1
|
McDonald BA, Cardenas CE, O'Connell N, Ahmed S, Naser MA, Wahid KA, Xu J, Thill D, Zuhour RJ, Mesko S, Augustyn A, Buszek SM, Grant S, Chapman BV, Bagley AF, He R, Mohamed ASR, Christodouleas J, Brock KK, Fuller CD. Investigation of autosegmentation techniques on T2-weighted MRI for off-line dose reconstruction in MR-linac workflow for head and neck cancers. Med Phys 2024; 51:278-291. [PMID: 37475466 PMCID: PMC10799175 DOI: 10.1002/mp.16582] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/03/2022] [Revised: 06/01/2023] [Accepted: 06/12/2023] [Indexed: 07/22/2023] Open
Abstract
BACKGROUND In order to accurately accumulate delivered dose for head and neck cancer patients treated with the Adapt to Position workflow on the 1.5T magnetic resonance imaging (MRI)-linear accelerator (MR-linac), the low-resolution T2-weighted MRIs used for daily setup must be segmented to enable reconstruction of the delivered dose at each fraction. PURPOSE In this pilot study, we evaluate various autosegmentation methods for head and neck organs at risk (OARs) on on-board setup MRIs from the MR-linac for off-line reconstruction of delivered dose. METHODS Seven OARs (parotid glands, submandibular glands, mandible, spinal cord, and brainstem) were contoured on 43 images by seven observers each. Ground truth contours were generated using a simultaneous truth and performance level estimation (STAPLE) algorithm. Twenty total autosegmentation methods were evaluated in ADMIRE: 1-9) atlas-based autosegmentation using a population atlas library (PAL) of 5/10/15 patients with STAPLE, patch fusion (PF), random forest (RF) for label fusion; 10-19) autosegmentation using images from a patient's 1-4 prior fractions (individualized patient prior [IPP]) using STAPLE/PF/RF; 20) deep learning (DL) (3D ResUNet trained on 43 ground truth structure sets plus 45 contoured by one observer). Execution time was measured for each method. Autosegmented structures were compared to ground truth structures using the Dice similarity coefficient, mean surface distance (MSD), Hausdorff distance (HD), and Jaccard index (JI). For each metric and OAR, performance was compared to the inter-observer variability using Dunn's test with control. Methods were compared pairwise using the Steel-Dwass test for each metric pooled across all OARs. Further dosimetric analysis was performed on three high-performing autosegmentation methods (DL, IPP with RF and 4 fractions [IPP_RF_4], IPP with 1 fraction [IPP_1]), and one low-performing (PAL with STAPLE and 5 atlases [PAL_ST_5]). For five patients, delivered doses from clinical plans were recalculated on setup images with ground truth and autosegmented structure sets. Differences in maximum and mean dose to each structure between the ground truth and autosegmented structures were calculated and correlated with geometric metrics. RESULTS DL and IPP methods performed best overall, all significantly outperforming inter-observer variability and with no significant difference between methods in pairwise comparison. PAL methods performed worst overall; most were not significantly different from the inter-observer variability or from each other. DL was the fastest method (33 s per case) and PAL methods the slowest (3.7-13.8 min per case). Execution time increased with a number of prior fractions/atlases for IPP and PAL. For DL, IPP_1, and IPP_RF_4, the majority (95%) of dose differences were within ± 250 cGy from ground truth, but outlier differences up to 785 cGy occurred. Dose differences were much higher for PAL_ST_5, with outlier differences up to 1920 cGy. Dose differences showed weak but significant correlations with all geometric metrics (R2 between 0.030 and 0.314). CONCLUSIONS The autosegmentation methods offering the best combination of performance and execution time are DL and IPP_1. Dose reconstruction on on-board T2-weighted MRIs is feasible with autosegmented structures with minimal dosimetric variation from ground truth, but contours should be visually inspected prior to dose reconstruction in an end-to-end dose accumulation workflow.
Collapse
Affiliation(s)
- Brigid A McDonald
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Carlos E Cardenas
- Department of Radiation Oncology, The University of Alabama at Birmingham, Birmingham, Alabama, USA
| | | | - Sara Ahmed
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Mohamed A Naser
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Kareem A Wahid
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | | | | | - Raed J Zuhour
- Department of Radiation Oncology, The University of Texas Medical Branch, Galveston, Texas, USA
| | - Shane Mesko
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Alexander Augustyn
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Samantha M Buszek
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Stephen Grant
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Bhavana V Chapman
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Alexander F Bagley
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Renjie He
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Abdallah S R Mohamed
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | | | - Kristy K Brock
- Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
- Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Clifton D Fuller
- Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| |
Collapse
|
2
|
Augurio A, Macchia G, Caravatta L, Lucarelli M, Di Gugliemo F, Vinciguerra A, Seccia B, De Sanctis V, Autorino R, Delle Curti C, Meregalli S, Perrucci E, Raspanti D, Cerrotta A. Contouring of emerging organs-at-risk (OARS) of the female pelvis and interobserver variability: A study by the Italian association of radiotherapy and clinical oncology (AIRO). Clin Transl Radiat Oncol 2023; 43:100688. [PMID: 37854671 PMCID: PMC10579954 DOI: 10.1016/j.ctro.2023.100688] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/30/2023] [Revised: 09/30/2023] [Accepted: 10/05/2023] [Indexed: 10/20/2023] Open
Abstract
Purpose To provide straightforward instructions for daily practice in delineating emerging organs-at-risk (OARs) of the female pelvis and to discuss the interobserver variability in a two-step multicenter study. Methods and materials A contouring atlas with anatomical boundaries for each emerging OAR was realized by radiation oncologists and radiologists who are experts in pelvic imaging, as per their knowledge and clinical practice. These contours were identified as quality benchmarks for the analysis subsequently carried out. Radiation oncologists not involved in setting the custom-built contouring atlas and interested in the treatment of gynecological cancer were invited to participate in this 2-step trial. In the first step all participants were supplied with a selected clinical case of locally advanced cervical cancer and had to identify emerging OARs (Levator ani muscle; Puborectalis muscle; Internal anal sphincter; External anal sphincter; Bladder base and trigone; Bladder neck; Iliac Bone Marrow; Lower Pelvis Bone Marrow; Lumbosacral Bone Marrow) based on their own personal knowledge of pelvic anatomy and experience. The suggested OARs and the contouring process were then presented at a subsequent webinar meeting with a contouring laboratory. Finally, in the second step, after the webinar meeting, each participant who had joined the study but was not involved in setting the benchmark received the custom-built contouring atlas with anatomical boundaries and was requested to delineate again the OARs using the tool provided. The Dice Similarity Coefficient (DSC) and the Jaccard Similarity Coefficient (JSC) were used to evaluate the spatial overlap accuracy of the different volume delineations and compared with the benchmark; the Hausdorff distance (HD) and the mean distance to agreement (MDA) to explore the distance between contours. All the results were reported as sample mean and standard deviation (SD). Results Fifteen radiation oncologists from different Institutions joined the study. The participants had a high agreement degree for pelvic bones sub-structures delineation according to DICE (IBM: 0.9 ± 0.02; LPBM: 0.91 ± 0.01). A moderate degree according to DICE was showed for ovaries (Right: 0.61 ± 0.16, Left: 0.72 ± 0.05), vagina (0.575 ± 0.13), bladder sub-structures (0.515 ± 0.08) and EAS (0.605 ± 0.05), whereas a low degree for the other sub-structures of the anal-rectal sphincter complex (LAM: 0.345 ± 0.07, PRM: 0.41 ± 0.10, and IAS: 0.4 ± 0.07). Conclusion This study found a moderate to low level of agreement in the delineation of the female pelvis emerging OARs, with a high degree of variability among observers. The development of delineation tools should be encouraged to improve the routine contouring of these OARs and increase the quality and consistency of radiotherapy planning.
Collapse
Affiliation(s)
- A. Augurio
- Department of Radiation Oncology, SS. Annunziata Hospital, Via Dei Vestini, 66100 Chieti, Italy
| | - G. Macchia
- Radiation Oncology Unit, Gemelli Molise Hospital, Università Cattolica del Sacro Cuore, Largo Agostino Gemelli, 1, 86100 Campobasso, Italy
| | - L. Caravatta
- Department of Radiation Oncology, SS. Annunziata Hospital, Via Dei Vestini, 66100 Chieti, Italy
| | - M. Lucarelli
- Department od Radiotion Oncology, SS Annunziata Hospital, "G. D'Annunzio" University, Via dei Vestini, 66100 Chieti, Italy
| | - F. Di Gugliemo
- Department od Radiotion Oncology, SS Annunziata Hospital, "G. D'Annunzio" University, Via dei Vestini, 66100 Chieti, Italy
| | - A. Vinciguerra
- Department of Radiation Oncology, SS. Annunziata Hospital, Via Dei Vestini, 66100 Chieti, Italy
| | - B. Seccia
- Department of Neuroscience, Imaging and Clinical Sciences, “G. D’Annunzio” University, Via Luigi Polacchi 11, 66100 Chieti, Italy
| | - V. De Sanctis
- Radiotherapy Oncology, Department of Medicine and Surgery and Translational Medicine, Sapienza University of Rome, S. Andrea Hospital, Via di Grottarossa 1035, 00189 Rome, Italy
| | - R. Autorino
- Oncological Radiotherapy Unit, Department of Diagnostic Imaging, Oncological Radiotherapy and Hematology, Fondazione Policlinico Universitario A. Gemelli IRCCS, Via Giuseppe Moscati, 31, 00168 Rome, Italy
| | - C. Delle Curti
- Radioterapia Oncologica, Fondazione IRCS, Istituto Nazionale dei Tumori di Milano, Via Giacomo Venezian, 1, 20133 Milano, Italy
| | - S. Meregalli
- Radiotherapy Unit, Azienda Ospedaliera San Gerardo, Via G. B. Pergolesi, 33, 20900 Monza, Italy
| | - E. Perrucci
- Radiation Oncology Section, Perugia General Hospital, Piazzale Giorgio Menghini, 3, 06129 Perugia, Italy
| | - D. Raspanti
- Temasinergie S.p.A., Via Marcello Malpighi 120, Faenza, Italy
| | - A. Cerrotta
- Radioterapia Oncologica, Fondazione IRCS, Istituto Nazionale dei Tumori di Milano, Via Giacomo Venezian, 1, 20133 Milano, Italy
| |
Collapse
|
3
|
Washington B, Cheek D, Fabian D, Kudrimoti M, Pokhrel D, Wang C, Thayer-Freeman C, Luo W. Effects of Interfraction Dose Variations of Target and Organs at Risk on Clinical Outcomes in High Dose Rate Brachytherapy for Cervical Cancer. Cancers (Basel) 2023; 15:4862. [PMID: 37835556 PMCID: PMC10571581 DOI: 10.3390/cancers15194862] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/15/2023] [Revised: 09/22/2023] [Accepted: 09/26/2023] [Indexed: 10/15/2023] Open
Abstract
Meeting dose prescription is critical to control tumors in radiation therapy. Interfraction dose variations (IDVs) from the prescribed dose in high dose rate brachytherapy (HDR) would cause the target dose to deviate from the prescription but their clinical effect has not been widely discussed in the literature. Our previous study found that IDVs followed a left-skewed distribution. The clinical effect of the IDVs in 100 cervical cancer HDR patients will be addressed in this paper. An in-house Monte Carlo (MC) program was used to simulate clinical outcomes by convolving published tumor dose response curves with IDV distributions. The optimal dose and probability of risk-free local control (RFLC) were calculated using the utility model. The IDVs were well-fitted by the left-skewed Beta distribution, which caused a 3.99% decrease in local control probability and a 1.80% increase in treatment failure. Utility with respect to IDV uncertainty increased the RFLC probability by 6.70% and predicted an optimal dose range of 83 Gy-91 Gy EQD2. It was also found that a 10 Gy dose escalation would not affect toxicity. In conclusion, HRCTV IDV uncertainty reduced LC probabilities and increased treatment failure rates. A dose escalation may help mitigate such effects.
Collapse
Affiliation(s)
- Brien Washington
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Dennis Cheek
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Denise Fabian
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Mahesh Kudrimoti
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Damodar Pokhrel
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Chi Wang
- Department of Internal Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA;
| | - Cameron Thayer-Freeman
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| | - Wei Luo
- Department of Radiation Medicine, University of Kentucky, 800 Rose Street, Lexington, KY 40536, USA; (B.W.); (D.C.); (D.F.); (M.K.); (D.P.); (C.T.-F.)
| |
Collapse
|
4
|
Guzene L, Beddok A, Nioche C, Modzelewski R, Loiseau C, Salleron J, Thariat J. Assessing Interobserver Variability in the Delineation of Structures in Radiation Oncology: A Systematic Review. Int J Radiat Oncol Biol Phys 2023; 115:1047-1060. [PMID: 36423741 DOI: 10.1016/j.ijrobp.2022.11.021] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/13/2022] [Revised: 11/04/2022] [Accepted: 11/09/2022] [Indexed: 11/23/2022]
Abstract
PURPOSE The delineation of target volumes and organs at risk is the main source of uncertainty in radiation therapy. Numerous interobserver variability (IOV) studies have been conducted, often with unclear methodology and nonstandardized reporting. We aimed to identify the parameters chosen in conducting delineation IOV studies and assess their performances and limits. METHODS AND MATERIALS We conducted a systematic literature review to highlight major points of heterogeneity and missing data in IOV studies published between 2018 and 2021. For the main used metrics, we did in silico analyses to assess their limits in specific clinical situations. RESULTS All disease sites were represented in the 66 studies examined. Organs at risk were studied independently of tumor site in 29% of reviewed IOV studies. In 65% of studies, statistical analyses were performed. No gold standard (GS; ie, reference) was defined in 36% of studies. A single expert was considered as the GS in 21% of studies, without testing intraobserver variability. All studies reported both absolute and relative indices, including the Dice similarity coefficient (DSC) in 68% and the Hausdorff distance (HD) in 42%. Limitations were shown in silico for small structures when using the DSC and dependence on irregular shapes when using the HD. Variations in DSC values were large between studies, and their thresholds were inconsistent. Most studies (51%) included 1 to 10 cases. The median number of observers or experts was 7 (range, 2-35). The intraclass correlation coefficient was reported in only 9% of cases. Investigating the feasibility of studying IOV in delineation, a minimum of 8 observers with 3 cases, or 11 observers with 2 cases, was required to demonstrate moderate reproducibility. CONCLUSIONS Implementation of future IOV studies would benefit from a more standardized methodology: clear definitions of the gold standard and metrics and a justification of the tradeoffs made in the choice of the number of observers and number of delineated cases should be provided.
Collapse
Affiliation(s)
- Leslie Guzene
- Department of Radiation Oncology, University Hospital of Amiens, Amiens, France
| | - Arnaud Beddok
- Department of Radiation Oncology, Institut Curie, Paris/Saint-Cloud/Orsay, France; Laboratory of Translational Imaging in Oncology (LITO), InsermUMR, Institut Curie, Orsay, France
| | - Christophe Nioche
- Laboratory of Translational Imaging in Oncology (LITO), InsermUMR, Institut Curie, Orsay, France
| | - Romain Modzelewski
- LITIS - EA4108-Quantif, Normastic, University of Rouen, and Nuclear Medicine Department, Henri Becquerel Center, Rouen, France
| | - Cedric Loiseau
- Department of Radiation Oncology, Centre François Baclesse; ARCHADE Research Community Caen, France; Département de Biostatistiques, Institut de Cancérologie de Lorraine, Vandœuvre-lès-Nancy, France
| | - Julia Salleron
- Département de Biostatistiques, Institut de Cancérologie de Lorraine, Vandœuvre-lès-Nancy, France
| | - Juliette Thariat
- Department of Radiation Oncology, Centre François Baclesse; ARCHADE Research Community Caen, France; Laboratoire de Physique Corpusculaire, Caen, France; Unicaen-Université de Normandie, Caen, France.
| |
Collapse
|
5
|
Washington B, Randall M, Fabian D, Cheek D, Wang C, Luo W. Statistical Analysis of Interfraction Dose Variations of High-Risk Clinical Target Volume and Organs at Risk for Cervical Cancer High-Dose-Rate Brachytherapy. Adv Radiat Oncol 2022; 7:101019. [PMID: 36110265 PMCID: PMC9468354 DOI: 10.1016/j.adro.2022.101019] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/02/2022] [Accepted: 06/29/2022] [Indexed: 11/23/2022] Open
Abstract
Purpose High-dose-rate (HDR) brachytherapy for cervical cancer treatment includes significant uncertainties. The aim of this study was to quantify the interfraction dosimetric variation (IDV) of the high-risk clinical target volume (HRCTV) from the prescribed dose and the corresponding effect on organ-at-risk (OAR) dose based on a comprehensive statistical analysis. Methods and Materials Fifty patients with cervical cancer treated with high-dose-rate intracavity brachytherapy from October 2019 to December 2020 were retrospectively analyzed. The OARs of interest were the rectum, bladder, sigmoid, and bowel. The dosimetric parameters evaluated for all patients was the dose absorbed by 90% of the HRCTV ( D 90 ) and the dose absorbed by 0.1 ( D 0.1 c c ) and 2 cm3 ( D 2 c c ) of each respective OAR. The HRCTV variations were from the prescribed dose and the OAR variations were from the corresponding tolerance dose. Distribution fitting of the HRCTV variations was determined to quantify the IDV. Comparative statistics of the HRCTV variations with the OAR variations were conducted to determine correlations. Results The mean HRCTV variation from the prescribed dose was -2.53% ± 8.74%. The HRCTV variations and OAR variations showed moderate to weak linear correlations despite the variations being relative to each other, with the bladder D 2 c c having the strongest correlation. There was a 30.0% (±2.62%, 95% confidence interval) probability of underdosing the HRCTV (-5% variation from prescription) and a 23.3% (±2.62%, 95% confidence interval) probability of overdosing the HRCTV (+5% variation from prescription). This tendency to underdose the HRCTV was a consequence of HRCTV IDV not being normally distributed. Conclusions HRCTV dosimetric variations and OAR variations were complexly correlated with the bladder D 2 c c having the strongest correlation. HRCTV IDV was best described as a left-skewed distribution that indicates a tendency of underdosing the HRCTV. The clinical significance of such dose variations is expected and will be further investigated.
Collapse
Affiliation(s)
- Brien Washington
- Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky
| | - Marcus Randall
- Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky
| | - Denise Fabian
- Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky
| | - Dennis Cheek
- Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky
| | - Chi Wang
- Department of Internal Medicine, University of Kentucky, Lexington, Kentucky
| | - Wei Luo
- Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky
| |
Collapse
|
6
|
Li Z, Zhu Q, Zhang L, Yang X, Li Z, Fu J. A deep learning-based self-adapting ensemble method for segmentation in gynecological brachytherapy. Radiat Oncol 2022; 17:152. [PMID: 36064571 PMCID: PMC9446699 DOI: 10.1186/s13014-022-02121-3] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/24/2022] [Accepted: 08/29/2022] [Indexed: 11/10/2022] Open
Abstract
Purpose Fast and accurate outlining of the organs at risk (OARs) and high-risk clinical tumor volume (HRCTV) is especially important in high-dose-rate brachytherapy due to the highly time-intensive online treatment planning process and the high dose gradient around the HRCTV. This study aims to apply a self-configured ensemble method for fast and reproducible auto-segmentation of OARs and HRCTVs in gynecological cancer. Materials and methods We applied nnU-Net (no new U-Net), an automatically adapted deep convolutional neural network based on U-Net, to segment the bladder, rectum and HRCTV on CT images in gynecological cancer. In nnU-Net, three architectures, including 2D U-Net, 3D U-Net and 3D-Cascade U-Net, were trained and finally ensembled. 207 cases were randomly chosen for training, and 30 for testing. Quantitative evaluation used well-established image segmentation metrics, including dice similarity coefficient (DSC), 95% Hausdorff distance (HD95%), and average surface distance (ASD). Qualitative analysis of automated segmentation results was performed visually by two radiation oncologists. The dosimetric evaluation was performed by comparing the dose-volume parameters of both predicted segmentation and human contouring. Results nnU-Net obtained high qualitative and quantitative segmentation accuracy on the test dataset and performed better than previously reported methods in bladder and rectum segmentation. In quantitative evaluation, 3D-Cascade achieved the best performance in the bladder (DSC: 0.936 ± 0.051, HD95%: 3.503 ± 1.956, ASD: 0.944 ± 0.503), rectum (DSC: 0.831 ± 0.074, HD95%: 7.579 ± 5.857, ASD: 3.6 ± 3.485), and HRCTV (DSC: 0.836 ± 0.07, HD95%: 7.42 ± 5.023, ASD: 2.094 ± 1.311). According to the qualitative evaluation, over 76% of the test data set had no or minor visually detectable errors in segmentation. Conclusion This work showed nnU-Net’s superiority in segmenting OARs and HRCTV in gynecological brachytherapy cases in our center, among which 3D-Cascade shows the highest accuracy in segmentation across different applicators and patient anatomy. Supplementary Information The online version contains supplementary material available at 10.1186/s13014-022-02121-3.
Collapse
Affiliation(s)
- Zhen Li
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China
| | - Qingyuan Zhu
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China
| | - Lihua Zhang
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China
| | - Xiaojing Yang
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China
| | - Zhaobin Li
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China.
| | - Jie Fu
- Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China.
| |
Collapse
|
7
|
Lindegaard JC, Petric P, Schmid MP, Nesvacil N, Haie-Meder C, Fokdal LU, Sturdza AE, Hoskin P, Mahantshetty U, Segedin B, Bruheim K, Huang F, Rai B, Cooper R, van der Steen-Banasik E, Van Limbergen E, Pieters BR, Tan LT, Nout RA, De Leeuw AAC, Kirchheiner K, Spampinato S, Jürgenliemk-Schulz I, Tanderup K, Kirisits C, Pötter R. Prognostic Implications of Uterine Cervical Cancer Regression During Chemoradiation Evaluated by the T-Score in the Multicenter EMBRACE I Study. Int J Radiat Oncol Biol Phys 2022; 113:379-389. [PMID: 35157992 DOI: 10.1016/j.ijrobp.2022.02.005] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/07/2021] [Revised: 01/31/2022] [Accepted: 02/03/2022] [Indexed: 12/11/2022]
Abstract
PURPOSE A simple scoring system (T-score, TS) for integrating findings from clinical examination and magnetic resonance imaging (MRI) of the primary tumor at diagnosis has shown strong prognostic capability for predicting local control and survival in locally advanced cervical cancer treated with chemoradiation and MRI-guided brachytherapy (BT). The aim was to validate the performance of TS using the multicenter EMBRACE I study and to evaluate the prognostic implications of TS regression obtained during initial chemoradiation. METHODS AND MATERIALS EMBRACE I recruited 1416 patients, of whom 1318 were available for TS. Patients were treated with chemoradiation followed by MRI-guided BT. A ranked ordinal scale of 0 to 3 points was used to assess 8 anatomic locations typical for local invasion of cervical cancer. TS was calculated separately at diagnosis (TSD) and at BT (TSBT) by the sum of points obtained from the 8 locations at the 2 occasions. RESULTS Median TSD and TSBT was 5 and 4, respectively. TS regression was observed in 71% and was an explanatory variable for BT technique (intracavitary vs intracavitary/interstitial) and major dose-volume histogram parameters for BT, such as high-risk clinical target (CTVHR), CTVHR D90 (minimal dose to 90% of the target volume), D2cm3 bladder (minimal dose to the most exposed 2 cm3 of the bladder), and D2cm3 rectum. TS regression (TSBT≤5) was associated with improved local control and survival and with less morbidity compared with patients with TSBT remaining high (>5) despite initial chemoradiation. TS regression was significant in multivariate analysis for both local control and survival when analyzed in consort with already established prognostic parameters related to the patient, disease, and treatment. CONCLUSIONS TS was validated in a multicenter setting and proven to be a strong multidisciplinary platform for integration of clinical findings and imaging with the ability to quantitate local tumor regression and its prognostic implications regarding BT technique, dose-volume histogram parameters, local control, survival, and morbidity.
Collapse
Affiliation(s)
| | - Primoz Petric
- Department of Oncology, Aarhus University Hospital, Aarhus, Denmark; Department of Radiation Oncology, University Hospital Zürich, Switzerland
| | - Maximilian Paul Schmid
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| | - Nicole Nesvacil
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| | | | | | - Alina Emiliana Sturdza
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| | - Peter Hoskin
- Mount Vernon Cancer Centre, Northwood, United Kingdom
| | - Umesh Mahantshetty
- Homi Bhabha Cancer Hospital & Research Centre, Visakhapatnam, (A Unit of Tata Memorial Centre, Mumbai), India
| | - Barbara Segedin
- Department of Radiotherapy, Institute of Oncology Ljubljana, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
| | - Kjersti Bruheim
- Department of Oncology, Oslo University Hospital, Oslo, Norway
| | - Fleur Huang
- Department of Oncology, Cross Cancer Institute and University of Alberta, Edmonton, Alberta, Canada
| | - Bhavana Rai
- Department of Radiotherapy and Oncology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | - Rachel Cooper
- St James's University Hospital, Leeds Cancer Centre, Leeds, United Kingdom
| | | | | | - Bradley Rumwell Pieters
- Department of Radiation Oncology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands
| | - Li-Tee Tan
- Department of Oncology, Addenbrooke's Hospital, Cambridge University Hospitals, United Kingdom
| | - Remi A Nout
- Department of Radiation Oncology, Leiden University Medical Center, The Netherlands
| | | | - Kathrin Kirchheiner
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| | - Sofia Spampinato
- Department of Oncology, Aarhus University Hospital, Aarhus, Denmark
| | - Ina Jürgenliemk-Schulz
- Department of Radiation Oncology, University Medical Centre Utrecht, Utrecht, Netherlands
| | - Kari Tanderup
- Department of Oncology, Aarhus University Hospital, Aarhus, Denmark
| | - Christian Kirisits
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| | - Richard Pötter
- Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
| |
Collapse
|
8
|
Sherer MV, Lin D, Elguindi S, Duke S, Tan LT, Cacicedo J, Dahele M, Gillespie EF. Metrics to evaluate the performance of auto-segmentation for radiation treatment planning: A critical review. Radiother Oncol 2021; 160:185-191. [PMID: 33984348 DOI: 10.1016/j.radonc.2021.05.003] [Citation(s) in RCA: 78] [Impact Index Per Article: 26.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2020] [Revised: 05/01/2021] [Accepted: 05/03/2021] [Indexed: 12/18/2022]
Abstract
Advances in artificial intelligence-based methods have led to the development and publication of numerous systems for auto-segmentation in radiotherapy. These systems have the potential to decrease contour variability, which has been associated with poor clinical outcomes and increased efficiency in the treatment planning workflow. However, there are no uniform standards for evaluating auto-segmentation platforms to assess their efficacy at meeting these goals. Here, we review the most frequently used evaluation techniques which include geometric overlap, dosimetric parameters, time spent contouring, and clinical rating scales. These data suggest that many of the most commonly used geometric indices, such as the Dice Similarity Coefficient, are not well correlated with clinically meaningful endpoints. As such, a multi-domain evaluation, including composite geometric and/or dosimetric metrics with physician-reported assessment, is necessary to gauge the clinical readiness of auto-segmentation for radiation treatment planning.
Collapse
Affiliation(s)
- Michael V Sherer
- Department of Radiation Medicine and Applied Sciences, University of California San Diego, La Jolla, United States
| | - Diana Lin
- Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, United States
| | - Sharif Elguindi
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, United States
| | - Simon Duke
- Department of Oncology, Cambridge University Hospitals, United Kingdom
| | - Li-Tee Tan
- Department of Oncology, Cambridge University Hospitals, United Kingdom
| | - Jon Cacicedo
- Department of Radiation Oncology, Cruces University Hospital/BioCruces Health Research Institute, Osakidetza, Barakaldo, Spain
| | - Max Dahele
- Department of Radiation Oncology, Amsterdam University Medical Center, Amsterdam, The Netherlands
| | - Erin F Gillespie
- Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, United States.
| |
Collapse
|
9
|
Education and training for image-guided adaptive brachytherapy for cervix cancer—The (GEC)-ESTRO/EMBRACE perspective. Brachytherapy 2020; 19:827-836. [DOI: 10.1016/j.brachy.2020.06.012] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/07/2020] [Accepted: 06/09/2020] [Indexed: 11/21/2022]
|
10
|
Shaaer A, Paudel M, Davidson M, Semple M, Nicolae A, Mendez LC, Chung H, Loblaw A, Tseng CL, Morton G, Ravi A. Dosimetric evaluation of MRI-to-ultrasound automated image registration algorithms for prostate brachytherapy. Brachytherapy 2020; 19:599-606. [PMID: 32712028 DOI: 10.1016/j.brachy.2020.06.014] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/05/2020] [Revised: 06/13/2020] [Accepted: 06/15/2020] [Indexed: 11/26/2022]
Abstract
PURPOSE Identifying dominant intraprostatic lesions (DILs) on transrectal ultrasound (TRUS) images during prostate high-dose-rate brachytherapy treatment planning remains a significant challenge. Multiparametric MRI (mpMRI) is the tool of choice for DIL identification; however, the geometry of the prostate on mpMRI and on the TRUS may differ significantly, requiring image registration. This study assesses the dosimetric impact attributed to differences in DIL contours generated using commonly available MRI to TRUS automated registration: rigid, semi-rigid, and deformable image registration, respectively. METHODS AND MATERIALS Ten patients, each with mpMRI and TRUS data sets, were included in this study. Five radiation oncologists with expertise in TRUS-based high-dose-rate brachytherapy were asked cognitively to transfer the DIL from the mpMRI images of each patient to the TRUS image. The contours were analyzed for concordance using simultaneous truth and performance level estimation (STAPLE) algorithm. The impact of DIL contour differences due to registration variability was evaluated by comparing the STAPLE-DIL dosimetry from the reference (STAPLE) plan with that from the evaluation plans (manual and automated registration) for each patient. The dosimetric impact of the automatic registration approach was also validated using a margin expansion that normalizes the volume of the autoregistered DILs to the volumes of the STAPLE-DILs. Dose metrics including D90, Dmean, V150, and V200 to the prostate and DIL were reported. For urethra and rectum, D10 and V80 were reported. RESULTS Significant differences in DIL coverage between reference and evaluation plans were found regardless of the algorithm methodology. No statistical difference was reported in STAPLE-DIL dosimetry when manual registration was used. A margin of 1.5 ± 0.8 mm, 1.1 ± 0.8 mm, and 2.5 ± 1.6 mm was required to be added for rigid, semi-rigid, and deformable registration, respectively, to mitigate the difference in STAPLE-DIL coverage between the evaluation and reference plans. CONCLUSION The dosimetric impact of integrating an MRI-delineated DIL into a TRUS-based brachytherapy workflow has been validated in this study. The results show that rigid, semi-rigid, and deformable registration algorithms lead to a significant undercoverage of the DIL D90 and Dmean. A margin of at least 1.5 ± 0.8 mm, 1.1 ± 0.8 mm, and 2.5 ± 1.6 mm is required to be added to the rigid, semi-rigid, and deformable DIL registration to be suitable for DIL-boosting during prostate brachytherapy.
Collapse
Affiliation(s)
- Amani Shaaer
- Department of physics, Ryerson University, Toronto, Ontario, Canada; Biomedical Physics Department, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | - Moti Paudel
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Melanie Davidson
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Mark Semple
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Alexandru Nicolae
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Lucas Castro Mendez
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Hans Chung
- Department of Radiation Oncology, University of Toronto, Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Andrew Loblaw
- Department of Radiation Oncology, University of Toronto, Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Chia-Lin Tseng
- Department of Radiation Oncology, University of Toronto, Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Gerard Morton
- Department of Radiation Oncology, University of Toronto, Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | - Ananth Ravi
- Department of Medical Physics, University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada.
| |
Collapse
|