1
|
Maniscalco A, Mathew E, Parsons D, Visak J, Arbab M, Alluri P, Li X, Wandrey N, Lin MH, Rahimi A, Jiang S, Nguyen D. Multimodal radiotherapy dose prediction using a multi-task deep learning model. Med Phys 2024; 51:3932-3949. [PMID: 38710210 PMCID: PMC11147699 DOI: 10.1002/mp.17115] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/09/2023] [Revised: 03/26/2024] [Accepted: 04/21/2024] [Indexed: 05/08/2024] Open
Abstract
BACKGROUND In radiation therapy (RT), accelerated partial breast irradiation (APBI) has emerged as an increasingly preferred treatment modality over conventional whole breast irradiation due to its targeted dose delivery and shorter course of treatment. APBI can be delivered through various modalities including Cobalt-60-based systems and linear accelerators with C-arm, O-ring, or robotic arm design. Each modality possesses distinct features, such as beam energy or the degrees of freedom in treatment planning, which influence their respective dose distributions. These modality-specific considerations emphasize the need for a quantitative approach in determining the optimal dose delivery modality on a patient-specific basis. However, manually generating treatment plans for each modality across every patient is time-consuming and clinically impractical. PURPOSE We aim to develop an efficient and personalized approach for determining the optimal RT modality for APBI by training predictive models using two different deep learning-based convolutional neural networks. The baseline network performs a single-task (ST), predicting dose for a single modality. Our proposed multi-task (MT) network, which is capable of leveraging shared information among different tasks, can concurrently predict dose distributions for various RT modalities. Utilizing patient-specific input data, such as a patient's computed tomography (CT) scan and treatment protocol dosimetric goals, the MT model predicts patient-specific dose distributions across all trained modalities. These dose distributions provide patients and clinicians quantitative insights, facilitating informed and personalized modality comparison prior to treatment planning. METHODS The dataset, comprising 28 APBI patients and their 92 treatment plans, was partitioned into training, validation, and test subsets. Eight patients were dedicated to the test subset, leaving 68 treatment plans across 20 patients to divide between the training and validation subsets. ST models were trained for each modality, and one MT model was trained to predict doses for all modalities simultaneously. Model performance was evaluated across the test dataset in terms of Mean Absolute Percent Error (MAPE). We conducted statistical analysis of model performance using the two-tailed Wilcoxon signed-rank test. RESULTS Training times for five ST models ranged from 255 to 430 min per modality, totaling 1925 min, while the MT model required 2384 min. MT model prediction required an average of 1.82 s per patient, compared to ST model predictions at 0.93 s per modality. The MT model yielded MAPE of 1.1033 ± 0.3627% as opposed to the collective MAPE of 1.2386 ± 0.3872% from ST models, and the differences were statistically significant (p = 0.0003, 95% confidence interval = [-0.0865, -0.0712]). CONCLUSION Our study highlights the potential benefits of a MT learning framework in predicting RT dose distributions across various modalities without notable compromises. This MT architecture approach offers several advantages, such as flexibility, scalability, and streamlined model management, making it an appealing solution for clinical deployment. With such a MT model, patients can make more informed treatment decisions, physicians gain more quantitative insight for pre-treatment decision-making, and clinics can better optimize resource allocation. With our proposed goal array and MT framework, we aim to expand this work to a site-agnostic dose prediction model, enhancing its generalizability and applicability.
Collapse
Affiliation(s)
- Austen Maniscalco
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Ezek Mathew
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - David Parsons
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Justin Visak
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Mona Arbab
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Prasanna Alluri
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Xingzhe Li
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Narine Wandrey
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Mu-Han Lin
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Asal Rahimi
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Steve Jiang
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| | - Dan Nguyen
- Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
| |
Collapse
|
2
|
Li F, Mail N, Stefania diMayorca M, McCaw TJ, Ozhasoglu C, Lalonde R, Chang J, Huq MS. Single isocenter HyperArc treatment of multiple intracranial metastases: Targeting accuracy. J Appl Clin Med Phys 2024; 25:e14234. [PMID: 38059673 PMCID: PMC10795440 DOI: 10.1002/acm2.14234] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2023] [Revised: 10/12/2023] [Accepted: 10/25/2023] [Indexed: 12/08/2023] Open
Abstract
PURPOSE/OBJECTIVES (A) To examine the alignment accuracy of CBCT guidance for brain metastases with off centered isocenters, (B) to test dose delivery and targeting accuracy for single isocenter treatments with multiple brain metastases. We report the results of the end-to-end test for Truebeam stereotactic radiosurgery (SRS). MATERIALS/METHODS An anthropomorphic CT head phantom was drilled with five MOSFET inserts and two PTW Pinpoint chamber inserts. The phantom was simulated, planned, and delivered. For the purpose of comparing the accuracy of alignment, CBCTs were acquired with the isocenter centered and offset superiorly 8 cm, inferiorly 8 cm, anteriorly 7 cm, posteriorly 7 cm, and right 5 cm. There were six degrees of freedom corrections applied to the plans, as well as intentional rotational and translational errors for dose comparisons. Dose accuracy checks were performed with MOSFET and PTW Pinpoint chamber, and targeting accuracy was assessed with GafChromic films. RESULT (A) Compared to centered CBCT, off-centered CBCT scan showed some alignment errors, with a maximum difference of 0.6-degree pitch and 0.9 mm translation when the phantom was placed 8 cm inferior off center. (B) For the single isocenter plan, measured doses of the five MOSFET were 95%-100% of the planned dose, whereas the multiple isocenter plans were 96%-100%. With intentional setup errors of 1-degree pitch, doses were 97.1%-100.4% compared to the perfect setup. The same was found for the two pinpoint chamber readings with 1-degree rotation and 1 mm translation. (C) Targeting accuracy for targets at the isocenter is 0.67 mm, within the machine specification of 0.75 mm. Targeting accuracy for isocenters 6-12 cm away from the target is in the range 0.67-1.18 mm. CONCLUSION (A) Single isocenter HyperArc treatments for multiple brain metastases are feasible and targeting accuracy is clinically acceptable. (B) The vertex in a cranial scan is very important for proper alignment.
Collapse
Affiliation(s)
- Fang Li
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | - Noor Mail
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | | | - Travis J. McCaw
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | - Cihat Ozhasoglu
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | - Ronald Lalonde
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | - Jina Chang
- Radiation OncologyUPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| | - Mohammed Saiful Huq
- Radiation OncologyUniversity of Pittsburgh School of Medicine and UPMC Hillman Cancer CenterPittsburghPennsylvaniaUSA
| |
Collapse
|
3
|
Hernández KVD, Unterkirhers S, Schneider U. Quality assessment of automatically planned O-Ring linac SBRT plans for pelvic lymph node metastases, finding the optimal minimum target size by comparison with robotic SBRT. J Appl Clin Med Phys 2023; 24:e14143. [PMID: 37738649 PMCID: PMC10691630 DOI: 10.1002/acm2.14143] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2023] [Revised: 08/18/2023] [Accepted: 08/21/2023] [Indexed: 09/24/2023] Open
Abstract
PURPOSE The purpose of this study is to assess the quality of automatic planned O-Ring Halcyon linac SBRT plans for pelvic lymph node metastases and to establish an absolute PTV volume threshold as a plan quality prediction criterion. Compliance of the plans to institutional SBRT plan evaluation criteria and differences in plan quality and treatment delivery times between Halcyon Linac and CyberKnife robotic SBRT were evaluated. METHODS Twenty-one CyberKnife treatment plans were replanned for Halcyon. Prescription doses range was 26-40 Gy in mean three fractions. The mean/median planning target volume was 4.0/3.6 cm3 . Institutional criteria for the plan evaluation were: New Conformity Index (NCI), Conformity Index (CI), Modified Gradient Index (MGI), selectivity index reciprocal (PIV/TVPIV ), and the target coverage by prescription isodose (%PIV). Statistical analysis based on the receiver operating characteristic (ROC) curve was used to determine a plan quality predictor threshold of the PTV volume. Comparative analysis of normal tissue complication probabilities (NTCP) was used to assess the risk of toxicity in healthy tissues. RESULTS Seventy-one percent (n = 15)/95% (n = 20) of Halcyon and 81% (n = 17)/100% (n = 21) of CK plans fulfilled all ideal/tolerance criteria. For PTVs above a found optimal threshold of 2.6 cm3 (71%, n = 15), no statistically significant difference was observed between the CI, NCI, PIV/TVPIV , and MGI indexes of both groups, while the coverage (%PIV) was statistically but not clinically significantly different between cohorts. Significantly shorter delivery times are expected with Halcyon. No significant differences in NTCP were observed. CONCLUSION All but one automatically optimized Halcyon treatment plans demonstrated ideal or acceptable performance. PTV threshold of 2.6 cm3 can be used as decision criteria in clinical settings. The results of our study demonstrated the promising performance of the Halcyon for pelvic SBRT, although plan-specific QA is required to verify machine performance during plan delivery.
Collapse
Affiliation(s)
| | | | - Uwe Schneider
- Science FacultyUniversity of ZürichZürichSwitzerland
- Medical PhysicsRadiotherapy HirslandenZürichSwitzerland
| |
Collapse
|
4
|
Huang Y, Liu Z. Dosimetric performance evaluation of the Halcyon treatment platform for stereotactic radiotherapy: A pooled study. Medicine (Baltimore) 2023; 102:e34933. [PMID: 37682167 PMCID: PMC10489306 DOI: 10.1097/md.0000000000034933] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/13/2023] [Accepted: 08/04/2023] [Indexed: 09/09/2023] Open
Abstract
With the advancement of radiotherapy equipment, stereotactic radiotherapy (SRT) has been increasingly used. Among the many radiotherapy devices, Halcyon shows promising applications. This article reviews the dosimetric performance such as plan quality, plan complexity, and gamma passing rates of SRT plans with Halcyon to determine the effectiveness and safety of Halcyon SRT plans. This article retrieved the last 5 years of PubMed studies on the effectiveness and safety of the Halcyon SRT plans. Two authors independently reviewed the titles and abstracts to decide whether to include the studies. A search was conducted to identify publications relevant to evaluating the dosimetric performance of SRT plans on Halcyon using the key strings Halcyon, stereotactic radiosurgery, SRT, stereotactic body radiotherapy, and stereotactic ablative radiotherapy. A total of 18 eligible publications were retrieved. Compared to SRT plans on the TrueBeam, the Halcyon has advantages in terms of plan quality, plan complexity, and gamma passing rates. The high treatment speed of SRT plans on the Halcyon is impressive, while the results of its plan evaluation are also encouraging. As a result, Halcyon offers a new option for busy radiotherapy units while significantly improving patient comfort in treatment. For more accurate results, additional relevant publications will need to be followed up in subsequent studies.
Collapse
Affiliation(s)
- Yangyang Huang
- Department of Radiotherapy, the Second Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Zongwen Liu
- Department of Radiotherapy, the Second Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| |
Collapse
|