On the Use of Artificial Intelligence for Dosimetry of Radiopharmaceutical Therapies

Brosch-Lenz, Julia Franziska; Delker, Astrid; Schmidt, Fabian; Tran-Gia, Johannes

doi:10.1055/a-2179-6872

Cited by 3 publications

(2 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning (DL) has been successfully employed for different computational medical imaging tasks [ 23 – 29 ]. There have been some attempts to use DL-based voxel-wise internal dosimetry in previous studies [ 15 , 30 – 33 ].…”

Section: Introductionmentioning

confidence: 99%

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

Mansouri,

Salimi,

Akhavanallaf

et al. 2024

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

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.

show abstract

Section: Introductionmentioning

confidence: 99%

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

Mansouri,

Salimi,

Akhavanallaf

et al. 2024

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

show abstract

“…Only through comparable MRT dosimetry across sites, principles for patient-specific MRT can be developed and applied clinically in the future. In addition, harmonization is crucial for the application of artificial intelligence in SPECT/CT imaging [ 3 ]. Due to the often small number of specific cases in nuclear medicine, it is possible to increase the amount of training data and thus decisively improve the efficiency of neural networks by data sharing across several centers.…”

Section: Introductionmentioning

confidence: 99%

Position dependence of recovery coefficients in 177Lu-SPECT/CT reconstructions – phantom simulations and measurements

Leube,

Claeys,

Gustafsson

et al. 2024

EJNMMI Phys

Self Cite

View full text Add to dashboard Cite

Background Although the importance of quantitative SPECT has increased tremendously due to newly developed therapeutic radiopharmaceuticals, there are still no accreditation programs to harmonize SPECT imaging. Work is currently underway to develop an accreditation for quantitative 177Lu SPECT/CT. The aim of this study is to verify whether the positioning of the spheres within the phantom has an influence on the recovery and thus needs to be considered in SPECT harmonization. In addition, the effects of these recovery coefficients on a potential partial volume correction as well as absorbed-dose estimates are investigated. Methods Using a low-dose CT of a SPECT/CT acquisition, a computerized version of the NEMA body phantom was created using a semi-automatic threshold-based method. Based on the mass-density map, the detector orbit, and the sphere centers, realistic SPECT acquisitions of all possible 720 sphere configurations of both the PET and the SPECT versions of the NEMA Body Phantom were generated using Monte Carlo simulations. SPECT reconstructions with different numbers of updates were performed without (CASToR) and with resolution modeling (STIR). Recovery coefficients were calculated for all permutations, reconstruction methods, and phantoms, and their dependence on the sphere positioning was investigated. Finally, the simulation-based findings were validated using SPECT/CT acquisitions of six different sphere configurations. Results Our analysis shows that sphere positioning has a significant impact on the recovery for both of the reconstruction methods and the phantom type. Although resolution modeling resulted in significantly higher recovery, the relative variation in recovery within the 720 permutations was even larger. When examining the extreme values of the recovery, reconstructions without resolution modeling were influenced primarily by the sphere position, while with resolution modeling the volume of the two adjacent spheres had a larger influence. The SPECT measurements confirmed these observations, and the recovery curves showed good overall agreement with the simulated data. Conclusion Our study shows that sphere positioning has a significant impact on the recovery obtained in NEMA sphere phantom measurements and should therefore be considered in a future SPECT accreditation. Furthermore, the single-measurement method normally performed for PVC should be reconsidered to account for the position dependency.

show abstract

Future Perspectives of Artificial Intelligence in Bone Marrow Dosimetry and Individualized Radioligand Therapy

Moraitis,

Küper,

Tran-Gia

et al. 2024

Seminars in Nuclear Medicine

View full text Add to dashboard Cite

On the Use of Artificial Intelligence for Dosimetry of Radiopharmaceutical Therapies

Cited by 3 publications

References 67 publications

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

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

Position dependence of recovery coefficients in 177Lu-SPECT/CT reconstructions – phantom simulations and measurements

Future Perspectives of Artificial Intelligence in Bone Marrow Dosimetry and Individualized Radioligand Therapy

Contact Info

Product

Resources

About