Clinical Target Volume Delineation Research Articles

Proposals for the delineation of neck clinical target volume for definitive Radiation therapy in patients with oral/ oropharyngeal squamous cell cancer based on lymph node distribution

Purpose/Objective(s)To establish the distribution pattern of cervical lymph node metastasis (LNM) and propose optimized clinical target volume (CTV) boundaries specific to oral/ oropharyngeal squamous cell cancer (OSCC/OPSCC). Materials/Methods531 patients with pathologically confirmed OSCC/OPSCC were enrolled from January 2013 to June 2022. Patients were stratified into two groups based on the minimal distance from the lesion's edge to the body's midline: ≤1 cm or > 1 cm. The geometric center of cervical metastatic LN was marked on a template CT. LN distribution probability maps were established. The relationships between the LN distribution and consensus guidelines were analyzed to propose modifications for CTV boundaries specific to OSCC/OPSCC. ResultsA total of 1962 positive LNs were enrolled. Compared with the > 1 cm group, the ≤ 1 cm group has following feature tendencies: male smokers, younger, median organs, large gross lesion, infiltrative growth pattern, contralateral LNM. The most frequently involved level of LNM was ipsilateral II, but ipsilateral Ib had the highest involvement rate in the > 1 cm OSCC group. In addition, tongue cancer had a higher incidence of LN extranodal extension (ENE), which mainly distributes in ipsilateral level II. The skip metastasis was prone to from level III to Vb (3.5 %) in LN(+)/ENE (-), and level Ib to VIa (3.7 %) in LN(+)/ENE (+). Accordingly, we proposed the following modifications: 1. only including lateral and posterior margin of submandibular gland within 5 mm; 2. retracting posterior boundary of level II to front edge of levator scapula muscle, and descending the upper boundary to transverse process of C2 vertebra only for OSCC; 3. including posterior third of thyroglossal muscle or anterior edge of sternocleidomastoid muscle; 4. sparing level Va in case of only level II involvement; 5. including upper area of the thyroid cartilage plate in case of level Ib LN(+)/ENE (+); 6. sparing level VIIa is considered. ConclusionThis is the first description of LN topographic spread patterns for OSCC/OPSCC. Modified CTV for prophylactic irradiation was proposed to spare the organs at risk and minimize adverse effects.

Nodal recurrence mapping and clinical target volumes after resection of intrahepatic cholangiocarcinoma or combined hepatocellular-cholangiocarcinoma

BackgroundScarce evidence exists for clinical target volume (CTV) definitions of regional lymph nodes (LNs) in intrahepatic cholangiocarcinoma (iCCA) or combined hepatocellular-cholangiocarcinoma (cHCC-CCA). We investigated the mapping pattern of nodal recurrence after surgery for iCCA and cHCC-CCA and provided evidence for the nodal CTV definition. MethodsWe retrospectively reviewed the medical records of patients with iCCA or cHCC-CCA who underwent surgery between 2010 and 2020. Eligibility criteria included patients pathologically diagnosed with iCCA or cHCC-CCA after surgery and a first recurrent event in regional LNs during follow-up. All recurrent LNs were registered onto reference computed tomography images based on the vascular structures to reconstruct the node mapping. Fifty-three patients were eligible. LN regions were classified into four risk groups. ResultsHepatic hilar and portal vein-vena cava were the most common recurrent regions, with recurrence rates of 62.3 % and 39.6 % (high-risk regions), respectively. Recurrence rates in the left gastric, diaphragmatic, common hepatic, superior mesenteric vessels, celiac trunk, and paracardial regions ranged from 15.1 % to 30.2 % (intermediate-risk regions). There were fewer recurrences in the para-aortic (16a1, a2, b1) and splenic artery and hilum regions, with rates <10 % (low-risk regions). No LN recurrence was observed in the para-oesophageal or para-aortic region (16b2) (very low-risk regions). Based on node mapping, the CTV should include high- and intermediate-risk regions for pathologically negative LN patients during postoperative radiotherapy. Low-risk regions should be included for pathologically positive LN patients. ConclusionWe provide evidence for CTV delineation in patients with iCCA and cHCC-CCA based on recurrent LN mapping.

The individualized delineation of clinical target volume for primary nasopharyngeal carcinoma based on invasion risk of substructures: A prospective, real-world study with a large population

Background and purposeThe delineation of clinical target volume (CTV) for primary nasopharyngeal carcinoma (NPC) is currently controversial and the international guideline still recommend a uniform border for CTV regardless of the tumor extent. We conducted this prospective, real-world study to evaluate the clinical outcomes of our individualized CTV delineation method based on distance plus substructures. Materials and MethodsWe preliminarily investigated the local extension patterns of NPC on 354 newly diagnosed patients and defined the structures surrounding the nasopharynx as Level-1 to Level-4 substructures stratified by the risk of invasion. We then enrolled patients with newly diagnosed NPC without distant metastasis to investigate our individualized CTV delineation protocol. All patients received intensity modulated radiotherapy. CTV1 and CTV2 were prescribed doses of 60 Gy and 54 Gy in 30 ∼ 33 fractions. The primary endpoint was local recurrence-free survival (LRFS); secondary endpoints included regional control and survival, estimated using the Kaplan-Meier method. The local failure patterns were also analyzed. ResultsFrom January 2008 to December 2012 and from January 2013 to September 2019, 356 and 648 patients were enrolled, named as training set and validation set, respectively. With a median follow-up of 104.6 (interquartile, 73.1–126.9) and 51.4 (39.5–78.5) months, 31 (8.7 %) and 38 (5.9 %) patients in training and validation sets experienced local recurrence, and the 5-year LRFS was 93.0 % and 93.2 %, respectively; 63 (17.7 %) and 39 (6 %) patients died in training and validation sets, and the 5-year overall survival (OS) was 88.5 % and 93.4 %, respectively. For the whole study cohort (N = 1004) with a median follow-up of 66.6 (41.5–98.0) months, the 5-year LRFS and OS was 93.2 % and 91.5 %. The grade 3 late toxicities included xerostomia, subcutaneous fibrosis, hearing impairment, trismus, visuality impairment and skin atrophy, with a total incidence of 1.5 %. Sixty-seven of 69 (97.1 %) local recurrence was in high-dose area. ConclusionOur individualized CTV delineation method can achieve favorable local tumor control and long-term survival outcomes with acceptable late toxicities.

Segmentation of 71 Anatomical Structures Necessary for the Evaluation of Guideline-Conforming Clinical Target Volumes in Head and Neck Cancers.

The delineation of the clinical target volumes (CTVs) for radiation therapy is time-consuming, requires intensive training and shows high inter-observer variability. Supervised deep-learning methods depend heavily on consistent training data; thus, State-of-the-Art research focuses on making CTV labels more homogeneous and strictly bounding them to current standards. International consensus expert guidelines standardize CTV delineation by conditioning the extension of the clinical target volume on the surrounding anatomical structures. Training strategies that directly follow the construction rules given in the expert guidelines or the possibility of quantifying the conformance of manually drawn contours to the guidelines are still missing. Seventy-one anatomical structures that are relevant to CTV delineation in head- and neck-cancer patients, according to the expert guidelines, were segmented on 104 computed tomography scans, to assess the possibility of automating their segmentation by State-of-the-Art deep learning methods. All 71 anatomical structures were subdivided into three subsets of non-overlapping structures, and a 3D nnU-Net model with five-fold cross-validation was trained for each subset, to automatically segment the structures on planning computed tomography scans. We report the DICE, Hausdorff distance and surface DICE for 71 + 5 anatomical structures, for most of which no previous segmentation accuracies have been reported. For those structures for which prediction values have been reported, our segmentation accuracy matched or exceeded the reported values. The predictions from our models were always better than those predicted by the TotalSegmentator. The sDICE with 2 mm margin was larger than 80% for almost all the structures. Individual structures with decreased segmentation accuracy are analyzed and discussed with respect to their impact on the CTV delineation following the expert guidelines. No deviation is expected to affect the rule-based automation of the CTV delineation.

Optimization of supraclavicular lymph node clinical target volume delineation in high-risk breast cancer: a single center experience and recommendation

BackgroundProphylactic irradiation of supraclavicular lymph node drainage areas can improve the regional control rate of lymph node-positive or lymph node-negative disease but a locally-advanced stage breast cancer, and it can reduce breast cancer-related mortality. However, many controversies exist in the clinical target volume delineation of supraclavicular lymph node drainage in patients with breast cancer.MethodsWe retrospectively analyzed 42 patients with breast cancer and supraclavicular lymph node metastasis at our hospital between January 2017 and December 2021. Among these cases, 32 were locally advanced and 10 were stage IV at initial treatment. A patient with breast cancer who did not undergo dissection of the supraclavicular and infraclavicular lymph nodes at our hospital was selected as a standard patient. A contrast-enhanced computed tomography (CT) scan for positioning was used as a template image, and blood vessels, muscles, and bony landmarks were used as references for positioning. The metastatic supraclavicular lymph nodes were identified in all enrolled patients and projected into the template CT images.ResultsThe metastastic pattern of supraclavicular lymph node in breast cancer was proposed: distribution along the posterolateral border of the internal jugular vein (medial supraclavicular group) and along the transverse jugular vein (lateral supraclavicular group). We theorized that the lateral and posterior borders of the clinical target volume in the supraclavicular region should include the lymph nodes in the posterior triangle of the neck (level V) in high-risk individuals. If the metastatic axillary lymph node is extensive, then the superior border of the supraclavicular region should be moved upward appropriately.ConclusionsThis study analyzed patients with breast cancer and supraclavicular lymph node metastasis at initial treatment, explored the metastastic pattern of supraclavicular lymph node, and applied anatomical knowledge to further optimize the target volume delineation of supraclavicular lymph node drainage area in high-risk breast cancer.

Integration of electro-anatomical maps in the planning phase of stereotactic arrhythmia radioablation for refractory ventricular tachycardia

Abstract Introduction Despite the efficacy of stereotactic arrhythmia radioablation (STAR) for the treatment of refractory ventricular tachycardia (VT), the planning phase to precisely delineate the arrhythmogenic substrate still suffers from limitations due to the incompatibility of data formats between 3D electro-anatomical mapping systems (EAM) and STAR planning software (CyberKnife). Purpose Within the STOPSTORM European consortium, we present a workflow to integrate all 3D EAM data (e.g. voltage, activation time, anatomical tags) into the STAR planning software using the open source software Slicer3D to improve targeting of the arrhythmogenic substrate. Method Two protocols were compared retrospectively. Protocol 1 (STAR gold-standard) comprised the manual contouring of the clinical target volume (CTV) on planning 4D-CAT SCAN drawn side-by-side using EAM (Carto3, Biosense Webster) screen-shots and diagnostic reports. Protocol 2 comprised semiautomatic segmentation of CTV on planning 4D-CAT SCAN using imported EAMs annotated with landmarks for VT target delimitation. The semiautomatic segmentation of CTVs included three steps: (1) semiautomatic segmentation of ventricular myocardium (VM), the blood-pool of the concerned ventricle and at least two additional cardiac cavities; (2) 3D alignment of EAM with the segmented cavities; and (3), 3D delineation of CTV identified as the transmural section of VM segment surrounding the EAM’s VT target delimited by the annotated landmarks. The volume overlap (Dice coefficient) and the absolute difference of the CTVs obtained with the two protocols were compared. Both measurements were expressed in percentage as follows: Dice coefficient 0% means no overlap and 100% complete overlap; absolute difference of volumes 0% means no difference and 100% means that one CTV was twice the size of the CTV generated by the other protocol. Results Both protocols were applied to four patients (65.5 ± 8.5 yo) with refractory VT successfully treated with STAR. Figure 1 shows an example (Patient 1) of the CTV contouring using the CARTO EAMs aligned with the CT images in Slicer3D. Figure 2 illustrates significant mismatch between the segmented CTVs of the two protocols. The median volume overlap of the CTVs was 44%, while the median volume absolute difference was 35%. Table 1 reports detailed results. Conclusion Our study reveals a high variability of CTV delineation for STAR using two different protocols. The intraprocedural assignment of anatomical landmarks for VT target delimitation on EAM appears promising to improve CTV delineation. Further research is needed to evaluate the correlation between STAR outcome and CTV delineation accuracy.

Impact of digital positron emission tomography/computed tomography on the delineation of clinical target volume in advanced lung cancer.

The present study investigated the differences between digital [18F]-Fluorodeoxyglucose (FDG) positron emission tomography [PET]/computed tomography [CT] (dPET/CT) and conventional PET/CT (cPET/CT) in delineating the clinical target volume (CTV) in patients with advanced lung cancer in the involved field radiation therapy (IFRT) era. Patients with advanced lung cancer were scanned using two dual-imaging protocols (dPET/CT and cPET/CT). Two virtual delineations contoured with reference to dPET/CT and cPET/CT images were created for each patient by five radiation oncologists. Changes in the delineation of target volumes in each patient were examined. A total of 10 patients [male/female, 9/1; median age, 65 years (range, 58-80 years)] were enrolled between April 2020 and September 2020. Significant changes in the delineation of CTVs were uncommon between dPET/CT and cPET/CT. A notable increase in CTVn was observed in 10% of the patients (1/10; P<0.05; Smirnov-Grubbs analysis). In this patient, a node that was not assessed as lymph node metastasis when cPET/CT was used was assessed as lymph node metastasis when dPET/CT was used and was included in the CTVn by all five radiation oncologists. In patients with advanced lung cancer, notable changes in CTV delineations are uncommon, regardless of whether dPET/CT or cPET/CT is used. However, in some cases, CTVn delineation with reference to dPET/CT may improve the treatment outcomes of IFRT for advanced lung cancer.

Artificial Intelligence–Based Autosegmentation: Advantages in Delineation, Absorbed Dose-Distribution, and Logistics

PurposeTo compare the performance of artificial intelligence (AI) in auto-contouring compared to a human practitioner in terms of precision, differences in dose distribution and time consumption. Methods and MaterialsDatasets of previously irradiated patients in three different segments (head and neck- (HNC), breast- (BC) and prostate cancer (PC)) were retrospectively collected. An experienced radiation oncologist (MD) performed organs-at-risk (OARs) and standard clinical target volume (CTV) delineations as baseline structures for comparison. AI-based auto-contours were generated in two additional CT copies; therefore, three groups were assessed: MD alone, AI alone, AI plus MD corrections (AI+C). Differences in Dice similarity coefficient (DSC) and person-hour burden were assessed. Furthermore, changes in clinically relevant dose-volume parameters were evaluated and compared. ResultsSeventy-five previously treated cases were collected (25 per segment) for the analysis. When compared to MD contours, the mean DSC scores were higher than 0.7 for 74% and 80% of AI and AI+C, respectively. After corrections, 17.1% structures presented DSC score deviations higher than 0.1 and 10.4% dose-volume parameters significantly changed in AI-contoured structures. The time consumption assessment yielded mean person-hour reductions of 68%, 51% and 71% for BC, PC and HNC, respectively. ConclusionIn great extent, AI yielded clinically acceptable OARs and certain CTVs in the explored anatomical segments. Sparse correction and assessment requirements place AI+C as a standard workflow. Minimal clinically relevant differences in OAR-exposure were identified. A substantial amount of person-hours could be repurposed with this technology.

Prior information guided auto-segmentation of clinical target volume of tumor bed in postoperative breast cancer radiotherapy

BackgroundAccurate delineation of clinical target volume of tumor bed (CTV-TB) is important but it is also challenging due to surgical effects and soft tissue contrast. Recently a few auto-segmentation methods were developed to improve the process. However, those methods had comparatively low segmentation accuracy. In this study the prior information was introduced to aid auto-segmentation of CTV-TB based on a deep-learning model.MethodsTo aid the delineation of CTV-TB, the tumor contour on preoperative CT was transformed onto postoperative CT via deformable image registration. Both original and transformed tumor contours were used for prior information in training an auto-segmentation model. Then, the CTV-TB contour on postoperative CT was predicted by the model. 110 pairs of preoperative and postoperative CT images were used with a 5-fold cross-validation strategy. The predicted contour was compared with the clinically approved contour for accuracy evaluation using dice similarity coefficient (DSC) and Hausdorff distance.ResultsThe average DSC of the deep-learning model with prior information was improved than the one without prior information (0.808 vs. 0.734, P < 0.05). The average DSC of the deep-learning model with prior information was higher than that of the traditional method (0.808 vs. 0.622, P < 0.05).ConclusionsThe introduction of prior information in deep-learning model can improve segmentation accuracy of CTV-TB. The proposed method provided an effective way to automatically delineate CTV-TB in postoperative breast cancer radiotherapy.

Clinical outcome after high dose rate intracavitary brachytherapy with traditional point 'A' dose prescription in locally advanced carcinoma of uterine cervix: dosimetric analysis from the perspective of computed tomography imaging-based 3-dimensional treatment planning.

To analyze tumour response and toxicity with respect to cumulative radiotherapy dose to target and organs at risk (OARs) with computed tomography (CT)-based image guided adaptive brachytherapy planning for locally advanced carcinoma cervix. Patients were treated with two-dimensional concurrent chemoradiotherapy to whole pelvis followed by intracavitary brachytherapy (ICBT) with dose prescription to point 'A'. CT image-based delineation of high-risk clinical target volume (HR-CTV), urinary bladder, rectum and sigmoid colon was done with generation of dose-volume histogram (DVH) data and optimization of doses to target and OARs. Follow up assessments were done for response of disease and toxicity with generation of data for statistical analysis. One hundred thirty-six patients were enrolled in the study. Delineated volume of HR-CTV ranged from 20.9 to 37.1 mL, with median value of 30.2 mL. The equivalent dose in 2 Gy per fraction (EQD2) for point 'A' ranged from 71.31 to 79.75 Gy with median value of 75.1 Gy and EQD2 HR-CTV D90 ranged from 71.9 to 89.7 Gy with median value of 85.1 Gy. 69.2% of patients showed complete response and after median follow-up of 25 months, 50 patients remained disease free, of whom, 74.0% had received ≥85 Gy to HR-CTV D90 versus 26.0% receiving <85 Gy to HR-CTV D90. s Amidst the unavailability of magnetic resonance imaging facilities in low middle income countries, incorporation of CT-image based treatment planning into routine practice for ICBT provides the scope to delineate volumes of target and OARs and to generate DVH data, which can prove to be a better surrogate for disease response and toxicity.

Towards deep-learning (DL) based fully automated target delineation for rectal cancer neoadjuvant radiotherapy using a divide-and-conquer strategy: a study with multicenter blind and randomized validation

PurposeManual clinical target volume (CTV) and gross tumor volume (GTV) delineation for rectal cancer neoadjuvant radiotherapy is pivotal but labor-intensive. This study aims to propose a deep learning (DL)-based workflow towards fully automated clinical target volume (CTV) and gross tumor volume (GTV) delineation for rectal cancer neoadjuvant radiotherapy.Materials & methodsWe retrospectively included 141 patients with Stage II-III mid-low rectal cancer and randomly grouped them into training (n = 121) and testing (n = 20) cohorts. We adopted a divide-and-conquer strategy to address CTV and GTV segmentation using two separate DL models with DpuUnet as backend-one model for CTV segmentation in the CT domain, and the other for GTV in the MRI domain. The workflow was validated using a three-level multicenter-involved blind and randomized evaluation scheme. Dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (95HD) metrics were calculated in Level 1, four-grade expert scoring was performed in Level 2, and head-to-head Turing test in Level 3.ResultsFor the DL-based CTV contours over the testing cohort, the DSC and 95HD (mean ± SD) were 0.85 ± 0.06 and 7.75 ± 6.42 mm respectively, and 96.4% cases achieved clinical viable scores (≥ 2). The positive rate in the Turing test was 52.3%. For GTV, the DSC and 95HD were 0.87 ± 0.07 and 4.07 ± 1.67 mm respectively, and 100% of the DL-based contours achieved clinical viable scores (≥ 2). The positive rate in the Turing test was 52.0%.ConclusionThe proposed DL-based workflow exhibited promising accuracy and excellent clinical viability towards automated CTV and GTV delineation for rectal cancer neoadjuvant radiotherapy.

Prior knowledge based deep learning auto-segmentation in magnetic resonance imaging-guided radiotherapy of prostate cancer

Background and purposeAutomation is desirable for organ segmentation in radiotherapy. This study compared deep learning methods for auto-segmentation of organs-at-risk (OARs) and clinical target volume (CTV) in prostate cancer patients undergoing fractionated magnetic resonance (MR)-guided adaptive radiation therapy. Models predicting dense displacement fields (DDFMs) between planning and fraction images were compared to patient-specific (PSM) and baseline (BM) segmentation models. Materials and methodsA dataset of 92 patients with planning and fraction MR images (MRIs) from two institutions were used. DDFMs were trained to predict dense displacement fields (DDFs) between the planning and fraction images, which were subsequently used to propagate the planning contours of the bladder, rectum, and CTV to the daily MRI. The training was performed either with true planning-fraction image pairs or with planning images and their counterparts deformed by known DDFs. The BMs were trained on 53 planning images, while to generate PSMs, the BMs were fine-tuned using the planning image of a given single patient. The evaluation included Dice similarity coefficient (DSC), the average (HDavg) and the 95th percentile (HD95) Hausdorff distance (HD). ResultsThe DDFMs with DSCs for bladder/rectum of 0.76/0.76 performed worse than PSMs (0.91/0.90) and BMs (0.89/0.88). The same trend was observed for HDs. For CTV, DDFM and PSM performed similarly yielding DSCs of 0.87 and 0.84, respectively. ConclusionsDDFMs were found suitable for CTV delineation after rigid alignment. However, for OARs they were outperformed by PSMs, as they predicted only limited deformations even in the presence of substantial anatomical changes.

Art for Advanced Radiation Oncology – Synchronizing the Past and the Future for a Better Patient Care

"Professor Maria Antonietta Gambacorta is a specialist in Radiation Oncology with a PhD in Brachytherapy. She is responsible for the Radiotherapy Oncology hospitalizations and the Interventional Oncology Center and the coordinator of the Multidisciplinary Group for the treatment of rectal tumors within the Comprehensive Cancer Centre of the Fondazione Policlinico Agostino Gemelli, Rome. Due to her level of expertise, she was elected member of the Board of Directors of the Italian Association of Radiation Oncology (AIRO); and member of the board for the drafting of the Guidelines on rectal cancer of the Italian Association of Medical Oncology. Professor Gambacorta is a member of European Society of Radiation Oncology (ESTRO) where she carries out teaching activities in international courses of the ESTRO school. She has published more than 170 scientific publications in peer-reviewed journals, including the International consensus guidelines for Clinical Target Volume delineation in rectal cancer. In this interview, Professor Gambacorta shares some insights about the Art for ART project, implemented in the Radiotherapy department of Policinico Agostino Gemelli, Rome."

GTV Based Automatic Delineation of Clinical Target Volume for Cervical Cancer

The delineation of gross tumor volume (GTV) and clinical target volume (CTV) are two critical steps in the radiotherapy planning for cervical cancer. GTV defines the primary treatment region for the gross tumor, while CTV is the area surrounding GTV that includes a certain probability (5% to 10%) of subclinical lesions. In contrast to GTV, CTV delineation relies on predefined and judgment-based boundaries, and the high variability among users makes this task particularly challenging. In this study, we evaluated the potential relationship between GTV and CTV and developed an automatic CTV delineation algorithm for cervical cancer based on the fusion of GTV information. We introduced position and shape constraints of GTV to improve the accuracy of CTV delineation. The GTV-Net deep learning method was used to segment the CTV images of cervical cancer. The method aimed to use the delineation results of the GTV region for one-hot coding and add human anatomy experience in the clinical field to guide the CTV segmentation. This retrospective study included 545 cervical cancer patients who received radiation therapy from June 2017 to May 2019, including postoperative and radical treatment groups. The CTV and GTV regions were manually delineated by human experts. Numerous experiments were conducted to evaluate the performance of the network. First, compared with different network architectures, the Dice similarity coefficient (DSC) and 95% Hausdorff distance (95HD) of GTV-Net were both improved. Then, we compared the GTV-Net method with two resident physicians. Our GTV-Net method outperformed both resident physicians. In the postoperative group, our method improved the DSC by 4% compared to 3D-UNet, reaching 76.55%, and increased by about 2.57% compared to V-Net's 73.98%, with an improvement of approximately 1.23% compared to the two resident physicians. In the radical treatment group, compared to 3D-UNet's 78.76%, our method increased the DSC by about 3.25%, reaching 82%, and increased by approximately 2.08% compared to V-Net's 79.92%, with an improvement of about 1.35% compared to the two resident physicians. Compared with 3D-UNet, the average 95HD in the postoperative group decreased from 1.489 to 1.457, and in the radical treatment group, it decreased from 1.454 to 1.433. The results of 95HD also showed some improvement compared to V-Net. This study is the first to introduce GTV information for automatic segmentation of the clinical target area for cervical cancer. In this experiment, we observed a positive gain in CTV target automatic delineation guided by GTV information compared to solely performing CTV segmentation, with an improvement in Dice similarity of more than 4% and Hausdorff distance of more than 6% in the experimental dataset. In addition, GTV-guided CTV automatic delineation has also shown promising results on multicenter data, which will better serve the clinical field.

Clinical prior Knowledge-Based One-Shot Learning for Automatic Delineation of Clinical Target Volumes in Adaptation Radiotherapy of Cervical Cancer

Rapid and accurate delineation of clinical target volumes (CTV) of cervical cancer is the crux to ensure the efficiency and benefits of adaptation radiotherapy (ART). However, contour propagation using deformation image registration (DIR) is difficult to ensure the accuracy of CTV contours due to the significant tumor recession in next fraction, and the tumor progress in each fraction is not considered by conventional automatic delineation methods based on deep learning (DL). Currently, one-shot learning (OSL) is feasible to learn the tumor progress from former fractions to improve the accuracy of automatically delineating CTV. We retrospectively collected 45 patients with cervical cancer from January 2021 to May 2022 in our department. All patients consist of a pair of planning CT and daily CT in ART. A personalized automatic delineation method based on one-shot learning was developed to delineate CTV in daily CT by learning the clinical prior knowledge from the CTV contours and images of planning CT. The performance of our proposed method was evaluated by dice similarity coefficient (DSC), 95% Harsdorff distance (95HD) and average surface distance (ASD) with human experts, and its automatic delineation performance were compared with DIR and DL in daily CT. Our automatic delineation method OSL performed the best results in all evaluation metrics (denoted by mean ± standard deviation) as shown in Table 1, it is superior to method DL: 0.92 & 0.90 of DSC, 2.33 mm & 2.68 mm of HD95, 0.68 mm & 0.82 mm of ASD, P < 0.05 for DSC and ASD. Specifically, our method is significantly superior to the automatic delineation results by method DIR: 0.92 & 0.84 of DSC, 2.33 mm & 4.11 mm of HD95, 0.68 mm & 1.52 mm of ASD, P < 0.05 for all. In addition, OSL can significantly overcome the delineation problems in fuzzy boundary and delineation missing and perform better generalization for some unusual images, compared with DIR and DL. We proposed an automatic delineation method based on one-shot learning for CTV of cervical cancer in ART, the results demonstrated that the proposed method could improve the precision and generalization of automatically delineating CTV compared against current popular methods. Therefore, it is potential to improve the quality and efficiency of ART for personalized patients and have a positive impact on tumor control and patient survival.

Automated Contouring of Cervical Lymph Nodes and Clinical Target Volumes for Nasopharyngeal Carcinoma Based on Deep Learning and Experience Constraints

Application of artificial intelligence (AI) for automated contouring of tumor volumes and organs at risk (OARs) for radiotherapy of nasopharyngeal carcinoma (NPC) leads to improved contouring accuracy and efficiency. However, few studies have involved the automated contouring of gross tumor volume of cervical lymph nodes (GTVn) and clinical target volumes (CTVs). In this work, we proposed an AI automated contouring tool for GTVn and CTVs for radiotherapy of NPC on the plain scans of planning compute tomography (CT). In this retrospective study, plain scan datasets of planning CT covering the nasopharynx and neck from 139 patients with NPC between March 2022 and December 2022 were collected and divided into training, validation, and testing cohorts of 95, 24, and 20 patients, respectively. Ground truth contours of primary gross tumor volume (GTVp), GTVn (divided into GTVn_L in left neck and GTVn_R in right neck), CTVs (including high risk CTV1 contains GTVp and low risk CTV2 contains GTVp and cervical nodal levels) and OARs were delineated and were defined by consensus of two experts. We first proposed a three-dimensional (3D) U-net using GTVp and OARs as experience constrains to guide the automated delineation of GTVn and CTVs. The average Dice similarity coefficient (DSC) and average surface distance (ASD) were used to quantify the performance of the AI tool. Next, five prospective patients were enrolled for clinical evaluation of our AI tool. DSC between automated contours and radiation oncologist-revised contours and time consuming of the revision were record. Clinical characteristics of 139 retrospective and 5 prospective patients are list in Table 1. In the independent testing set of 20 patients, our AI tool showed high performance in GTVn and CTVs contouring when compared with the ground truth contours. The mean DSC were 0.73 ± 0.07, 0.74 ± 0.05, 0.93 ± 0.03, and 0.88 ± 0.03, and the mean ASD were 1.01 ± 0.43 mm, 1.14 ± 0.61 mm, 0.51 ± 0.13 mm, 1.17 ± 0.43 mm for GTVn_L, GTVn_R, CTV1 and CTV2, respectively. In the five prospective patients, mean DSC were 0.74 ± 0.07, 0.74 ± 0.10, 0.95 ± 0.01 and 0.89 ± 0.04, respectively. The median time consuming for GTVn and CTVs revision was 2minutes and 10 seconds (range, 1 minutes to 3 minutes). The proposed AI tool integrating clinical experience as constrains showed high accuracy for contouring GTVn and CTVs of NPC. With the assistance of AI contours, contouring efficiency could be probably increased, which is promising in online adaptive radiotherapy of NPC.

Proposals for the Delineation of Neck Clinical Target Volume Specific to Oral Squamous Cell Cancer Based on Lymph Node Distribution

To establish the regional lymph node (LN) distribution probability map and draw the neck clinical target volume specific to oral squamous cell cancer (OSCC). A total of 293 patients with pathologically proven OSCC were enrolled from Jan 2014 to Jun 2022. The patients were divided into two groups according to the ≤1 cm or>1 cm of minimal distance from the lesion's edge to the midline of body. The center point of the LNs with a minimal axial diameter of ≥10 mm was marked on a single treatment planning computed tomography scan. The neck LN levels I to X were based on 2013 updated international consensus guidelines. LN distribution probability maps were established. The relationships between the LN distribution and consensus guidelines were analyzed to propose modifications for clinical target volume boundaries specific to OSCC. A total of 1033 LNs were marked. Compared with the >1 cm group, The ≤1 cm group has following feature tendencies: strong addiction to smoking and drinking, predilection for tongue and oral floor cancer, larger primary lesion with mainly infiltrative growth pattern, more frequent extra-nodal extension with mainly contralateral LNs. For the ≤1cm group, the distribution probability of LNs was: ipsilateral II (42.67%), ipsilateral Ib (17.35%), contralateral II (11.18%), ipsilateral III (8.74%), contralateral Ib (7.46%), Ia (6.43%). The other group was: ipsilateral Ib (43.14%), ipsilateral II (37.64%), contralateral Ib (4.31%), ipsilateral III (3.92%), contralateral II (3.53%), Ia (0.78%). Accordingly, the proposals for the delineation of neck clinical target volume were: 1. the inferior boundary of level Ia extends until the level of the thyroid cartilage plate; 2. the upper boundary of level Ib descends to the level of the mandibular angle, and the inner posterior boundary only includes the lateral and posterior margin of the submandibular gland within 5mm; 3. the upper boundary of level II changes to the transverse process of the second cervical vertebra, and the posterior boundary retracts to the posterior edge of the levator scapula. 4. the anterior inner border of level IVa contains only the internal jugular arteriovenous sheath. This is the first description of LN topographic spread patterns for OSCC. Optimization of CTV delineation is helpful to improve the treatment outcome and reduce adverse events.

A Multicenter Study on Deep Learning for Glioblastoma Auto-Segmentation with Prior Knowledge in Multimodal Imaging

A precise radiotherapy plan is required to ensure accurate delineation of gross tumor volumes (GTV) and clinical target volumes (CTV1 and CTV2) of glioblastomas (GBMs). However, traditional manual delineation is labor intensive and highly dependent on oncologists' experience. To construct and evaluate a deep learning-based automatic delineation method using prior knowledge in multimodal medical imaging to automate precise GTV, CTV1 and CTV2 contouring in GBM patients. We retrospectively collected the CT and MRI scans of 55 eligible patients with histologically proven high-grade glioma (HGG) from an institute, these scans were performed with non-enhanced CT (CT), contrast-enhanced T1-weighted (T1C) and T2-FLAIR (T2F) sequences. We proposed a two-stage automatic segmentation framework (PKMI-Net) for GTV, CTV1 and CTV2 based on deep learning using prior knowledge in multimodal medical imaging, and its segmentation performance was evaluated with dice similarity coefficient (DSC), 95% Harsdorff distance (HD95), average surface distance (ASD) and relative volume difference (RVD). To further investigate the generalizability of our method, we designed and conducted two evaluation strategies (Mix and Cross) on four multicenter datasets (including 55 patients, 37 patients, 21 patients and 35 patients). The evaluation results with an 11-patient test set from the single institute were summarized in Table 1, the proposed method demonstrated the best accuracy in segmenting, respectively, GTV, CTV1 and CTV, achieving a DSC of 0.94, 0.95 and 0.92; HD95 of 2.07 mm, 1.18 mm and 3.80 mm; ASD of 0.69 mm, 0.39 mm and 1.13 mm and RVE of 5.50%, 3.97% and 9.68%. In the multicenter evaluation, the segmentation performance of our method implemented with the Cross strategy was comparable to that with the Mix strategy, demonstrating that our method had high and stable generalizability across multicenter datasets in automatically segmenting GTV, CTV1 and CTV2. Our proposed method achieved promising results in automatically segmenting gliomas across various datasets, which could improve the quality and efficiency of glioblastoma radiotherapy.

A Comprehensive Deep Learning Framework for Automatic Target Volumes Segmentation in Post-Operative Stereotactic Partial Breast Irradiation (S-PBI)

In S-PBI, accurate delineation of post-surgical tumor bed volume (TBV) and clinical target volume (CTV) are crucial tasks to achieve effective radiotherapy outcomes. However, manual contouring is labor intensive, time consuming, and largely relies on the experience of clinicians. We aimed to propose a deep learning (DL) approach which mimics physicians' contouring practice to accurately segment target volumes in post-operative breast CT images. Our approach incorporated domain knowledge into a 3D U-Net based DL model for breast target volumes (TBV and CTV) delineation. Our TBV segmentation approach was inspired by the marker-guidance procedure in manual delineation, where the visual clues provided by the markers assist physicians in defining TBV. For this purpose, a distance-transformation coupled with a Gaussian filter was adopted to convert markers' locations on the CT images to saliency maps. Subsequently, the CT images and the corresponding saliency maps formed a two-channel input for the segmentation model. For CTV segmentation, TBV was incorporated as an input in addition to the CT images, guiding the model to encode the location-related image features. The architecture allowed the network to emulate the oncologist's manual delineation where CTV is derived from TBV via a margin expansion, followed by correcting the extensions for anatomical barriers of tumor invasion (e.g., skin, chest wall). We retrospectively collected 175 prone CT images from 35 post-operative breast cancer patients who received 5-fraction partial breast irradiation (PBI) regimen on a Co-60 prone based S-PBI unit. The 35 patients were randomly split into 25, 5, and 5 for model training, validation, and testing respectively. We evaluated the performance of the developed DL model on the testing dataset by comparing the predicted volumes with the manually delineated contours (ground truth) using Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (HD95), and average symmetric surface distance (ASD). For TBV segmentation, our model achieved mean (standard deviation) of 0.76 (±2.7), 6.76 (±1.83) mm, and 1.9 (±0.66) mm for DSC, HD95, and ASD respectively. For CTV segmentation, our model achieved 0.94 (±0.02), 2.46 (±0.5) mm, and 0.53 (±0.14) mm for DSC, HD95, and ASD respectively. The proposed auto-segmentation approach generated TBV and CTV masks in ∼11 seconds per CT volume, implying significantly improved efficiency compared to manual contouring. We developed a comprehensive DL framework mimicking clinical contouring practice for auto-segmentation of target volumes in S-PBI. The results demonstrated high levels of agreement between the predicted contours and physicians' manual contours. The approach is promising for improving the efficiency and accuracy of the on-line treatment planning workflow, such as adaptive based S-PBI.

DCE-MRI-Based Atlas for Prostate Bed Recurrence after Radical Prostatectomy: Consistency of CTV Delineation with the Available Contouring Guidelines

To create a magnetic resonance image (MRI)-based atlas of prostate bed recurrence for prostate cancer patients undergoing salvage radiotherapy (sRT) after radical prostatectomy and to evaluate the consistency of contouring guidelines for the prostatic fossa clinical target volume (PF-CTV) delineation. Patients with biochemical recurrence and prostate bed lesions at MRI before sRT were included. A reference patient was identified and the lesion(s) of each patient was mapped by a deformable co-registration among T2 weighted (T2w) images. The structure sets were combined to create a 3D recurrence incidence map, overlapped on T2w reference set, then transferred to the planning CT. The volume and the location of the lesions were extracted. On the reference CT, 5 PF-CTV contouring strategies were implemented: Radiation Therapy Oncology Group (RTOG); European Organization for Research and Treatment of Cancer (EORTC); Francophone Group of Urological Radiotherapy (GFRU); Faculty of Radiation Oncology Genito-urinary Group (FROGG) and Princess Margaret Hospital (PMH). Coverage was assessed after identifying the centroid of each lesion: if the centroid was included in the PF-CTV, the coverage was considered complete (fully covered-FC), otherwise uncovered (UC). The coverage was evaluated for each contouring strategy and compared with the chi-squared test; the strength of the correlation was assessed with the phi coefficient. A total of 112 patients with 124 recurrences were identified. The lesions were located at the vesicourethral anastomosis (VUA), around the bladder neck (BN) and beyond the bladder (RV) in 63, 36 and 25 cases. Regardless the contouring strategy, 89.1% of the lesions were FC. Coverage was higher (p<0.001) for BN (95.6%) over RV (83.2%) and VUA (73.7%) lesions. FC/UC rates were 66.1%/33.9%, 83.9%/16.1%, 89.5%/10.5%, 83.1%/16.9%, 87.1%/12.9% for EORTC, FROGG, GFRU, RTOG and PMH, respectively. At pair-wise comparison, the EORTC definition had lower coverage (p<0.001) and was poorly correlated to the other strategies (phi values ranging 0.385-0.540, p <0.001). Conversely, the correlations between the other definitions were >0.700, the highest between PMH-FROG (0.812, p<0.001). After selecting VUA lesions only (N = 63), the probability of FC was 60.3%, 74.6%, 82.5%, 71.4% and 79.4% for EORTC, FROGG, GFRU, RTOG and PMH, respectively. The EORTC definition performed poorly than GFRU (p = 0.006) and PMH (p = 0.020). When considering lesions above the VUA (N = 61), the performance improved (FC: 72.1%, 93.4%, 96.7%, 95.1% and 95.1% for EORTC, FROGG, GFRU, RTOG and PMH, respectively) but the EORTC was significantly less covering and its correlation with the others was ≤ 0.1. According to the criterium of coverage our data do not support the use of the EORTC definition. Here we provide an MRI-based atlas of prostatic bed recurrence that can optimize sRT contours.