Search

Background: Adaptive radiotherapy (ART) workflows have been increasingly adopted to achieve dose escalation and tissue sparing under shifting anatomic conditions, but the necessity of recontouring and the associated time burden hinders a real‐time or online ART workflow. In response to this challenge, approaches to auto‐segmentation involving deformable image registration, atlas‐based segmentation, and deep learning‐based segmentation (DLS) have been developed. Despite the particular promise shown by DLS methods, implementing these approaches in a clinical setting remains a challenge, namely due to the difficulty of curating a data set of sufficient size and quality so as to achieve generalizability in a trained model. Purpose: To address this challenge, we have developed an intentional deep overfit learning (IDOL) framework tailored to the auto‐segmentation task. However, certain limitations were identified, particularly the insufficiency of the personalized dataset to effectively overfit the model. In this study, we introduce a personalized hyperspace learning (PHL)‐IDOL segmentation framework capable of generating datasets that induce the model to overfit specific patient characteristics for medical image segmentation. Methods: The PHL‐IDOL model is trained in two stages. In the first, a conventional, general model is trained with a diverse set of patient data (n = 100 patients) consisting of CT images and clinical contours. Following this, the general model is tuned with a data set consisting of two components: (a) selection of a subset of the patient data (m < n) using the similarity metrics (mean square error (MSE), peak signal‐to‐noise ratio (PSNR), structural similarity index (SSIM), and the universal quality image index (UQI) values); (b) adjust the CT and the clinical contours using a deformed vector generated from the reference patient and the selected patients using (a). After training, the general model, the continual model, the conventional IDOL model, and the proposed PHL‐IDOL model were evaluated using the volumetric dice similarity coefficient (VDSC) and the Hausdorff distance 95% (HD95%) computed for 18 structures in 20 test patients. Results: Implementing the PHL‐IDOL framework resulted in improved segmentation performance for each patient. The Dice scores increased from 0.81±$ \pm $0.05 with the general model, 0.83±0.04$ \pm 0.04$ for the continual model, 0.83±0.04$ \pm 0.04$ for the conventional IDOL model to an average of 0.87±0.03$ \pm 0.03$ with the PHL‐IDOL model. Similarly, the Hausdorff distance decreased from 3.06±0.99$ \pm 0.99$ with the general model, 2.84±0.69$ \pm 0.69$ for the continual model, 2.79±0.79$ \pm 0.79$ for the conventional IDOL model and 2.36±0.52$ \pm 0.52$ for the PHL‐IDOL model. All the standard deviations were decreased by nearly half of the values comparing the general model and the PHL‐IDOL model. Conclusion: The PHL‐IDOL framework applied to the auto‐segmentation task achieves improved performance compared to the general DLS approach, demonstrating the promise of leveraging patient‐specific prior information in a task central to online ART workflows. [ABSTRACT FROM AUTHOR]

Background: Phenomenological relative biological effectiveness (RBE) models for proton therapy, based on the dose‐averaged linear energy transfer (LET), have been developed to address the apparent RBE increase towards the end of the proton range. The results of these phenomenological models substantially differ due to varying empirical assumptions and fitting functions. In contrast, more theory‐based approaches are used in carbon ion radiotherapy, such as the microdosimetric kinetic model (MKM). However, implementing microdosimetry‐based models in LET‐based proton therapy treatment planning systems poses challenges. Purpose: This work presents a LET‐based version of the MKM that is practical for clinical use in proton radiotherapy. Methods: At first, we derived an approximation of the Mayo Clinic Florida (MCF) MKM for relatively‐sparsely ionizing radiation such as protons. The mathematical formalism of the proposed model is equivalent to the original MKM, but it maintains some key features of the MCF MKM, such as the determination of model parameters from measurable cell characteristics. Subsequently, we carried out Monte Carlo calculations with PHITS in different simulated scenarios to establish a heuristic correlation between microdosimetric quantities and the dose averaged LET of protons. Results: A simple allometric function was found able to describe the relationship between the dose‐averaged LET of protons and the dose‐mean lineal energy, which includes the contributions of secondary particles. The LET‐based MKM was used to model the in vitro clonogenic survival RBE of five human and rodent cell lines (A549, AG01522, CHO, T98G, and U87) exposed to pristine and spread‐out Bragg peak (SOBP) proton beams. The results of the LET‐based MKM agree well with the biological data in a comparable or better way with respect to the other models included in the study. A sensitivity analysis on the model results was also performed. Conclusions: The LET‐based MKM integrates the predictive theoretical framework of the MCF MKM with a straightforward mathematical description of the RBE based on the dose‐averaged LET, a physical quantity readily available in modern treatment planning systems for proton therapy. [ABSTRACT FROM AUTHOR]

Background: Kilo‐voltage cone‐beam computed tomography (CBCT) is a prevalent modality used for adaptive radiotherapy (ART) due to its compatibility with linear accelerators and ability to provide online imaging. However, the widely‐used Feldkamp‐Davis‐Kress (FDK) reconstruction algorithm has several limitations, including potential streak aliasing artifacts and elevated noise levels. Iterative reconstruction (IR) techniques, such as total variation (TV) minimization, dictionary‐based methods, and prior information‐based methods, have emerged as viable solutions to address these limitations and improve the quality and applicability of CBCT in ART. Purpose: One of the primary challenges in IR‐based techniques is finding the right balance between minimizing image noise and preserving image resolution. To overcome this challenge, we have developed a new reconstruction technique called high‐resolution CBCT (HRCBCT) that specifically focuses on improving image resolution while reducing noise levels. Methods: The HRCBCT reconstruction technique builds upon the conventional IR approach, incorporating three components: the data fidelity term, the resolution preservation term, and the regularization term. The data fidelity term ensures alignment between reconstructed values and measured projection data, while the resolution preservation term exploits the high resolution of the initial Feldkamp‐Davis‐Kress (FDK) algorithm. The regularization term mitigates noise during the IR process. To enhance convergence and resolution at each iterative stage, we applied Iterative Filtered Backprojection (IFBP) to the data fidelity minimization process. Results: We evaluated the performance of the proposed HRCBCT algorithm using data from two physical phantoms and one head and neck patient. The HRCBCT algorithm outperformed all four different algorithms; FDK, Iterative Filtered Back Projection (IFBP), Compressed Sensing based Iterative Reconstruction (CSIR), and Prior Image Constrained Compressed Sensing (PICCS) methods in terms of resolution and noise reduction for all data sets. Line profiles across three line pairs of resolution revealed that the HRCBCT algorithm delivered the highest distinguishable line pairs compared to the other algorithms. Similarly, the Modulation Transfer Function (MTF) measurements, obtained from the tungsten wire insert on the CatPhan 600 physical phantom, showed a significant improvement with HRCBCT over traditional algorithms. Conclusion: The proposed HRCBCT algorithm offers a promising solution for enhancing CBCT image quality in adaptive radiotherapy settings. By addressing the challenges inherent in traditional IR methods, the algorithm delivers high‐definition CBCT images with improved resolution and reduced noise throughout each iterative step. Implementing the HR CBCT algorithm could significantly impact the accuracy of treatment planning during online adaptive therapy. [ABSTRACT FROM AUTHOR]

Background: In order to compute the relative biological effectiveness (RBE) of ion radiation therapy with the Mayo Clinic Florida microdosimetric kinetic model (MCF MKM), it is necessary to process entire microdosimetric distributions. Therefore, a posteriori RBE recalculations (i.e., for a different cell line or another biological endpoint) would require whole spectral information. It is currently not practical to compute and store all this data for each clinical voxel. Purpose: To develop a methodology that allows to store a limited amount of physical information without losing accuracy in the RBE calculations nor the possibility of a posteriori RBE recalculations. Methods: Computer simulations for four monoenergetic 12C ion beams and a 12C ion spread‐out Bragg peak (SOBP) were performed to assess lineal energy distributions as a function of the depth within a water phantom. These distributions were used in combination with the MCF MKM to compute the in vitro clonogenic survival RBE for human salivary gland tumor cells (HSG cell line) and human skin fibroblasts (NB1RGB cell line). The RBE values were also calculated with a new abridged microdosimetric distribution methodology (AMDM) and compared with the reference RBE calculations using the entire distributions. Results: The maximum relative deviation between the RBE values computed using the entire distributions and the AMDM was 0.61% (monoenergetic beams) and 0.49% (SOBP) for the HSG cell line, while 0.45% (monoenergetic beams) and 0.26% (SOBP) for the NB1RGB cell line. Conclusion: The excellent agreement between the RBE values computed using the entire lineal energy distributions and the AMDM represents a milestone for the clinical implementation of the MCF MKM. [ABSTRACT FROM AUTHOR]