Jong Hwi Jeong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jong Hwi Jeong has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Haksoo Kim
Proton Therapy Center, National Cancer Center, Goyang, Korea
5 shared publications - Young Kyung Lim
Proton Therapy Center, National Cancer Center, Goyang, Korea
5 shared publications - Dongho Shin
Proton Therapy Center, National Cancer Center, Goyang-si, Republic of Korea
4 shared publications - Se Byeong Lee
Proton Therapy Center, National Cancer Center, Goyang, Korea
4 shared publications Chankyu Kim
Department of Radiation Oncology, Proton Therapy Center, National Cancer Center, Goyang-si, South Korea
3 shared publicationsDongho Shin
Department of Radiation Oncology, Proton Therapy Center, National Cancer Center, Goyang-si, South Korea
3 shared publications
Latest Publications
- Open Access
- 17 December 2025
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 15, 1690416
Deep learning-based dose prediction for prostate cancer with empty bladder protocol: a framework for efficient and personalized radiotherapy planning
Radiation therapy (RT) is a cornerstone in the management of localized and locally advanced prostate cancer, traditionally delivered with a full bladder (FB) protocol to reduce radiation exposure to surrounding organs. However, consistent bladder filling is difficult to maintain, leading to workflow delays, anatomical inconsistencies, and variable toxicity outcomes. Recent evidence, including the ongoing RELIEF trial at Mayo Clinic, suggests that an empty bladder (EB) protocol provides comparable toxicity outcomes to FB while improving patient comfort and treatment consistency. To address the increased anatomical variability associated with EB protocols, we developed a deep learning (DL)-based dose prediction model tailored to EB patients. A conditional generative adversarial network (cGAN) with a modified 3D U-Net architecture was trained on 90 FB cases and fine-tuned on 20 EB cases stratified into stereotactic body radiotherapy (SBRT) and intensity-modulated radiotherapy (IMRT). Model performance was evaluated against clinical manual plans using mean absolute percentage error (MAPE) and dose-volume histogram (DVH) metrics. The EB Fine-tuning model(SBRT/IMRT) achieved superior accuracy compared with the general FB-trained model, with an average MAPE of 3.53 ± 0.40% versus 4.87 ± 0.86%. DVH analyses demonstrated improved agreement with manual plans for planning target volumes and organs at risk, with discrepancies consistently within 2.5 Gy or 3%. These results demonstrate that fine-tuning with EB-specific data enhances prediction accuracy and clinical relevance of the DL-based model. The proposed framework supports efficient EB treatment planning, provides reference dose distributions for quality assurance, and offers educational value to clinicians adopting EB protocols. By combining automation with clinical applicability, this approach facilitates broader adoption of EB radiotherapy in prostate cancer while improving workflow reproducibility and patient-centered care.
- 17 December 2025
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 15, 1690416
Open AccessDeep learning-based dose prediction for prostate cancer with empty bladder protocol: a framework for efficient and personalized radiotherapy planning
Radiation therapy (RT) is a cornerstone in the management of localized and locally advanced prostate cancer, traditionally delivered with a full bladder (FB) protocol to reduce radiation exposure to surrounding organs. However, consistent bladder filling is difficult to maintain, leading to workflow delays, anatomical inconsistencies, and variable toxicity outcomes. Recent evidence, including the ongoing RELIEF trial at Mayo Clinic, suggests that an empty bladder (EB) protocol provides comparable toxicity outcomes to FB while improving patient comfort and treatment consistency. To address the increased anatomical variability associated with EB protocols, we developed a deep learning (DL)-based dose prediction model tailored to EB patients. A conditional generative adversarial network (cGAN) with a modified 3D U-Net architecture was trained on 90 FB cases and fine-tuned on 20 EB cases stratified into stereotactic body radiotherapy (SBRT) and intensity-modulated radiotherapy (IMRT). Model performance was evaluated against clinical manual plans using mean absolute percentage error (MAPE) and dose-volume histogram (DVH) metrics. The EB Fine-tuning model(SBRT/IMRT) achieved superior accuracy compared with the general FB-trained model, with an average MAPE of 3.53 ± 0.40% versus 4.87 ± 0.86%. DVH analyses demonstrated improved agreement with manual plans for planning target volumes and organs at risk, with discrepancies consistently within 2.5 Gy or 3%. These results demonstrate that fine-tuning with EB-specific data enhances prediction accuracy and clinical relevance of the DL-based model. The proposed framework supports efficient EB treatment planning, provides reference dose distributions for quality assurance, and offers educational value to clinicians adopting EB protocols. By combining automation with clinical applicability, this approach facilitates broader adoption of EB radiotherapy in prostate cancer while improving workflow reproducibility and patient-centered care.
- Open Access
Feature Importance Analysis of a Deep Learning Model for Predicting Late Bladder Toxicity Occurrence in Uterine Cervical Cancer Patients
(1) In this study, we developed a deep learning (DL) model that can be used to predict late bladder toxicity. (2) We collected data obtained from 281 uterine cervical cancer patients who underwent definitive radiation therapy. The DL model was trained using 16 features, including patient, tumor, treatment, and dose parameters, and its performance was compared with that of a multivariable logistic regression model using the following metrics: accuracy, prediction, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). In addition, permutation feature importance was calculated to interpret the DL model for each feature, and the lightweight DL model was designed to focus on the top five important features. (3) The DL model outperformed the multivariable logistic regression model on our dataset. It achieved an F1-score of 0.76 and an AUROC of 0.81, while the corresponding values for the multivariable logistic regression were 0.14 and 0.43, respectively. The DL model identified the doses for the most exposed 2 cc volume of the bladder (BD2cc) as the most important feature, followed by BD5cc and the ICRU bladder point. In the case of the lightweight DL model, the F-score and AUROC were 0.90 and 0.91, respectively. (4) The DL models exhibited superior performance in predicting late bladder toxicity compared with the statistical method. Through the interpretation of the model, it further emphasized its potential for improving patient outcomes and minimizing treatment-related complications with a high level of reliability.
Open AccessFeature Importance Analysis of a Deep Learning Model for Predicting Late Bladder Toxicity Occurrence in Uterine Cervical Cancer Patients
(1) In this study, we developed a deep learning (DL) model that can be used to predict late bladder toxicity. (2) We collected data obtained from 281 uterine cervical cancer patients who underwent definitive radiation therapy. The DL model was trained using 16 features, including patient, tumor, treatment, and dose parameters, and its performance was compared with that of a multivariable logistic regression model using the following metrics: accuracy, prediction, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). In addition, permutation feature importance was calculated to interpret the DL model for each feature, and the lightweight DL model was designed to focus on the top five important features. (3) The DL model outperformed the multivariable logistic regression model on our dataset. It achieved an F1-score of 0.76 and an AUROC of 0.81, while the corresponding values for the multivariable logistic regression were 0.14 and 0.43, respectively. The DL model identified the doses for the most exposed 2 cc volume of the bladder (BD2cc) as the most important feature, followed by BD5cc and the ICRU bladder point. In the case of the lightweight DL model, the F-score and AUROC were 0.90 and 0.91, respectively. (4) The DL models exhibited superior performance in predicting late bladder toxicity compared with the statistical method. Through the interpretation of the model, it further emphasized its potential for improving patient outcomes and minimizing treatment-related complications with a high level of reliability.
- Open Access
Development of a Real-Time Pixel Array-Type Detector for Ultrahigh Dose-Rate Beams
Although research into ultrahigh dose-rate (UHDR) radiation therapy is ongoing, there is a significant lack of experimental measurements for two-dimensional (2D) dose-rate distributions. Additionally, conventional pixel-type detectors result in significant beam loss. In this study, we developed a pixel array-type detector with adjustable gaps and a data acquisition system to evaluate its effectiveness in measuring UHDR proton beams in real time. We measured a UHDR beam at the Korea Institute of Radiological and Medical Sciences using an MC-50 cyclotron, which produced a 45-MeV energy beam with a current range of 10–70 nA, to confirm the UHDR beam conditions. To minimize beam loss during measurement, we adjusted the gap and high voltage on the detector and determined the collection efficiency of the developed detector through Monte Carlo simulation and experimental measurements of the 2D dose-rate distribution. We also verified the accuracy of the real-time position measurement using the developed detector with a 226.29-MeV PBS beam at the National Cancer Center of the Republic of Korea. Our results indicate that, for a current of 70 nA with an energy beam of 45 MeV generated using the MC-50 cyclotron, the dose rate exceeded 300 Gy/s at the center of the beam, indicating UHDR conditions. Simulation and experimental measurements show that fixing the gap at 2 mm and the high voltage at 1000 V resulted in a less than 1% loss of collection efficiency when measuring UHDR beams. Furthermore, we achieved real-time measurements of the beam position with an accuracy of within 2% at five reference points. In conclusion, our study developed a beam monitoring system that can measure UHDR proton beams and confirmed the accuracy of the beam position and profile through real-time data transmission.
Open AccessDevelopment of a Real-Time Pixel Array-Type Detector for Ultrahigh Dose-Rate Beams
Although research into ultrahigh dose-rate (UHDR) radiation therapy is ongoing, there is a significant lack of experimental measurements for two-dimensional (2D) dose-rate distributions. Additionally, conventional pixel-type detectors result in significant beam loss. In this study, we developed a pixel array-type detector with adjustable gaps and a data acquisition system to evaluate its effectiveness in measuring UHDR proton beams in real time. We measured a UHDR beam at the Korea Institute of Radiological and Medical Sciences using an MC-50 cyclotron, which produced a 45-MeV energy beam with a current range of 10–70 nA, to confirm the UHDR beam conditions. To minimize beam loss during measurement, we adjusted the gap and high voltage on the detector and determined the collection efficiency of the developed detector through Monte Carlo simulation and experimental measurements of the 2D dose-rate distribution. We also verified the accuracy of the real-time position measurement using the developed detector with a 226.29-MeV PBS beam at the National Cancer Center of the Republic of Korea. Our results indicate that, for a current of 70 nA with an energy beam of 45 MeV generated using the MC-50 cyclotron, the dose rate exceeded 300 Gy/s at the center of the beam, indicating UHDR conditions. Simulation and experimental measurements show that fixing the gap at 2 mm and the high voltage at 1000 V resulted in a less than 1% loss of collection efficiency when measuring UHDR beams. Furthermore, we achieved real-time measurements of the beam position with an accuracy of within 2% at five reference points. In conclusion, our study developed a beam monitoring system that can measure UHDR proton beams and confirmed the accuracy of the beam position and profile through real-time data transmission.
- Open Access
Interobserver Variability Prediction of Primary Gross Tumor in a Patient with Non-Small Cell Lung Cancer
This research addresses the problem of interobserver variability (IOV), in which different oncologists manually delineate varying primary gross tumor volume (pGTV) contours, adding risk to targeted radiation treatments. Thus, a method of IOV reduction is urgently needed. Hypothesizing that the radiation oncologist’s IOV may shrink with the aid of IOV maps, we propose IOV prediction network (IOV-Net), a deep-learning model that uses the fuzzy membership function to produce high-quality maps based on computed tomography (CT) images. To test the prediction accuracy, a ground-truth pGTV IOV map was created using the manual contour delineations of radiation therapy structures provided by five expert oncologists. Then, we tasked IOV-Net with producing a map of its own. The mean squared error (prediction vs. ground truth) and its standard deviation were 0.0038 and 0.0005, respectively. To test the clinical feasibility of our method, CT images were divided into two groups, and oncologists from our institution created manual contours with and without IOV map guidance. The Dice similarity coefficient and Jaccard index increased by ~6 and 7%, respectively, and the Hausdorff distance decreased by 2.5 mm, indicating a statistically significant IOV reduction (p < 0.05). Hence, IOV-net and its resultant IOV maps have the potential to improve radiation therapy efficacy worldwide.
Open AccessInterobserver Variability Prediction of Primary Gross Tumor in a Patient with Non-Small Cell Lung Cancer
This research addresses the problem of interobserver variability (IOV), in which different oncologists manually delineate varying primary gross tumor volume (pGTV) contours, adding risk to targeted radiation treatments. Thus, a method of IOV reduction is urgently needed. Hypothesizing that the radiation oncologist’s IOV may shrink with the aid of IOV maps, we propose IOV prediction network (IOV-Net), a deep-learning model that uses the fuzzy membership function to produce high-quality maps based on computed tomography (CT) images. To test the prediction accuracy, a ground-truth pGTV IOV map was created using the manual contour delineations of radiation therapy structures provided by five expert oncologists. Then, we tasked IOV-Net with producing a map of its own. The mean squared error (prediction vs. ground truth) and its standard deviation were 0.0038 and 0.0005, respectively. To test the clinical feasibility of our method, CT images were divided into two groups, and oncologists from our institution created manual contours with and without IOV map guidance. The Dice similarity coefficient and Jaccard index increased by ~6 and 7%, respectively, and the Hausdorff distance decreased by 2.5 mm, indicating a statistically significant IOV reduction (p < 0.05). Hence, IOV-net and its resultant IOV maps have the potential to improve radiation therapy efficacy worldwide.
- Open Access
Assessment of a Therapeutic X-ray Radiation Dose Measurement System Based on a Flexible Copper Indium Gallium Selenide Solar Cell
Several detectors have been developed to measure radiation doses during radiotherapy. However, most detectors are not flexible. Consequently, the airgaps between the patient surface and detector could reduce the measurement accuracy. Thus, this study proposes a dose measurement system based on a flexible copper indium gallium selenide (CIGS) solar cell. Our system comprises a customized CIGS solar cell (with a size 10 × 10 cm2 and thickness 0.33 mm), voltage amplifier, data acquisition module, and laptop with in-house software. In the study, the dosimetric characteristics, such as dose linearity, dose rate independence, energy independence, and field size output, of the dose measurement system in therapeutic X-ray radiation were quantified. For dose linearity, the slope of the linear fitted curve and the R-square value were 1.00 and 0.9999, respectively. The differences in the measured signals according to changes in the dose rates and photon energies were <2% and <3%, respectively. The field size output measured using our system exhibited a substantial increase as the field size increased, contrary to that measured using the ion chamber/film. Our findings demonstrate that our system has good dosimetric characteristics as a flexible in vivo dosimeter. Furthermore, the size and shape of the solar cell can be easily customized, which is an advantage over other flexible dosimeters based on an a-Si solar cell.
Open AccessAssessment of a Therapeutic X-ray Radiation Dose Measurement System Based on a Flexible Copper Indium Gallium Selenide Solar Cell
Several detectors have been developed to measure radiation doses during radiotherapy. However, most detectors are not flexible. Consequently, the airgaps between the patient surface and detector could reduce the measurement accuracy. Thus, this study proposes a dose measurement system based on a flexible copper indium gallium selenide (CIGS) solar cell. Our system comprises a customized CIGS solar cell (with a size 10 × 10 cm2 and thickness 0.33 mm), voltage amplifier, data acquisition module, and laptop with in-house software. In the study, the dosimetric characteristics, such as dose linearity, dose rate independence, energy independence, and field size output, of the dose measurement system in therapeutic X-ray radiation were quantified. For dose linearity, the slope of the linear fitted curve and the R-square value were 1.00 and 0.9999, respectively. The differences in the measured signals according to changes in the dose rates and photon energies were <2% and <3%, respectively. The field size output measured using our system exhibited a substantial increase as the field size increased, contrary to that measured using the ion chamber/film. Our findings demonstrate that our system has good dosimetric characteristics as a flexible in vivo dosimeter. Furthermore, the size and shape of the solar cell can be easily customized, which is an advantage over other flexible dosimeters based on an a-Si solar cell.
- Open Access
Dose Profile Modulation of Proton Minibeam for Clinical Application
The feasibility of proton minibeam radiation therapy (pMBRT) using a multislit collimator (MSC) and a scattering device was evaluated for clinical use at a clinical proton therapy facility. We fabricated, through Monte Carlo (MC) simulations, not only an MSC with a high peak-to-valley dose ratio (PVDR) at the entrance of the proton beam, to prevent radiation toxicity, but also a scattering device to modulate the PVDR in depth. The slit width and center-to-center distance of the diverging MSC were 2.5 mm and 5.0 mm at the large end, respectively, and its thickness and available field size were 100 mm and 76 × 77.5 mm2, respectively. Spatially fractionated dose distributions were measured at various depths using radiochromic EBT3 films and also tested on bacterial cells. MC simulation showed that the thicker the MSC, the higher the PVDR at the phantom surface. Dosimetric evaluations showed that lateral dose profiles varied according to the scatterer’s thickness, and the depths satisfying PVDR = 1.1 moved toward the surface as their thickness increased. The response of the bacterial cells to the proton minibeams’ depth was also established, in a manner similar to the dosimetric pattern. Conclusively, these results strongly suggest that pMBRT can be implemented in clinical centers by using MSC and scatterers.
Open AccessDose Profile Modulation of Proton Minibeam for Clinical Application
The feasibility of proton minibeam radiation therapy (pMBRT) using a multislit collimator (MSC) and a scattering device was evaluated for clinical use at a clinical proton therapy facility. We fabricated, through Monte Carlo (MC) simulations, not only an MSC with a high peak-to-valley dose ratio (PVDR) at the entrance of the proton beam, to prevent radiation toxicity, but also a scattering device to modulate the PVDR in depth. The slit width and center-to-center distance of the diverging MSC were 2.5 mm and 5.0 mm at the large end, respectively, and its thickness and available field size were 100 mm and 76 × 77.5 mm2, respectively. Spatially fractionated dose distributions were measured at various depths using radiochromic EBT3 films and also tested on bacterial cells. MC simulation showed that the thicker the MSC, the higher the PVDR at the phantom surface. Dosimetric evaluations showed that lateral dose profiles varied according to the scatterer’s thickness, and the depths satisfying PVDR = 1.1 moved toward the surface as their thickness increased. The response of the bacterial cells to the proton minibeams’ depth was also established, in a manner similar to the dosimetric pattern. Conclusively, these results strongly suggest that pMBRT can be implemented in clinical centers by using MSC and scatterers.
- Open Access
- 17 August 2021
- Published by Springer Nature in Radiation Oncology
- Vol. 16 (1) , 1-13
Deep learning method for prediction of patient-specific dose distribution in breast cancer
Background: Patient-specific dose prediction improves the efficiency and quality of radiation treatment planning and reduces the time required to find the optimal plan. In this study, a patient-specific dose prediction model was developed for a left-sided breast clinical case using deep learning, and its performance was compared with that of conventional knowledge-based planning using RapidPlan™. Methods: Patient-specific dose prediction was performed using a contour image of the planning target volume (PTV) and organs at risk (OARs) with a U-net-based modified dose prediction neural network. A database of 50 volumetric modulated arc therapy (VMAT) plans for left-sided breast cancer patients was utilized to produce training and validation datasets. The dose prediction deep neural network (DpNet) feature weights of the previously learned convolution layers were applied to the test on a cohort of 10 test sets. With the same patient data set, dose prediction was performed for the 10 test sets after training in RapidPlan. The 3D dose distribution, absolute dose difference error, dose-volume histogram, 2D gamma index, and iso-dose dice similarity coefficient were used for quantitative evaluation of the dose prediction. Results: The mean absolute error (MAE) and one standard deviation (SD) between the clinical and deep learning dose prediction models were 0.02 ± 0.04%, 0.01 ± 0.83%, 0.16 ± 0.82%, 0.52 ± 0.97, − 0.88 ± 1.83%, − 1.16 ± 2.58%, and − 0.97 ± 1.73% for D95%, Dmean in the PTV, and the OARs of the body, left breast, heart, left lung, and right lung, respectively, and those measured between the clinical and RapidPlan dose prediction models were 0.02 ± 0.14%, 0.87 ± 0.63%, − 0.29 ± 0.98%, 1.30 ± 0.86%, − 0.32 ± 1.10%, 0.12 ± 2.13%, and − 1.74 ± 1.79, respectively. Conclusions: In this study, a deep learning method for dose prediction was developed and was demonstrated to accurately predict patient-specific doses for left-sided breast cancer. Using the deep learning framework, the efficiency and accuracy of the dose prediction were compared to those of RapidPlan. The doses predicted by deep learning were superior to the results of the RapidPlan-generated VMAT plan.
- 17 August 2021
- Published by Springer Nature in Radiation Oncology
- Vol. 16 (1) , 1-13
Open AccessDeep learning method for prediction of patient-specific dose distribution in breast cancer
Background: Patient-specific dose prediction improves the efficiency and quality of radiation treatment planning and reduces the time required to find the optimal plan. In this study, a patient-specific dose prediction model was developed for a left-sided breast clinical case using deep learning, and its performance was compared with that of conventional knowledge-based planning using RapidPlan™. Methods: Patient-specific dose prediction was performed using a contour image of the planning target volume (PTV) and organs at risk (OARs) with a U-net-based modified dose prediction neural network. A database of 50 volumetric modulated arc therapy (VMAT) plans for left-sided breast cancer patients was utilized to produce training and validation datasets. The dose prediction deep neural network (DpNet) feature weights of the previously learned convolution layers were applied to the test on a cohort of 10 test sets. With the same patient data set, dose prediction was performed for the 10 test sets after training in RapidPlan. The 3D dose distribution, absolute dose difference error, dose-volume histogram, 2D gamma index, and iso-dose dice similarity coefficient were used for quantitative evaluation of the dose prediction. Results: The mean absolute error (MAE) and one standard deviation (SD) between the clinical and deep learning dose prediction models were 0.02 ± 0.04%, 0.01 ± 0.83%, 0.16 ± 0.82%, 0.52 ± 0.97, − 0.88 ± 1.83%, − 1.16 ± 2.58%, and − 0.97 ± 1.73% for D95%, Dmean in the PTV, and the OARs of the body, left breast, heart, left lung, and right lung, respectively, and those measured between the clinical and RapidPlan dose prediction models were 0.02 ± 0.14%, 0.87 ± 0.63%, − 0.29 ± 0.98%, 1.30 ± 0.86%, − 0.32 ± 1.10%, 0.12 ± 2.13%, and − 1.74 ± 1.79, respectively. Conclusions: In this study, a deep learning method for dose prediction was developed and was demonstrated to accurately predict patient-specific doses for left-sided breast cancer. Using the deep learning framework, the efficiency and accuracy of the dose prediction were compared to those of RapidPlan. The doses predicted by deep learning were superior to the results of the RapidPlan-generated VMAT plan.
- Open Access
- 19 January 2021
- Published by Springer Nature in Research Square
Deep Learning Method for Prediction of Patient-Specific Dose Distribution in Breast Cancer
BackgroundPatient-specific dose prediction improves the efficiency and quality of radiation treatment planning and reduces the time required to find the optimal plan. In this study, a patient-specific dose prediction model was developed for a left-sided breast clinical case using deep learning and its performance was compared with that of conventional knowledge-based planning using RapidPlan™.MethodsPatient-specific dose prediction was performed using a contour image of the planning target volume (PTV) and organs at risk (OARs) with a U-net-based modified dose prediction neural network. A database of 50 volumetric modulated arc therapy (VMAT) plans for left-sided breast cancer patients was utilized to produce training and validation datasets. The dose prediction deep neural network (DpNet) feature weights of the previously learned convolution layers were applied to the test on a cohort of 10 test sets. With the same patient data set, dose prediction was performed for the 10 test sets after training in RapidPlan. The 3D dose distribution, absolute dose difference error, dose-volume histogram, 2D gamma index, and iso-dose dice similarity coefficient were used for quantitative evaluation of the dose prediction.ResultsThe mean absolute error (MAE) and one standard deviation (SD) between the clinical and deep learning dose prediction models were 0.02 ± 0.04%, 0.01 ± 0.83%, 0.16 ±0.82%, 0.52 ± 0.97% -0.88 ± 1.83%, -1.16 ± 2.58%, and -0.97 ± 1.73% for D95%, Dmean in the PTV, and the OARs of the body, left breast, heart, left lung, and right lung, respectively, and those measured between the clinical and RapidPlan dose prediction models were 0.02 ± 0.04%, 0.87 ± 0.63%, -0.29 ± 0.98%, 1.30 ± 0.86%, -0.32 ± 1.10%, 0.12 ± 2.13, and -1.74 ±1.79%, respectively.ConclusionsIn this study, a deep learning method for dose prediction was developed and was demonstrated to predict patient-specific doses for left-sided breast cancer accurately. Using the deep learning framework, the efficiency and accuracy of the dose prediction were compared to those of RapidPlan. The doses predicted by deep learning were superior to the results of the RapidPlan-generated VMAT plan. The proposed model will be able to maintain treatment plan quality and increase efficiency through patient dose prediction.
- 19 January 2021
- Published by Springer Nature in Research Square
Open AccessDeep Learning Method for Prediction of Patient-Specific Dose Distribution in Breast Cancer
BackgroundPatient-specific dose prediction improves the efficiency and quality of radiation treatment planning and reduces the time required to find the optimal plan. In this study, a patient-specific dose prediction model was developed for a left-sided breast clinical case using deep learning and its performance was compared with that of conventional knowledge-based planning using RapidPlan™.MethodsPatient-specific dose prediction was performed using a contour image of the planning target volume (PTV) and organs at risk (OARs) with a U-net-based modified dose prediction neural network. A database of 50 volumetric modulated arc therapy (VMAT) plans for left-sided breast cancer patients was utilized to produce training and validation datasets. The dose prediction deep neural network (DpNet) feature weights of the previously learned convolution layers were applied to the test on a cohort of 10 test sets. With the same patient data set, dose prediction was performed for the 10 test sets after training in RapidPlan. The 3D dose distribution, absolute dose difference error, dose-volume histogram, 2D gamma index, and iso-dose dice similarity coefficient were used for quantitative evaluation of the dose prediction.ResultsThe mean absolute error (MAE) and one standard deviation (SD) between the clinical and deep learning dose prediction models were 0.02 ± 0.04%, 0.01 ± 0.83%, 0.16 ±0.82%, 0.52 ± 0.97% -0.88 ± 1.83%, -1.16 ± 2.58%, and -0.97 ± 1.73% for D95%, Dmean in the PTV, and the OARs of the body, left breast, heart, left lung, and right lung, respectively, and those measured between the clinical and RapidPlan dose prediction models were 0.02 ± 0.04%, 0.87 ± 0.63%, -0.29 ± 0.98%, 1.30 ± 0.86%, -0.32 ± 1.10%, 0.12 ± 2.13, and -1.74 ±1.79%, respectively.ConclusionsIn this study, a deep learning method for dose prediction was developed and was demonstrated to predict patient-specific doses for left-sided breast cancer accurately. Using the deep learning framework, the efficiency and accuracy of the dose prediction were compared to those of RapidPlan. The doses predicted by deep learning were superior to the results of the RapidPlan-generated VMAT plan. The proposed model will be able to maintain treatment plan quality and increase efficiency through patient dose prediction.
- Open Access
- 27 November 2019
- Published by Springer Nature in Radiation Oncology
- Vol. 14 (1) , 1-13
Comparative clinical evaluation of atlas and deep-learning-based auto-segmentation of organ structures in liver cancer
Background: Accurate and standardized descriptions of organs at risk (OARs) are essential in radiation therapy for treatment planning and evaluation. Traditionally, physicians have contoured patient images manually, which, is time-consuming and subject to inter-observer variability. This study aims to a) investigate whether customized, deep-learning-based auto-segmentation could overcome the limitations of manual contouring and b) compare its performance against a typical, atlas-based auto-segmentation method organ structures in liver cancer. Methods: On-contrast computer tomography image sets of 70 liver cancer patients were used, and four OARs (heart, liver, kidney, and stomach) were manually delineated by three experienced physicians as reference structures. Atlas and deep learning auto-segmentations were respectively performed with MIM Maestro 6.5 (MIM Software Inc., Cleveland, OH) and, with a deep convolution neural network (DCNN). The Hausdorff distance (HD) and, dice similarity coefficient (DSC), volume overlap error (VOE), and relative volume difference (RVD) were used to quantitatively evaluate the four different methods in the case of the reference set of the four OAR structures. Results: The atlas-based method yielded the following average DSC and standard deviation values (SD) for the heart, liver, right kidney, left kidney, and stomach: 0.92 ± 0.04 (DSC ± SD), 0.93 ± 0.02, 0.86 ± 0.07, 0.85 ± 0.11, and 0.60 ± 0.13 respectively. The deep-learning-based method yielded corresponding values for the OARs of 0.94 ± 0.01, 0.93 ± 0.01, 0.88 ± 0.03, 0.86 ± 0.03, and 0.73 ± 0.09. The segmentation results show that the deep learning framework is superior to the atlas-based framwork except in the case of the liver. Specifically, in the case of the stomach, the DSC, VOE, and RVD showed a maximum difference of 21.67, 25.11, 28.80% respectively. Conclusions: In this study, we demonstrated that a deep learning framework could be used more effectively and efficiently compared to atlas-based auto-segmentation for most OARs in human liver cancer. Extended use of the deep-learning-based framework is anticipated for auto-segmentations of other body sites.
- 27 November 2019
- Published by Springer Nature in Radiation Oncology
- Vol. 14 (1) , 1-13
Open AccessComparative clinical evaluation of atlas and deep-learning-based auto-segmentation of organ structures in liver cancer
Background: Accurate and standardized descriptions of organs at risk (OARs) are essential in radiation therapy for treatment planning and evaluation. Traditionally, physicians have contoured patient images manually, which, is time-consuming and subject to inter-observer variability. This study aims to a) investigate whether customized, deep-learning-based auto-segmentation could overcome the limitations of manual contouring and b) compare its performance against a typical, atlas-based auto-segmentation method organ structures in liver cancer. Methods: On-contrast computer tomography image sets of 70 liver cancer patients were used, and four OARs (heart, liver, kidney, and stomach) were manually delineated by three experienced physicians as reference structures. Atlas and deep learning auto-segmentations were respectively performed with MIM Maestro 6.5 (MIM Software Inc., Cleveland, OH) and, with a deep convolution neural network (DCNN). The Hausdorff distance (HD) and, dice similarity coefficient (DSC), volume overlap error (VOE), and relative volume difference (RVD) were used to quantitatively evaluate the four different methods in the case of the reference set of the four OAR structures. Results: The atlas-based method yielded the following average DSC and standard deviation values (SD) for the heart, liver, right kidney, left kidney, and stomach: 0.92 ± 0.04 (DSC ± SD), 0.93 ± 0.02, 0.86 ± 0.07, 0.85 ± 0.11, and 0.60 ± 0.13 respectively. The deep-learning-based method yielded corresponding values for the OARs of 0.94 ± 0.01, 0.93 ± 0.01, 0.88 ± 0.03, 0.86 ± 0.03, and 0.73 ± 0.09. The segmentation results show that the deep learning framework is superior to the atlas-based framwork except in the case of the liver. Specifically, in the case of the stomach, the DSC, VOE, and RVD showed a maximum difference of 21.67, 25.11, 28.80% respectively. Conclusions: In this study, we demonstrated that a deep learning framework could be used more effectively and efficiently compared to atlas-based auto-segmentation for most OARs in human liver cancer. Extended use of the deep-learning-based framework is anticipated for auto-segmentations of other body sites.
- 24 May 2019
- Published by Morressier GmbH
Monte Carlo investigation of dose enhancement due to gold nanoparticles in particle beam therapy
Purpose: We investigated the dose enhancement effect due to gold nanoparticles in particle beam (12C, 4He, proton) irradiation. Methods: For M |
- 24 May 2019
- Published by Morressier GmbH
Monte Carlo investigation of dose enhancement due to gold nanoparticles in particle beam therapy
Purpose: We investigated the dose enhancement effect due to gold nanoparticles in particle beam (12C, 4He, proton) irradiation. Methods: For M |