DblurDoseNet: A deep residual learning network for voxel radionuclide dosimetry compensating for single-photon emission computerized tomography imaging resolution

Linear Boltzmann equation solver for voxel-level dosimetry in radiopharmaceutical therapy: Comparison with Monte Carlo and kernel convolution

BACKGROUND

With recent interest in patient-specific dosimetry for radiopharmaceutical therapy (RPT) and selective internal radiation therapy (SIRT), an increasing number of voxel-based algorithms are being evaluated. Monte Carlo (MC) radiation transport, generally considered to be the most accurate among different methods for voxel-level absorbed dose estimation, can be computationally inefficient for routine clinical use.

PURPOSE

This work demonstrates a recently implemented grid-based linear Boltzmann transport equation (LBTE) solver for fast and accurate voxel-based dosimetry in RPT and SIRT and benchmarks it against MC.

METHODS

A deterministic LBTE solver (Acuros MRT) was implemented within a commercial RPT dosimetry package (Velocity 4.1). The LBTE is directly discretized using an adaptive mesh refined grid and then the coupled photon-electron radiation transport is iteratively solved inside specified volumes to estimate radiation doses from both photons and charged particles in heterogeneous media. To evaluate the performance of the LBTE solver for RPT and SIRT applications, 177Lu SPECT/CT, 90Y PET/CT, and 131I SPECT/CT images of phantoms and patients were used. Multiple lesions (2-1052 mL) and normal organs were delineated for each study. Voxel dosimetry was performed with the LBTE solver, dose voxel kernel (DVK) convolution with density correction, and a validated in-house MC code using the same time-integrated activity and density maps as input to the different dose engines. The resulting dose maps, difference maps, and dose-volume-histogram (DVH) metrics were compared, to assess the voxel-level agreement. Evaluation of mean absorbed dose included comparison with structure-level estimates from OLINDA.

RESULTS

In the phantom inserts/compartments, the LBTE solver versus MC and DVK convolution demonstrated good agreement with mean absorbed dose and DVH metrics agreeing to within 5% except for the D90 and D70 metrics of a very low activity concentration insert of 90Y where the agreement was within 15%. In the patient studies (five patients imaged after 177Lu DOTATATE RPT, five after 90Y SIRT, and two after 131I radioimmunotherapy), in general, there was better agreement between the LBTE solver and MC than between LBTE solver and DVK convolution for mean absorbed dose and voxel-level evaluations. Across all patients for all three radionuclides, for soft tissue structures (kidney, liver, lesions), the mean absorbed dose estimates from the LBTE solver were in good agreement with those from MC (median difference < 1%, maximum 9%) and those from DVK (median difference < 5%, maximum 9%). The LBTE and OLINDA estimates for mean absorbed dose in kidneys and liver agreed to within 10%, but differences for lesions were larger with a maximum 14% for 177Lu, 23% for 90Y, and 26% for 131I. For bone regions, the agreement in mean absorbed doses between LBTE and both MC and DVK were similar (median < 11%, max 11%) while for lung the agreement between LBTE and MC (median < 1%, max 8%) was substantially better than between LBTE and DVK (median < 16%, max 33%). Voxel level estimates for soft tissue structures also showed good agreement between the LBTE solver and both MC and DVK with a median difference < 5% (maximum < 13%) for the DVH metrics with all three radionuclides. The largest difference in DVH metrics was for the D90 and D70 metric in lung and bone where the uptake was low. Here, the difference between LBTE and MC had a median value < 14% (maximum 23%) for bone and < 4% (maximum 37%) for lung, while the corresponding differences between LBTE and DVK were < 23% (maximum 31%) and < 67% (maximum 313%), respectively. For a typical patient with a matrix size of 166 × 166 × 129 (voxel size 3 × 3 × 3 mm3), voxel dosimetry using the LBTE solver was as fast as ∼2 min on a desktop computer.

CONCLUSION

Having established good agreement between the LBTE solver and MC for RPT and SIRT applications, the LBTE solver is a viable option for voxel dosimetry that can be faster than MC. Further analysis is being performed to encompass the broad range of radionuclides and conditions encountered clinically.

Image-based dosimetry for [225Ac]Ac-PSMA-I&T therapy and the effect of daughter-specific pharmacokinetics

PURPOSE

Although 221Fr and 213Bi have sufficient gamma emission probabilities, quantitative SPECT after [225Ac]Ac-PSMA-I&T therapy remains challenging due to low therapeutic activities. Furthermore, 221Fr and 213Bi may underlie a different pharmacokinetics due to alpha recoil. We conducted a quantitative SPECT study and a urine analysis to investigate the pharmacokinetics of 221Fr and 213Bi and the impact on image-based lesion and kidney dosimetry.

METHODS

Five patients (7.7 ± 0.2 MBq [225Ac]Ac-PSMA-I&T) underwent an abdominal SPECT/CT (1 h) at 24 and 48 h (Siemens Symbia T2, high-energy collimator, 440 keV/218 keV (width 20%), 78 keV (width 50%)). Quantitative SPECT was reconstructed using MAP-EM with attenuation and transmission-dependent scatter corrections and resolution modelling. Time-activity curves for kidneys (CT-based) and lesions (80% isocontour 24 h) were fitted mono-exponentially. Urine samples collected along with each SPECT/CT were measured in a gamma counter until secular equilibrium was reached.

RESULTS

Mean kidney and lesion effective half-lives were as follows: 213Bi, 27 ± 6/38 ± 10 h; 221Fr, 24 ± 6/38 ± 11 h; 78 keV, 23 ± 7/39 ± 13 h. The 213Bi-to-221Fr kidney SUV ratio increased by an average of 9% from 24 to 48 h. Urine analysis revealed an increasing 213Bi-to-225Ac ratio (24 h, 0.98 ± 0.15; 48 h, 1.08 ± 0.09). Mean kidney and lesion absorbed doses were 0.17 ± 0.06 and 0.36 ± 0.1 Sv RBE = 5 /MBq using 221Fr and 213Bi SPECT images, compared to 0.16 ± 0.05/0.18 ± 0.06 and 0.36 ± 0.1/0.38 ± 0.1 Sv RBE = 5 /MBq considering either the 221Fr or 213Bi SPECT.

CONCLUSION

SPECT/CT imaging and urine analysis showed minor differences of up to 10% in the daughter-specific pharmacokinetics. These variances had a minimal impact on the lesion and kidney dosimetry which remained within 8%.

A Deep-Learning-Based Partial-Volume Correction Method for Quantitative 177Lu SPECT/CT Imaging

With the development of new radiopharmaceutical therapies, quantitative SPECT/CT has progressively emerged as a crucial tool for dosimetry. One major obstacle of SPECT is its poor resolution, which results in blurring of the activity distribution. Especially for small objects, this so-called partial-volume effect limits the accuracy of activity quantification. Numerous methods for partial-volume correction (PVC) have been proposed, but most methods have the disadvantage of assuming a spatially invariant resolution of the imaging system, which does not hold for SPECT. Furthermore, most methods require a segmentation based on anatomic information. Methods: We introduce DL-PVC, a methodology for PVC of 177Lu SPECT/CT imaging using deep learning (DL). Training was based on a dataset of 10,000 random activity distributions placed in extended cardiac-torso body phantoms. Realistic SPECT acquisitions were created using the SIMIND Monte Carlo simulation program. SPECT reconstructions without and with resolution modeling were performed using the CASToR and STIR reconstruction software, respectively. The pairs of ground-truth activity distributions and simulated SPECT images were used for training various U-Nets. Quantitative analysis of the performance of these U-Nets was based on metrics such as the structural similarity index measure or normalized root-mean-square error, but also on volume activity accuracy, a new metric that describes the fraction of voxels in which the determined activity concentration deviates from the true activity concentration by less than a certain margin. On the basis of this analysis, the optimal parameters for normalization, input size, and network architecture were identified. Results: Our simulation-based analysis revealed that DL-PVC (0.95/7.8%/35.8% for structural similarity index measure/normalized root-mean-square error/volume activity accuracy) outperforms SPECT without PVC (0.89/10.4%/12.1%) and after iterative Yang PVC (0.94/8.6%/15.1%). Additionally, we validated DL-PVC on 177Lu SPECT/CT measurements of 3-dimensionally printed phantoms of different geometries. Although DL-PVC showed activity recovery similar to that of the iterative Yang method, no segmentation was required. In addition, DL-PVC was able to correct other image artifacts such as Gibbs ringing, making it clearly superior at the voxel level. Conclusion: In this work, we demonstrate the added value of DL-PVC for quantitative 177Lu SPECT/CT. Our analysis validates the functionality of DL-PVC and paves the way for future deployment on clinical image data.

Deep transformer-based personalized dosimetry from SPECT/CT images: a hybrid approach for [177Lu]Lu-DOTATATE radiopharmaceutical therapy

PURPOSE

Accurate dosimetry is critical for ensuring the safety and efficacy of radiopharmaceutical therapies. In current clinical dosimetry practice, MIRD formalisms are widely employed. However, with the rapid advancement of deep learning (DL) algorithms, there has been an increasing interest in leveraging the calculation speed and automation capabilities for different tasks. We aimed to develop a hybrid transformer-based deep learning (DL) model that incorporates a multiple voxel S-value (MSV) approach for voxel-level dosimetry in [177Lu]Lu-DOTATATE therapy. The goal was to enhance the performance of the model to achieve accuracy levels closely aligned with Monte Carlo (MC) simulations, considered as the standard of reference. We extended our analysis to include MIRD formalisms (SSV and MSV), thereby conducting a comprehensive dosimetry study.

METHODS

We used a dataset consisting of 22 patients undergoing up to 4 cycles of [177Lu]Lu-DOTATATE therapy. MC simulations were used to generate reference absorbed dose maps. In addition, MIRD formalism approaches, namely, single S-value (SSV) and MSV techniques, were performed. A UNEt TRansformer (UNETR) DL architecture was trained using five-fold cross-validation to generate MC-based dose maps. Co-registered CT images were fed into the network as input, whereas the difference between MC and MSV (MC-MSV) was set as output. DL results are then integrated to MSV to revive the MC dose maps. Finally, the dose maps generated by MSV, SSV, and DL were quantitatively compared to the MC reference at both voxel level and organ level (organs at risk and lesions).

RESULTS

The DL approach showed slightly better performance (voxel relative absolute error (RAE) = 5.28 ± 1.32) compared to MSV (voxel RAE = 5.54 ± 1.4) and outperformed SSV (voxel RAE = 7.8 ± 3.02). Gamma analysis pass rates were 99.0 ± 1.2%, 98.8 ± 1.3%, and 98.7 ± 1.52% for DL, MSV, and SSV approaches, respectively. The computational time for MC was the highest (~2 days for a single-bed SPECT study) compared to MSV, SSV, and DL, whereas the DL-based approach outperformed the other approaches in terms of time efficiency (3 s for a single-bed SPECT). Organ-wise analysis showed absolute percent errors of 1.44 ± 3.05%, 1.18 ± 2.65%, and 1.15 ± 2.5% for SSV, MSV, and DL approaches, respectively, in lesion-absorbed doses.

CONCLUSION

A hybrid transformer-based deep learning model was developed for fast and accurate dose map generation, outperforming the MIRD approaches, specifically in heterogenous regions. The model achieved accuracy close to MC gold standard and has potential for clinical implementation for use on large-scale datasets.

Theranostics and artificial intelligence: new frontiers in personalized medicine

The field of theranostics is rapidly advancing, driven by the goals of enhancing patient care. Recent breakthroughs in artificial intelligence (AI) and its innovative theranostic applications have marked a critical step forward in nuclear medicine, leading to a significant paradigm shift in precision oncology. For instance, AI-assisted tumor characterization, including automated image interpretation, tumor segmentation, feature identification, and prediction of high-risk lesions, improves diagnostic processes, offering a precise and detailed evaluation. With a comprehensive assessment tailored to an individual's unique clinical profile, AI algorithms promise to enhance patient risk classification, thereby benefiting the alignment of patient needs with the most appropriate treatment plans. By uncovering potential factors unseeable to the human eye, such as intrinsic variations in tumor radiosensitivity or molecular profile, AI software has the potential to revolutionize the prediction of response heterogeneity. For accurate and efficient dosimetry calculations, AI technology offers significant advantages by providing customized phantoms and streamlining complex mathematical algorithms, making personalized dosimetry feasible and accessible in busy clinical settings. AI tools have the potential to be leveraged to predict and mitigate treatment-related adverse events, allowing early interventions. Additionally, generative AI can be utilized to find new targets for developing novel radiopharmaceuticals and facilitate drug discovery. However, while there is immense potential and notable interest in the role of AI in theranostics, these technologies do not lack limitations and challenges. There remains still much to be explored and understood. In this study, we investigate the current applications of AI in theranostics and seek to broaden the horizons for future research and innovation.

Semi-supervised low-dose SPECT restoration using sinogram inner-structure aware graph neural network

Objective.To mitigate the potential radiation risk, low-dose single photon emission computed tomography (SPECT) is of increasing interest. Numerous deep learning-based methods have been developed to perform low-dose imaging while maintaining image quality. However, most existing methods seldom explore the unique inner-structure inherent within sinograms. In addition, traditional supervised learning methods require large-scale labeled data, where the normal-dose data serves as annotation and is intractable to acquire in low-dose imaging. In this study, we aim to develop a novel sinogram inner-structure-aware semi-supervised framework for the task of low-dose SPECT sinogram restoration.Approach.The proposed framework retains the strengths of UNet, meanwhile introducing a sinogram-structure-based non-local neighbors graph neural network (SSN-GNN) module and a window-based K-nearest neighbors GNN (W-KNN-GNN) module to effectively exploit the inherent inner-structure within SPECT sinograms. Moreover, the proposed framework employs the mean teacher semi-supervised learning approach to leverage the information available in abundant unlabeled low-dose sinograms.Main results.The datasets exploited in this study were acquired from the (Extended Cardiac-Torso) XCAT anthropomorphic digital phantoms, which provide realistic images for imaging research of various modalities. Quantitative as well as qualitative results demonstrate that the proposed framework achieves superior performance compared to several state-of-the-art reconstruction methods. To further validate the effectiveness of the proposed framework, ablation and robustness experiments were also performed. The experimental results show that each component of the proposed framework effectively improves the model performance, and the framework exhibits superior robustness with respect to various noise levels. Besides, the proposed semi-supervised paradigm showcases the efficacy of incorporating supplementary unlabeled low-dose sinograms.Significance.The proposed framework improves the quality of low-dose SPECT reconstructed images by utilizing sinogram inner-structure and incorporating supplementary unlabeled data, which provides an important tool for dose reduction without sacrificing the image quality.

Radiomics and Artificial Intelligence in Radiotheranostics: A Review of Applications for Radioligands Targeting Somatostatin Receptors and Prostate-Specific Membrane Antigens

Radiotheranostics refers to the pairing of radioactive imaging biomarkers with radioactive therapeutic compounds that deliver ionizing radiation. Given the introduction of very promising radiopharmaceuticals, the radiotheranostics approach is creating a novel paradigm in personalized, targeted radionuclide therapies (TRTs), also known as radiopharmaceuticals (RPTs). Radiotherapeutic pairs targeting somatostatin receptors (SSTR) and prostate-specific membrane antigens (PSMA) are increasingly being used to diagnose and treat patients with metastatic neuroendocrine tumors (NETs) and prostate cancer. In parallel, radiomics and artificial intelligence (AI), as important areas in quantitative image analysis, are paving the way for significantly enhanced workflows in diagnostic and theranostic fields, from data and image processing to clinical decision support, improving patient selection, personalized treatment strategies, response prediction, and prognostication. Furthermore, AI has the potential for tremendous effectiveness in patient dosimetry which copes with complex and time-consuming tasks in the RPT workflow. The present work provides a comprehensive overview of radiomics and AI application in radiotheranostics, focusing on pairs of SSTR- or PSMA-targeting radioligands, describing the fundamental concepts and specific imaging/treatment features. Our review includes ligands radiolabeled by 68Ga, 18F, 177Lu, 64Cu, 90Y, and 225Ac. Specifically, contributions via radiomics and AI towards improved image acquisition, reconstruction, treatment response, segmentation, restaging, lesion classification, dose prediction, and estimation as well as ongoing developments and future directions are discussed.

90Y SPECT scatter estimation and voxel dosimetry in radioembolization using a unified deep learning framework

PURPOSE

90Y SPECT-based dosimetry following radioembolization (RE) in liver malignancies is challenging due to the inherent scatter and the poor spatial resolution of bremsstrahlung SPECT. This study explores a deep-learning-based absorbed dose-rate estimation method for 90Y that mitigates the impact of poor SPECT image quality on dosimetry and the accuracy-efficiency trade-off of Monte Carlo (MC)-based scatter estimation and voxel dosimetry methods.

METHODS

Our unified framework consists of three stages: convolutional neural network (CNN)-based bremsstrahlung scatter estimation, SPECT reconstruction with scatter correction (SC) and absorbed dose-rate map generation with a residual learning network (DblurDoseNet). The input to the framework is the measured SPECT projections and CT, and the output is the absorbed dose-rate map. For training and testing under realistic conditions, we generated a series of virtual patient phantom activity/density maps from post-therapy images of patients treated with 90Y-RE at our clinic. To train the scatter estimation network, we use the scatter projections for phantoms generated from MC simulation as the ground truth (GT). To train the dosimetry network, we use MC dose-rate maps generated directly from the activity/density maps of phantoms as the GT (Phantom + MC Dose). We compared performance of our framework (SPECT w/CNN SC + DblurDoseNet) and MC dosimetry (SPECT w/CNN SC + MC Dose) using normalized root mean square error (NRMSE) and normalized mean absolute error (NMAE) relative to GT.

RESULTS

When testing on virtual patient phantoms, our CNN predicted scatter projections had NRMSE of 4.0% ± 0.7% on average. For the SPECT reconstruction with CNN SC, we observed a significant improvement on NRMSE (9.2% ± 1.7%), compared to reconstructions with no SC (149.5% ± 31.2%). In terms of virtual patient dose-rate estimation, SPECT w/CNN SC + DblurDoseNet had a NMAE of 8.6% ± 5.7% and 5.4% ± 4.8% in lesions and healthy livers, respectively; compared to 24.0% ± 6.1% and 17.7% ± 2.1% for SPECT w/CNN SC + MC Dose. In patient dose-rate maps, though no GT was available, we observed sharper lesion boundaries and increased lesion-to-background ratios with our framework. For a typical patient data set, the trained networks took ~ 1 s to generate the scatter estimate and ~ 20 s to generate the dose-rate map (matrix size: 512 × 512 × 194) on a single GPU (NVIDIA V100).

CONCLUSION

Our deep learning framework, trained using true activity/density maps, has the potential to outperform non-learning voxel dosimetry methods such as MC that are dependent on SPECT image quality. Across comprehensive testing and evaluations on multiple targeted lesions and healthy livers in virtual patients, our proposed deep learning framework demonstrated higher (66% on average in terms of NMAE) estimation accuracy than the current "gold-standard" MC method. The enhanced computing speed with our framework without sacrificing accuracy is highly relevant for clinical dosimetry following 90Y-RE.

On the Use of Artificial Intelligence for Dosimetry of Radiopharmaceutical Therapies

Routine clinical dosimetry along with radiopharmaceutical therapies is key for future treatment personalization. However, dosimetry is considered complex and time-consuming with various challenges amongst the required steps within the dosimetry workflow. The general workflow for image-based dosimetry consists of quantitative imaging, the segmentation of organs and tumors, fitting of the time-activity-curves, and the conversion to absorbed dose. This work reviews the potential and advantages of the use of artificial intelligence to improve speed and accuracy of every single step of the dosimetry workflow.

Future trends for patient-specific dosimetry methodology in molecular radiotherapy

Molecular radiotherapy is rapidly expanding, and new radiotherapeutics are emerging. The majority of treatments is still performed using empirical fixed activities and not tailored for individual patients. Molecular radiotherapy dosimetry is often seen as a promising candidate that would allow personalisation of treatments as outcome should ultimately depend on the absorbed doses delivered and not the activities administered. The field of molecular radiotherapy dosimetry has made considerable progress towards the feasibility of routine clinical dosimetry with reasonably accurate absorbed-dose estimates for a range of molecular radiotherapy dosimetry applications. A range of challenges remain with respect to the accurate quantification, assessment of time-integrated activity and absorbed dose estimation. In this review, we summarise a range of technological and methodological advancements, mainly focussed on beta-emitting molecular radiotherapeutics, that aim to improve molecular radiotherapy dosimetry to achieve accurate, reproducible, and streamlined dosimetry. We describe how these new technologies can potentially improve the often time-consuming considered process of dosimetry and provide suggestions as to what further developments might be required.

Physics-Guided Deep Scatter Estimation by Weak Supervision for Quantitative SPECT

Accurate scatter estimation is important in quantitative SPECT for improving image contrast and accuracy. With a large number of photon histories, Monte-Carlo (MC) simulation can yield accurate scatter estimation, but is computationally expensive. Recent deep learning-based approaches can yield accurate scatter estimates quickly, yet full MC simulation is still required to generate scatter estimates as ground truth labels for all training data. Here we propose a physics-guided weakly supervised training framework for fast and accurate scatter estimation in quantitative SPECT by using a 100× shorter MC simulation as weak labels and enhancing them with deep neural networks. Our weakly supervised approach also allows quick fine-tuning of the trained network to any new test data for further improved performance with an additional short MC simulation (weak label) for patient-specific scatter modelling. Our method was trained with 18 XCAT phantoms with diverse anatomies / activities and then was evaluated on 6 XCAT phantoms, 4 realistic virtual patient phantoms, 1 torso phantom and 3 clinical scans from 2 patients for 177Lu SPECT with single / dual photopeaks (113, 208 keV). Our proposed weakly supervised method yielded comparable performance to the supervised counterpart in phantom experiments, but with significantly reduced computation in labeling. Our proposed method with patient-specific fine-tuning achieved more accurate scatter estimates than the supervised method in clinical scans. Our method with physics-guided weak supervision enables accurate deep scatter estimation in quantitative SPECT, while requiring much lower computation in labeling, enabling patient-specific fine-tuning capability in testing.

A machine learning-based model for a dose point kernel calculation

PURPOSE

Absorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study reports on the design, implementation, and test of a multi-target regressor approach to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.

METHODS

DPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Regressor Chains (RC) with three different coefficients regularization/shrinkage models were used as base regressors. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the beta emitters sDPK were applied to a patient-specific case calculating the Voxel Dose Kernel (VDK) for a hepatic radioembolization treatment with [Formula: see text]Y.

RESULTS

The three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than [Formula: see text] in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than [Formula: see text] were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.

CONCLUSION

An ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required short computation times.

Voxel-Based Internal Dosimetry for 177Lu-Labeled Radiopharmaceutical Therapy Using Deep Residual Learning

Purpose

In this study, we propose a deep learning (DL)-based voxel-based dosimetry method in which dose maps acquired using the multiple voxel S-value (VSV) approach were used for residual learning.

Methods

Twenty-two SPECT/CT datasets from seven patients who underwent 177Lu-DOTATATE treatment were used in this study. The dose maps generated from Monte Carlo (MC) simulations were used as the reference approach and target images for network training. The multiple VSV approach was used for residual learning and compared with dose maps generated from deep learning. The conventional 3D U-Net network was modified for residual learning. The absorbed doses in the organs were calculated as the mass-weighted average of the volume of interest (VOI).

Results

The DL approach provided a slightly more accurate estimation than the multiple-VSV approach, but the results were not statistically significant. The single-VSV approach yielded a relatively inaccurate estimation. No significant difference was noted between the multiple VSV and DL approach on the dose maps. However, this difference was prominent in the error maps. The multiple VSV and DL approach showed a similar correlation. In contrast, the multiple VSV approach underestimated doses in the low-dose range, but it accounted for the underestimation when the DL approach was applied.

Conclusion

Dose estimation using the deep learning-based approach was approximately equal to that in the MC simulation. Accordingly, the proposed deep learning network is useful for accurate and fast dosimetry after radiation therapy using 177Lu-labeled radiopharmaceuticals.

A Machine Learning based model for a Dose Point Kernel calculation

PURPOSE

Absorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study shows applications of machine learning to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.

METHODS

DPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Three machine learning (ML) algorithms were trained using the MC DPKs. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the ML sDPK approach was applied to a patient-specific case calculating the dose voxel kernels (DVK) for a hepatic radioembolization treatment with \(^{90}\)Y.

RESULTS

The three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than \(10%\) in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than \(7 %\) were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.

CONCLUSION

An ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required remarkable short computation times.