limits defined by WHO standards. The CA three distinct classifications of sites namely, low, medium, and high, based on how comparable the various physicochemical traits and pollutant levels were throughout the sampling locations. Sites H-(4–6) were included in cluster 1, site H-(3) and (7) classified under Cluster 2 and finally, H-(1–2) indicated as Cluster 3. Thus, as stated by CA, 42.86 % of tested water samples were fall into the group of good quality and about 57.14 % of examined locations resides in the poor category of water. The findings indicate which characteristics, as determined by the CA technique, were accountable for the alteration in the water quality and raise the potential of runoff from cities, farms, and industries. Subsequently, MCDM techniques has been implemented to facilitate rapid comprehension of surface water behavior and address problems with water resource management. The obtained SAW index ranges from 0.5 to 0.94. Here, H-(1), (2), and (7) grouped as highly polluted zone, and they seemed to be connected to runoff from farms and untreated industrial and municipal trash as well as local pollution sources. From this innovative study, it is found that even though the water is fit for drinking at 3 places, it is medium to highly suitable for human consumption. Therefore, there needs to be treated in some way before being used in addition to being protected against contamination. Most often, the findings indicate that appropriate management actions are important to restore the water quality of this catchment in order to support a robust and promising aquatic ecosystem. Additionally, the significance of this restoration for objective ecological policy and decision-making processes is underscored.

Impact of motion management strategies on abdominal organ at risk delineation for magnetic resonance-guided radiotherapy

Background and purposeThe impact of respiratory motion management strategies for abdominal radiotherapy, such as abdominal compression (AC) and breath hold (BH), on abdominal organ at risk (OAR) delineation on magnetic resonance imaging (MRI) is unknown. This feasibility study compared the inter- and intra- observer delineation variation on MRI acquired with AC, BH for three critical abdominal OAR. Materials and methodsT2-weighted (W) 3D MRI in free-breathing (FB) and with AC, and T1W 3D mDixon exhale BH were acquired. Four observers blinded to motion management strategy used, delineated stomach, liver, and duodenum on all MRI. One case per strategy was repeated over 6 weeks later to quantify intra-observer variation. Simultaneous truth and performance level estimation (STAPLE) contours for each OAR were generated, median and IQR mean distance to agreement (mDTA) and maximum Hausdorff distance (HD) between observer and STAPLE contours were calculated. Observers scored organ visibility on each MRI using a four-point Likert scale. ResultsA total of 27 scans including repeats were delineated. Pooled mDTA for all OARs was 1.3 mm (0.5 mm) with AC, 1.4 mm (1.0 mm) with BH, and 1.3 mm (0.5 mm) in FB. Intra-observer mDTA was highest for all organs in FB with 10.8 mm for duodenum, 1.8 mm for liver, and 2.7 mm for stomach. The pooled mean perceptual quality score value was highest for AC across organs. ConclusionsNo motion management strategy demonstrated superior similarity across OAR, emphasizing the need for personalised approaches based on individual clinical and patient factors.

2149: Delineation variability on deep-learning-based super-resolution images on head-and-neck patients

2149: Delineation variability on deep-learning-based super-resolution images on head-and-neck patients

2204: Motion management strategy and interobserver variation in abdominal OAR delineation using MRI

2204: Motion management strategy and interobserver variation in abdominal OAR delineation using MRI

1337: Delineation variability for rectal boost RT: the role of RTTs in CBCT-guided online-adaptive RT

1337: Delineation variability for rectal boost RT: the role of RTTs in CBCT-guided online-adaptive RT

The prognostic role of MRI-based radiomics in tongue carcinoma: a multicentric validation study

PurposeRadiomics is an emerging field that utilizes quantitative features extracted from medical images to predict clinically meaningful outcomes. Validating findings is crucial to assess radiomics applicability. We aimed to validate previously published magnetic resonance imaging (MRI) radiomics models to predict oncological outcomes in oral tongue squamous cell carcinoma (OTSCC).Materials and methodsRetrospective multicentric study on OTSCC surgically treated from 2010 to 2019. All patients performed preoperative MRI, including contrast-enhanced T1-weighted (CE-T1), diffusion-weighted sequences and apparent diffusion coefficient map. We evaluated overall survival (OS), locoregional recurrence-free survival (LRRFS), cause-specific mortality (CSM). We elaborated different models based on clinical and radiomic data. C-indexes assessed the prediction accuracy of the models.ResultsWe collected 112 consecutive independent patients from three Italian Institutions to validate the previously published MRI radiomic models based on 79 different patients. The C-indexes for the hybrid clinical-radiomic models in the validation cohort were lower than those in the training cohort but remained > 0.5 in most cases. CE-T1 sequence provided the best fit to the models: the C-indexes obtained were 0.61, 0.59, 0.64 (pretreatment model) and 0.65, 0.69, 0.70 (posttreatment model) for OS, LRRFS and CSM, respectively.ConclusionOur clinical-radiomic models retain a potential to predict OS, LRRFS and CSM in heterogeneous cohorts across different centers. These findings encourage further research, aimed at overcoming current limitations, due to the variability of imaging acquisition, processing and tumor volume delineation.

Stereotactic arrhythmia radioablation for refractory ventricular tachycardia. A retrospective analysis and optimisation of the planning phase

Abstract Introduction Stereotactic arrhythmia radioablation (STAR) is used as a rescue treatment for refractory ventricular tachycardia (VT). However, the planning phase to accurately delineate the clinical target volume (CTV) still suffers from limitations due to the incompatibility of data formats between 3D electro-anatomical mapping systems (EAM) used for VT ablation and STAR planning software. The concordance between the CTVs as defined by the planning software (CTVSTAR) and the CTV based on EAM data (CTVEAM) remains unknown. Purpose Within the STOPSTORM European consortium, we compared the concordance between CTVSTAR and CTVEAM in our series of 13 patients (pts) treated by STAR for refractory VT. We also evaluated need for redo ablation of VT after STAR. Method Using the software Slicer3D, two protocols were retrospectively compared: Protocol 1 comprised the manual contouring of the CTVSTAR based on 4D-CT SCAN drawn side-by-side using EAM (Carto3, Biosense Webster) screenshots and diagnostic reports. Protocol 2 comprised semiautomatic segmentation of CTVEAM based on 4D-CT SCAN using imported EAMs annotated with landmarks for VT substrate delimitation. The delineation of CTVEAM included 3 steps: (1) semiautomatic segmentation of ventricular myocardium (VM), the blood-pool of the concerned ventricle and at least 2 additional cardiac cavities; (2) 3D-alignment of EAM with the segmented cavities; and (3), 3D-delineation of CTVSTAR identified as the transmural section of VM segment surrounding the EAM’s VT substrate delimited by STAR tags (Figure 1). Volume overlap (Dice coefficient, 100% meaning complete overlap) and volume absolute difference (VAD, 0% meaning similar volumes) between CTVs were compared. The following cumulative averages were computed over time: redo ablation (RedoA) and CTV. Results Figure 1 illustrates the last treated patient (#13) in whom the delineated volumes, CTVSTAR (red) and CTVEAM (blue), displayed good matching (Dice 72% and VAD 4. Figure 2 shows the comparison between CTVSTAR and CTVEAM for the 13 pts (65±7 yo) treated with STAR, resulting in a mean Dice of 47±17% and mean VAD of 40±51%. Figure 3 shows the cumulative RedoA and the cumulative averaged CTVSTAR and CTVEAM over time. Interestingly, the need for redo ablation decreased with increasing CTVs. Conclusion Our study demonstrates the efficacy of STAR treatment for refractory VT. Our confidence in STAR led to higher CTVs and lower RedoA over time. The high variability of CTV delineation using two different protocols suggests that STAR planning phase can be optimised by appointing intraprocedural anatomical landmarks to the VT substrate.

723: Delineation variations in RT planning of rectum cancer patients using the Swedish guideline

723: Delineation variations in RT planning of rectum cancer patients using the Swedish guideline

2059: Interobserver variability in pancreatic cancer delineation between the arterial and venous phases

2059: Interobserver variability in pancreatic cancer delineation between the arterial and venous phases

Correlation between rCBV Delineation Similarity and Overall Survival in a Prospective Cohort of High-Grade Gliomas Patients: The Hidden Value of Multimodal MRI?

The accuracy of target delineation in radiation treatment planning of high-grade gliomas (HGGs) is crucial to achieve high tumor control, while minimizing treatment-related toxicity. Magnetic resonance imaging (MRI) represents the standard imaging modality for delineation of gliomas with inherent limitations in accurately determining the microscopic extent of tumors. The purpose of this study was to assess the survival impact of multi-observer delineation variability of multiparametric MRI (mpMRI) and [18F]-FET PET/CT. Thirty prospectively included patients with histologically confirmed HGGs underwent a PET/CT and mpMRI including diffusion-weighted imaging (DWI: b0, b1000, ADC), contrast-enhanced T1-weighted imaging (T1-Gado), T2-weighted fluid-attenuated inversion recovery (T2Flair), and perfusion-weighted imaging with computation of relative cerebral blood volume (rCBV) and K2 maps. Nine radiation oncologists delineated the PET/CT and MRI sequences. Spatial similarity (Dice similarity coefficient: DSC) was calculated between the readers for each sequence. Impact of the DSC on progression-free survival (PFS) and overall survival (OS) was assessed using Kaplan-Meier curves and the log-rank test. The highest DSC mean values were reached for morphological sequences, ranging from 0.71 +/- 0.18 to 0.84 +/- 0.09 for T2Flair and T1Gado, respectively, while metabolic volumes defined by PET/CT achieved a mean DSC of 0.75 +/- 0.11. rCBV variability (mean DSC0.32 +/- 0.20) significantly impacted PFS (p = 0.02) and OS (p = 0.002). Our data suggest that the T1-Gado and T2Flair sequences were the most reproducible sequences, followed by PET/CT. Reproducibility for functional sequences was low, but rCBV inter-reader similarity significantly impacted PFS and OS.

Dosimetric Impact of Delineation and Motion Uncertainties on the Heart and Substructures in Lung Cancer Radiotherapy

AimsDelineation variations and organ motion produce difficult-to-quantify uncertainties in planned radiation doses to targets and organs at risk. Similar to manual contouring, most automatic segmentation tools generate single delineations per structure; however, this does not indicate the range of clinically acceptable delineations. This study develops a method to generate a range of automatic cardiac structure segmentations, incorporating motion and delineation uncertainty, and evaluates the dosimetric impact in lung cancer. Materials and methodsEighteen cardiac structures were delineated using a locally developed auto-segmentation tool. It was applied to lung cancer planning CTs for 27 curative (planned dose ≥50 Gy) cases, and delineation variations were estimated by using ten mapping-atlases to provide separate substructure segmentations. Motion-related cardiac segmentation variations were estimated by auto-contouring structures on ten respiratory phases for 9/27 cases that had 4D-planning CTs. Dose volume histograms (DVHs) incorporating these variations were generated for comparison. ResultsVariations in mean doses (Dmean), defined as the range in values across ten feasible auto-segmentations, were calculated for each cardiac substructure. Over the study cohort the median variations for delineation uncertainty and motion were 2.20–11.09 Gy and 0.72–4.06 Gy, respectively. As relative values, variations in Dmean were between 18.7%–65.3% and 7.8%–32.5% for delineation uncertainty and motion, respectively. Doses vary depending on the individual planned dose distribution, not simply on segmentation differences, with larger dose variations to cardiac structures lying within areas of steep dose gradient. ConclusionRadiotherapy dose uncertainties from delineation variations and respiratory-related heart motion were quantified using a cardiac substructure automatic segmentation tool. This predicts the ‘dose range’ where doses to structures are most likely to fall, rather than single DVH curves. This enables consideration of these uncertainties in cardiotoxicity research and for future plan optimisation. The tool was designed for cardiac structures, but similar methods are potentially applicable to other OARs.

Interobserver variation in tumor delineation of liver metastases using Magnetic Resonance Imaging

Background and purposeMagnetic Resonance Imaging (MRI) guided stereotactic body radiotherapy (SBRT) of liver metastases is an upcoming high-precision non-invasive treatment. Interobserver variation (IOV) in tumor delineation, however, remains a relevant uncertainty for planning target volume (PTV) margins. The aims of this study were to quantify IOV in MRI-based delineation of the gross tumor volume (GTV) of liver metastases and to detect patient-specific factors influencing IOV. Materials and methodsA total of 22 patients with liver metastases from three primary tumor origins were selected (colorectal(8), breast(6), lung(8)). Delineation guidelines and planning MRI-scans were provided to eight radiation oncologists who delineated all GTVs. All delineations were centrally peer reviewed to identify outliers not meeting the guidelines. Analyses were performed both in- and excluding outliers. IOV was quantified as the standard deviation (SD) of the perpendicular distance of each observer’s delineation towards the median delineation. The correlation of IOV with shape regularity, tumor origin and volume was determined. ResultsIncluding all delineations, average IOV was 1.6 mm (range 0.6–3.3 mm). From 160 delineations, in total fourteen single delineations were marked as outliers after peer review. After excluding outliers, the average IOV was 1.3 mm (range 0.6–2.3 mm). There was no significant correlation between IOV and tumor origin or volume. However, there was a significant correlation between IOV and regularity (Spearman’s ρs = -0.66; p = 0.002). ConclusionMRI-based IOV in tumor delineation of liver metastases was 1.3–1.6 mm, from which PTV margins for IOV can be calculated. Tumor regularity and IOV were significantly correlated, potentially allowing for patient-specific margin calculation.

Feasibility and safety of contrast-enhanced magnetic resonance-guided adaptive radiotherapy for upper abdominal tumors: A preliminary exploration

This study investigates the use of contrast-enhanced magnetic resonance (MR) in MR-guided adaptive radiotherapy (MRgART) for upper abdominal tumors. Contrast-enhanced T1-weighted MR (cT1w MR) using half doses of gadoterate was used to guide daily adaptive radiotherapy for tumors poorly visualized without contrast. The use of gadoterate was found to be feasible and safe in 5-fraction MRgART and could improve the contrast-to-noise ratio of MR images. And the use of cT1w MR could reduce the interobserver variation of adaptive tumor delineation compared to plain T1w MR (4.41 vs. 6.58, p < 0.001) and T2w MR (4.41 vs. 7.42, p < 0.001).

The role of 18F-FDG PET in minimizing variability in gross tumor volume delineation of soft tissue sarcomas

BackgroundAccurate gross tumor volume (GTV) delineation is a critical step in radiation therapy treatment planning. However, it is reader dependent and thus susceptible to intra- and inter-reader variability. GTV delineation of soft tissue sarcoma (STS) often relies on CT and MR images. PurposeThis study investigates the potential role of 18F-FDG PET in reducing intra- and inter-reader variability thereby improving reproducibility of GTV delineation in STS, without incurring additional costs or radiation exposure. Materials and MethodsThree readers performed independent GTV delineation of 61 patients with STS using first CT and MR followed by CT, MR, and 18F-FDG PET images. Each reader performed a total of six delineation trials, three trials per imaging modality group. Dice Similarity Coefficient (DSC) score and Hausdorff distance (HD) were used to assess both intra- and inter-reader variability using generated simultaneous truth and performance level estimation (STAPLE) GTVs as ground truth. Statistical analysis was performed using a Wilcoxon signed-ranked test. ResultsThere was a statistically significant decrease in both intra- and inter-reader variability in GTV delineation using CT, MR 18F-FDG PET images vs. CT and MR images. This was translated by an increase in the DSC score and a decrease in the HD for GTVs drawn from CT, MR and 18F-FDG PET images vs. GTVs drawn from CT and MR for all readers and across all three trials. ConclusionIncorporation of 18F-FDG PET into CT and MR images decreased intra- and inter-reader variability and subsequently increased reproducibility of GTV delineation in STS.

Clinical Evaluation of Deep Learning for Tumor Delineation on 18F-FDG PET/CT of Head and Neck Cancer.

Artificial intelligence (AI) may decrease 18F-FDG PET/CT-based gross tumor volume (GTV) delineation variability and automate tumor-volume-derived image biomarker extraction. Hence, we aimed to identify and evaluate promising state-of-the-art deep learning methods for head and neck cancer (HNC) PET GTV delineation. Methods: We trained and evaluated deep learning methods using retrospectively included scans of HNC patients referred for radiotherapy between January 2014 and December 2019 (ISRCTN16907234). We used 3 test datasets: an internal set to compare methods, another internal set to compare AI-to-expert variability and expert interobserver variability (IOV), and an external set to compare internal and external AI-to-expert variability. Expert PET GTVs were used as the reference standard. Our benchmark IOV was measured using the PET GTV of 6 experts. The primary outcome was the Dice similarity coefficient (DSC). ANOVA was used to compare methods, a paired t test was used to compare AI-to-expert variability and expert IOV, an unpaired t test was used to compare internal and external AI-to-expert variability, and post hoc Bland-Altman analysis was used to evaluate biomarker agreement. Results: In total, 1,220 18F-FDG PET/CT scans of 1,190 patients (mean age ± SD, 63 ± 10 y; 858 men) were included, and 5 deep learning methods were trained using 5-fold cross-validation (n = 805). The nnU-Net method achieved the highest similarity (DSC, 0.80 [95% CI, 0.77-0.86]; n = 196). We found no evidence of a difference between expert IOV and AI-to-expert variability (DSC, 0.78 for AI vs. 0.82 for experts; mean difference of 0.04 [95% CI, -0.01 to 0.09]; P = 0.12; n = 64). We found no evidence of a difference between the internal and external AI-to-expert variability (DSC, 0.80 internally vs. 0.81 externally; mean difference of 0.004 [95% CI, -0.05 to 0.04]; P = 0.87; n = 125). PET GTV-derived biomarkers of AI were in good agreement with experts. Conclusion: Deep learning can be used to automate 18F-FDG PET/CT tumor-volume-derived imaging biomarkers, and the deep-learning-based volumes have the potential to assist clinical tumor volume delineation in radiation oncology.

Interobserver delineation variability of computed tomography-based radiomic features of the parotid gland.

To assess the interobserver delineation variability of radiomic features of the parotid gland from computed tomography (CT) images and evaluate the correlation of these features for head and neck cancer (HNC) radiotherapy patients. Contrast-enhanced CT images of 20 HNC patients were utilized. The parotid glands were delineated by treating radiation oncologists (ROs), a selected RO and AccuContour auto-segmentation software. Dice similarity coefficients (DSCs) between each pair of observers were calculated. A total of 107 radiomic features were extracted, whose robustness to interobserver delineation was assessed using the intraclass correlation coefficient (ICC). Pearson correlation coefficients (r) were calculated to determine the relationship between the features. The influence of excluding unrobust features from normal tissue complication probability (NTCP) modeling was investigated for severe oral mucositis (grade ≥3). The average DSC was 0.84 (95% confidence interval, 0.83-0.86). Most of the shape features demonstrated robustness (ICC ≥0.75), while the first-order and texture features were influenced by delineation variability. Among the three observers investigated, 42 features were sufficiently robust, out of which 36 features exhibited weak correlation (|r|<0.8). No significant difference in the robustness level was found when comparing manual segmentation by a single RO or automated segmentation with the actual clinical contour data made by treating ROs. Excluding unrobust features from the NTCP model for severe oral mucositis did not deteriorate the model performance. Interobserver delineation variability had substantial impact on radiomic features of the parotid gland. Both manual and automated segmentation methods contributed similarly to this variation.