Explicitly encoding the cyclic nature of breathing signal allows for accurate breathing motion prediction in radiotherapy with minimal training data.

Background and purpose: Active breathing motion management in radiotherapy consists of motion monitoring, quantification and mitigation. It is impacted by associated latencies of a few 100 ms. Artificial neural networks can successfully predict breathing motion and eliminate latencies. However, they require usually a large dataset for training. The objective of this work was to demonstrate that explicitly encoding the cyclic nature of the breathing signal into the training data enables significant reduction of training datasets which can be obtained from healthy volunteers. Material and methods: Seventy surface scanner breathing signals from 25 healthy volunteers in anterior-posterior direction were used for training and validation (ratio 4:1) of long short-term memory models. The model performance was compared to a model using decomposition into phase, amplitude and a time-dependent baseline. Testing of the models was performed on 55 independent breathing signals in anterior-posterior direction from surface scanner (35 lung, 20 liver) of 30 patients with a mean breathing amplitude of (5.9 ± 6.7) mm. Results: Using the decomposed breathing signal allowed for a reduction of the absolute root-mean square error (RMSE) from 0.34 mm to 0.12 mm during validation. Testing using patient data yielded an average absolute RMSE of the breathing signal of (0.16 ± 0.11) mm with a prediction horizon of 500 ms. Conclusion: It was demonstrated that a motion prediction model can be trained with less than 100 datasets of healthy volunteers if breathing cycle parameters are considered. Applied to 55 patients, the model predicted breathing motion with a high accuracy.

Realistic 3D printed CT imaging tumor phantoms for validation of image processing algorithms.

Medical imaging phantoms are widely used for validation and verification of imaging systems and algorithms in surgical guidance and radiation oncology procedures. Especially, for the performance evaluation of new algorithms in the field of medical imaging, manufactured phantoms need to replicate specific properties of the human body, e.g., tissue morphology and radiological properties. Additive manufacturing (AM) technology provides an inexpensive opportunity for accurate anatomical replication with customization capabilities. In this study, we proposed a simple and cheap protocol using Fused Deposition Modeling (FDM) technology to manufacture realistic tumor phantoms based on the filament 3D printing technology. Tumor phantoms with both homogenous and heterogeneous radiodensity were fabricated. The radiodensity similarity between the printed tumor models and real tumor data from CT images of lung cancer patients was evaluated. Additionally, it was investigated whether a heterogeneity in the 3D printed tumor phantoms as observed in the tumor patient data had an influence on the validation of image registration algorithms. A radiodensity range between -217 to 226 HUs was achieved for 3D printed phantoms using different filament materials; this range of radiation attenuation is also observed in the human lung tumor tissue. The resulted HU range could serve as a lookup-table for researchers and phantom manufactures to create realistic CT tumor phantoms with the desired range of radiodensities. The 3D printed tumor phantoms also precisely replicated real lung tumor patient data regarding morphology and could also include life-like heterogeneity of the radiodensity inside the tumor models. An influence of the heterogeneity on accuracy and robustness of the image registration algorithms was not found.

A novel bone suppression algorithm in intensity-based 2D/3D image registration for real-time tumor motion monitoring: Development and phantom-based validation.

BACKGROUND: Real-time tumor motion monitoring (TMM) is a crucial process for intra-fractional respiration management in lung cancer radiotherapy. Since the tumor can be partly or fully located behind the ribs, the TMM is challenging. PURPOSE: The aim of this work was to develop a bone suppression (BS) algorithm designed for real-time 2D/3D marker-less TMM to increase the visibility of the tumor when overlapping with bony structures and consequently to improve the accuracy of TMM. METHOD: A BS method was implemented in the in-house developed software for ultrafast intensity-based 2D/3D tumor registration (Fast Image-based Registration [FIRE]). The method operates on both, digitally reconstructed radiograph (DRR) and intra-fractional X-ray images. The bony structures are derived from computed tomography data by thresholding during ray-casting, and the resulting bone DRR is subtracted from intra-fractional X-ray images to obtain a soft-tissue-only image for subsequent tumor registration. The accuracy of TMM utilizing BS was evaluated within a retrospective phantom study with nine different 3D-printed tumor phantoms placed in the in-house developed Advanced Radiation DOSimetry (ARDOS) breathing phantom. A 24 mm craniocaudal tumor motion, including rib eclipses, was simulated, and X-ray images were acquired on the Elekta Versa HD Linac in the lateral and posterior-anterior directions. An error assessment for BS images was evaluated with respect to the ground truth tumor position. RESULTS: A total error (root mean square error) of 0.87 ± 0.23 mm and 1.03 ± 0.26 mm was found for posterior-anterior and lateral imaging; the mean time for BS was 8.03 ± 1.54 ms. Without utilizing BS, TMM failed in all X-ray images since the registration algorithm focused on the rib position due to the predominant intensity of this tissue within DRR and X-ray images. CONCLUSION: The BS algorithm developed and implemented improved the accuracy, robustness, and stability of real-time TMM in lung cancer in a phantom study, even in the case of rib interlude where normal tumor registration fails.

3D printed patient-specific thorax phantom with realistic heterogenous bone radiopacity using filament printer technology.

Current medical imaging phantoms are usually limited by simplified geometry and radiographic skeletal homogeneity, which confines their usage for image quality assessment. In order to fabricate realistic imaging phantoms, replication of the entire tissue morphology and the associated CT numbers, defined as Hounsfield Unit (HU) is required. 3D printing is a promising technology for the production of medical imaging phantoms with accurate anatomical replication. So far, the majority of the imaging phantoms using 3D printing technologies tried to mimic the average HU of soft tissue human organs. One important aspect of the anthropomorphic imaging phantoms is also the replication of realistic radiodensities for bone tissues. In this study, we used filament printing technology to develop a CT-derived 3D printed thorax phantom with realistic bone-equivalent radiodensity using only one single commercially available filament. The generated thorax phantom geometry closely resembles a patient and includes direct manufacturing of bone structures while creating life-like heterogeneity within bone tissues. A HU analysis as well as a physical dimensional comparison were performed in order to evaluate the density and geometry agreement between the proposed phantom and the corresponding CT data. With the achieved density range (-482 to 968 HU) we could successfully mimic the realistic radiodensity of the bone marrow as well as the cortical bone for the ribs, vertebral body and dorsal vertebral column in the thorax skeleton. In addition, considering the large radiodensity range achieved a full thorax imaging phantom mimicking also soft tissues can become feasible. The physical dimensional comparison using both Extrema Analysis and Collision Detection methods confirmed a mean surface overlap of 90% and a mean volumetric overlap of 84,56% between the patient and phantom model. Furthermore, the reproducibility analyses revealed a good geometry and radiodensity duplicability in 24 printed cylinder replicas. Thus, according to our results, the proposed additively manufactured anthropomorphic thorax phantom has the potential to be efficiently used for validation of imaging- and radiation-based procedures in precision medicine.