Automatic marker-free target positioning and tracking for image-guided radiotherapy and interventions

Background:Images from onboard electronic portal imaging devices (EPID) contain dose information that can be converted into dose maps. A cycle-consistent generative adversarial network (CycleGAN)-based model was developed for two-dimensional (2D) EPID dosimetry, and the dose characteristics of the model were evaluated carefully.Materials and Methods:All the measurements were done on a linac equipped with an EPID detector. This experiment involved: (1) assessing the dose characteristics of the EPID and (2) using a commercial treatment planning system to calculate dose distributions in a slab phantom, which were taken as the ground truth in CycleGAN-based models for converting the EPID images to 2D dose maps. There were about 780 beams delivered to EPID through a slab phantom. There were two normalization methods (NM): I: based on the highest possible value: 65,535 (16 bit); II: by its own maximum pixel value. To evaluate the model, gamma analyses between the ground truth and the output were performed with in-house software; and the dose linearity of the model was checked carefully. A comparative analysis was conducted to evaluate the outcomes stemming from two distinct NMs applied to the input data.Results:The dose characteristics of the EPID demonstrated exceptional precision. Notably, the beam output factors exhibited considerable variations with the increasing thickness of the phantom. Specifically, when the phantom thickness surpassed 12 cm, the trend lines exhibited a pronounced linearity. Deep learning models efficiently transformed EPID images into planar dose maps, albeit exhibiting dose nonlinearity that could be mitigated by choose the suitable normalization medthods. The mean pass rates of gamma analyses (3 mm, 3%) of data normalized by the way I or II were 85.5%, 97.9%, respectively.Conclusion:EPID is an excellent flat-panel detector that captures images rich in dose information, which can be effectively transformed into precise planar dose maps using CycleGAN-based models. The trained model could be used in the quality assurance of treatment plans.

Introduction: Advanced radiotherapy methods include Intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT) requiring stringent quality assurance (QA) to ensure accurate dose delivery. This study assesses the effectiveness of IMRT and VMAT dosage verification by gamma index analysis and capability index (Cpml), which are statistical process control (SPC) tools, to assess QA process reliability and compare dosimetrical outcomes for head and neck and rectal cancer treatment plans. Materials and Methods: For 28 cases of head and neck (14 IMRT and 14 VMAT) were generated utilizing treatment planning system (TPS) Eclipse, which delivered 6 MV photon beam linear accelerator (Varian, Clinac IX). Dose distributions were measured using an aSi-1000 electronic portal imaging device (EPID) and OCTAVIUS Detector 1500 2D array, compared against TPS calculations using gamma index criteria (3%/3 mm). Control charts and Cpml evaluated QA process stability for 25 cases. Dosimetrical comparisons for rectal cancer assessed planning target volume coverage, organ-at-risk doses, and monitor units. Result: For head and neck the gamma passing rates showed strong agreement, with means of 99.24% ± 0.37 (EPID) and 98.76% ± 0.61 (2D array) for IMRT, and 99.00% ± 0.38 (EPID) and 98.91% ± 0.43 (2D array) for VMAT. For rectal cancer the capability index (Cpml) values of 1.36 (IMRT) and 1.47 (VMAT) indicated optimal QA performance. VMAT plans for rectal cancer reduced MUs and treatment time by approximately 50% while achieving comparable PTV coverage and OAR sparing. Conclusion: VMAT demonstrates efficiency advantages over IMRT, with robust QA processes, supported by SPC and capability index assessments, enhancing treatment precision and patient outcomes.

Objective To determine the planning target volume margins of head and neck cancers treated by image guided radiotherapy (IGRT).Methods 464 sets cone beam computed tomography (CBCT) images before setup correction and 126 sets CBCT images after correction were obtained from 51 head and neck cancer patients treated by IGRT in our department.The systematic and random errors were evaluated by either online or offline correction through registering the CBCT images to the planning CT.The data was divided into 3 groups according to the online correction times.Results The isocenter shift were 0.37 mm±2.37 mm, -0.43 mm±2.30 mm and 0.47 mm±2.65 mm in right-left (RL), anterior-posterior (AP) and superior-inferior (SI) directions respectively before correction, and it reduced to 0.08 mm±0.68 mm, -0.03 mm±0.74 mm and 0.03 mm±0.80 mm when evaluated by 126 sets corrected CBCT images.The planning target volume (PTV) margin from clinical target volume (CTV) before correction were:6.41 mm,6.15 mm and 7.10 mm based on two parameter model, and it reduced to 1.78 mm,1.80 mm and 1.97 mm after correction.The PTV margins were 3.8 mm,3.8 mm,4.0 mm;4.0 mm,4.0 mm,5.0 mm and 5.4 mm,5.2 mm,6.1 mm in RL, AP and SI respectively when online-correction times were more than 15 times, 11-15 times,5-10 times.Conclusions CBCT-based on online correction reduce the PTV margin for head and neck cancers treated by IGRT and ensure more precise dose delivery and less normal tissue complications. Key words: Intensity modulated radiotherapy; Image guided radiotherapy; Setup error; Treatment margin

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. Among the research challenges on a global scale, the investigation of deep transfer learning (DTL) in the field of electricity load forecasting represents a fundamental effort that can inform artificial intelligence applications in general. In this paper, a comprehensive study is reported regarding day-ahead electricity load forecasting. For this purpose, three sequence-to-sequence (Seq2seq) deep learning (DL) models are used, namely the multilayer perceptron (MLP), the convolutional neural network (CNN) and the ensemble learning model (ELM), which consists of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL, emphasizing the way the datasets are trained and fine-tuned for higher forecasting accuracy. In order to implement the forecasting strategies using deep learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1–24 h) for each month of the year for the Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. Four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only the Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for the Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) involves the training of the DL models using the Lesvos dataset, and the testing period is performed directly on the Chios dataset without fine-tuning. The fourth strategy is a multi-task deep learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. The results obtained demonstrate that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, DTL Case 1 and 2 outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it is concluded that the MLP model presents the best monthly forecasts with MAPE values of 6.24% and 6.01% for the first two cases, the CNN model presents the best monthly forecasts with MAPE values of 5.57% and 5.60%, respectively, and the ELM model achieves the best monthly forecasts with MAPE values of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.

PurposeTo conduct a head-to-head comparison between deep learning (DL) and radiomics models across institutions for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC) and to investigate the model robustness and generalizability through rigorous internal and external validation.MethodsThis retrospective study included 2304 preoperative images of 576 HCC lesions from two centers, with MVI status determined by postoperative histopathology. We developed DL and radiomics models for predicting the presence of MVI using B-mode ultrasound, contrast-enhanced ultrasound (CEUS) at the arterial, portal, and delayed phases, and a combined modality (B + CEUS). For radiomics, we constructed models with enlarged vs. original regions of interest (ROIs). A cross-validation approach was performed by training models on one center’s dataset and validating the other, and vice versa. This allowed assessment of the validity of different ultrasound modalities and the cross-center robustness of the models. The optimal model combined with alpha-fetoprotein (AFP) was also validated. The head-to-head comparison was based on the area under the receiver operating characteristic curve (AUC).ResultsThirteen DL models and 25 radiomics models using different ultrasound modalities were constructed and compared. B + CEUS was the optimal modality for both DL and radiomics models. The DL model achieved AUCs of 0.802–0.818 internally and 0.667–0.688 externally across the two centers, whereas radiomics achieved AUCs of 0.749–0.869 internally and 0.646–0.697 externally. The radiomics models showed overall improvement with enlarged ROIs (P < 0.05 for both CEUS and B + CEUS modalities). The DL models showed good cross-institutional robustness (P > 0.05 for all modalities, 1.6–2.1% differences in AUC for the optimal modality), whereas the radiomics models had relatively limited robustness across the two centers (12% drop-off in AUC for the optimal modality). Adding AFP improved the DL models (P < 0.05 externally) and well maintained the robustness, but did not benefit the radiomics model (P > 0.05).ConclusionCross-institutional validation indicated that DL demonstrated better robustness than radiomics for preoperative MVI prediction in patients with HCC, representing a promising solution to non-standardized ultrasound examination procedures.

4D-CBCT reconstruction using MV portal imaging during volumetric modulated arc therapy

Weak correlation between gamma passing rates and dose differences in target volumes and organs at risk (OARs) has been reported in several studies. Evaluation on the differences between planned dose-volume histogram (DVH) and reconstructed DVH from measurement was adopted and incorporated into patient-specific quality assurance (PSQA). However, it is difficult to develop a methodology allowing the evaluation of errors on DVHs accurately and quickly. To develop a DVH-based pretreatment PSQA for volumetric modulated arc therapy (VMAT) with combined deep learning (DL) and machine learning models to overcome the limitation of conventional gamma index (GI) and improve the efficiency of DVH-based PSQA. A DL model with a three-dimensional squeeze-and-excitation residual blocks incorporated into a modified U-net was developed to predict the measured PSQA DVHs of 208 head-and-neck (H&N) cancer patients underwent VMAT between 2018 and 2021 from two hospitals, in which 162 cases was randomly selected for training, 18 for validation, and 28 for testing. After evaluating the differences between treatment planning system (TPS) and PSQA DVHs predicted by DL model with multiple metrics, a pass or fail (PoF) classification model was developed using XGBoost algorithm. Evaluation of domain experts on dose errors between TPS and reconstructed PSQA DVHs was taken as ground truth for PoF classification model training. The prediction model was able to achieve a good agreement between predicted, measured, and TPS doses. Quantitative evaluation demonstrated no significant difference between predicted PSQA dose and measured dose for target and OARs, except for Dmean of PTV6900 (p=0.001), D50 of PTV6000 (p=0.014), D2 of PTV5400 (p=0.009), D50 of left parotid (p=0.015), and Dmax of left inner ear (p=0.007). The XGBoost model achieved an area under curves, accuracy, sensitivity, and specificity of 0.89 versus 0.88, 0.89 versus 0.86, 0. 71 versus 0.71, and 0.95 versus 0.91 with measured and predicted PSQA doses, respectively. The agreement between domain experts and the classification model was 86% for 28 test cases. The successful prediction of PSQA doses and classification of PoF for H&N VMAT PSQA indicating that this DVH-based PSQA method is promising to overcome the limitations of GI and to improve the efficiency and accuracy of VMAT delivery.

Electronic portal imaging devices (EPIDs) have been widely utilized for patient-specific quality assurance (PSQA) and their use for transit dosimetry applications is emerging. Yet there are no specific guidelines on the potential uses, limitations, and correct utilization of EPIDs for these purposes. The American Association of Physicists in Medicine (AAPM) Task Group 307 (TG-307) provides a comprehensive review of the physics, modeling, algorithms and clinical experience with EPID-based pre-treatment and transit dosimetry techniques. This review also includes the limitations and challenges in the clinical implementation of EPIDs, including recommendations for commissioning, calibration and validation, routine QA, tolerance levels for gamma analysis and risk-based analysis. Characteristics of the currently available EPID systems and EPID-based PSQA techniques are reviewed. The details of the physics, modeling, and algorithms for both pre-treatment and transit dosimetry methods are discussed, including clinical experience with different EPID dosimetry systems. Commissioning, calibration, and validation, tolerance levels and recommended tests, are reviewed, and analyzed. Risk-based analysis for EPID dosimetry is also addressed. Clinical experience, commissioning methods and tolerances for EPID-based PSQA system are described for pre-treatment and transit dosimetry applications. The sensitivity, specificity, and clinical results for EPID dosimetry techniques are presented as well as examples of patient-related and machine-related error detection by these dosimetry solutions. Limitations and challenges in clinical implementation of EPIDs for dosimetric purposes are discussed and acceptance and rejection criteria are outlined. Potential causes of and evaluations of pre-treatment and transit dosimetry failures are discussed. Guidelines and recommendations developed in this report are based on the extensive published data on EPID QA along with the clinical experience of the TG-307 members. TG-307 focused on the commercially available EPID-based dosimetric tools and provides guidance for medical physicists in the clinical implementation of EPID-based patient-specific pre-treatment and transit dosimetry QA solutions including intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT) treatments.

An overview is provided of the use of amorphous silicon electronic portal imaging devices (EPIDs) for dosimetric purposes in radiation therapy, focusing on 3D patient dose estimation. EPIDs were originally developed to provide on-treatment radiological imaging to assist with patient setup, but there has also been a natural interest in using them as dosimeters since they use the megavoltage therapy beam to form images. The current generation of clinically available EPID technology, amorphous-silicon (a-Si) flat panel imagers, possess many characteristics that make them much better suited to dosimetric applications than earlier EPID technologies. Features such as linearity with dose/dose rate, high spatial resolution, realtime capability, minimal optical glare, and digital operation combine with the convenience of a compact, retractable detector system directly mounted on the linear accelerator to provide a system that is well-suited to dosimetric applications. This review will discuss clinically available a-Si EPID systems, highlighting dosimetric characteristics and remaining limitations. Methods for using EPIDs in dosimetry applications will be discussed. Dosimetric applications using a-Si EPIDs to estimate three-dimensional dose in the patient during treatment will be overviewed. Clinics throughout the world are implementing increasingly complex treatments such as dynamic intensity modulated radiation therapy and volumetric modulated arc therapy, as well as specialized treatment techniques using large doses per fraction and short treatment courses (ie. hypofractionation and stereotactic radiosurgery). These factors drive the continued strong interest in using EPIDs as dosimeters for patient treatment verification.

Comparison of Workflow Efficiency of VMAT Craniospinal Irradiation Between Adult and Pediatric Settings

Hypofractionation for localized prostate cancer treatment is rapidly spreading in the medical community and it is supported by radiobiological evidences (lower α/β ratio compared with surrounding tissues). Stereotactic body radiation therapy (SBRT) is a technique to administer high doses with great precision, which is commonly performed with CyberKnife (CK) in prostate cancer treatment. Since the CyberKnife (CK) is not available at all radiotherapy center, alternative SBRT techniques are available such as Volumetric Modulated Arc Therapy (VMAT) and Helical Tomotherapy (HT). The aim of the present study was to compare the dosimetric differences between the CK, VMAT, and HT plans for localized prostate cancer treatment.Seventeenpatients have been recruited and replanned using VMAT and HT to this purpose: they received the treatment using the CK with a prescription of 36.25 Gy in 5 fractions; bladder, rectum and penis bulb were considered as organs at risk (OAR). In order to compare the techniques, we considered DVHs, PTV coverage, Conformity Index and new Conformity Index, Homogeneity Index, beam-on time and OARs received dose.The 3 treatments methods showed a comparable coverage of the lesion (PTV 95%: 99.8 ± 0.4% CK; 98.5 ± 0.8% VMAT; 99.4 ± 0.5% HT. P < .05) and good sparing of OARs. Nevertheless, the beam-on time showed a significant difference (37 ± 9 m CK; 7.1 ± 0.3 m VMAT; 17 ± 2 m HT. P < .05).Our results showed that, although CK is the best SBRT technique for prostate cancer treatment, in case this technology is not available, it can be replaced by a similar treatment delivered by VMAT technique. VMAT can be administrated only if it has an appropriate Image Guided Radiation Therapy (IGRT) tracking system.

Purpose: The aim of this study is to quantify dosimetric effects resulting from variation in pancreatic tumor position assessed by bony anatomy and implanted fiducial markers Methods: Twelve pancreatic cancer patients were retrospectively analyzed for this study. All patients received modulated arc therapy (VMAT) treatment using fiducial-based Image Guided Radiation Therapy (IGRT) to the intact pancreas. Using daily orthogonal kV and/or Cone beam CT images, the shift needed to co-register the daily pre-treatment images to reference CT from fiducial to bone (Fid-Bone) were recorded as Left-Right (LR), Anterior-Posterior (AP) and Superior-Inferior (SI). The original VMAT plan iso-center was shifted based on KV bone matching positions at 5 evenly spaced fractions. Dose coverage of the planning target volumes (PTVs) (V100%), mean dose to liver, kidney and stomach/duodenum were assessed in the modified plans. Results: A total of 306 fractions were analyzed. The absolute fiducial-bone positional shifts were greatest in the SI direction, (AP = 2.7 ± 3.0, LR = 2.8 ± 2.8, and SI 6.3 ± 7.9 mm, mean ± SD). The V100% was significantly reduced by 13.5%, (Fid-Bone = 95.3 ± 2.0 vs. 82.3 ± 11.8%, p=0.02). This varied widely among patients (Fid-Bone V100% Range = 2–60%), where 33% ofmore » patients had a reduction in V100% of more than 10%. The impact on OARs was greatest to the liver (Fid-Bone= 14.6 vs. 16.1 Gy, 10%), and stomach, (Fid-Bone = 23.9 vx. 25.5 Gy, 7%), however was not statistically significant (p=0.10 both). Conclusion: Compared to matching by fiducial markers, matching by bony anatomy would have substantially reduced the PTV coverage by 13.5%. This reinforces the importance of online position verification based on fiducial markers. Hence, implantation of fiducial markers is strongly recommended for pancreatic cancer patients undergoing intensity modulated radiation therapy treatments.« less

Prostate position relative to pelvic bony anatomy based on intraprostatic gold markers and electronic portal imaging

Intra- and Interfractionation Reproducibility of Lung Tumor Position using Active Breathing Control (ABC) and Cone-Beam CT (CBCT) Based Image-Guided Radiotherapy (IGRT)

COVID-19 is highly infectious and causes acute respiratory disease. Machine learning (ML) and deep learning (DL) models are vital in detecting disease from computerized chest tomography (CT) scans. The DL models outperformed the ML models. For COVID-19 detection from CT scan images, DL models are used as end-to-end models. Thus, the performance of the model is evaluated for the quality of the extracted feature and classification accuracy. There are four contributions included in this work. First, this research is motivated by studying the quality of the extracted feature from the DL by feeding these extracted to an ML model. In other words, we proposed comparing the end-to-end DL model performance against the approach of using DL for feature extraction and ML for the classification of COVID-19 CT scan images. Second, we proposed studying the effect of fusing extracted features from image descriptors, e.g., Scale-Invariant Feature Transform (SIFT), with extracted features from DL models. Third, we proposed a new Convolutional Neural Network (CNN) to be trained from scratch and then compared to the deep transfer learning on the same classification problem. Finally, we studied the performance gap between classic ML models against ensemble learning models. The proposed framework is evaluated using a CT dataset, where the obtained results are evaluated using five different metrics The obtained results revealed that using the proposed CNN model is better than using the well-known DL model for the purpose of feature extraction. Moreover, using a DL model for feature extraction and an ML model for the classification task achieved better results in comparison to using an end-to-end DL model for detecting COVID-19 CT scan images. Of note, the accuracy rate of the former method improved by using ensemble learning models instead of the classic ML models. The proposed method achieved the best accuracy rate of 99.39%.