A knowledge-based intensity-modulated radiation therapy treatment planning technique for locally advanced nasopharyngeal carcinoma radiotherapy

Abstract BackgroundTo investigate the feasibility of a knowledge-based automated intensity-modulated radiation therapy (IMRT) planning technique for locally advanced nasopharyngeal carcinoma (NPC) radiotherapy.Methods140 NPC patients treated with definitive radiation therapy with the step-and-shoot IMRT techniques were retrospectively selected and separated into a knowledge library (n=115) and a test library (n=25). For each patient in the knowledge library, the overlap volume histogram (OVH), target volume histogram (TVH) and dose objectives were extracted from the manually generated plan. 5-fold cross validation was performed to divide the patients in the knowledge library into 5 groups before validating one group by using the other 4 groups to train each neural network (NN) machine learning models. For patients in the test library, their OVH and TVH were then used by the trained models to predict a corresponding set of mean dose objectives, which were subsequently used to generate automated plans (APs) in Pinnacle planning system via an in-house developed automated scripting system. All APs were obtained after a single step of optimization. Manual plans (MPs) for the test patients were generated by an experienced medical physicist strictly following the established clinical protocols. The qualities of the APs and MPs were evaluated by an attending radiation oncologist. The dosimetric parameters for planning target volume (PTV) coverage and the organs-at-risk (OAR) sparing were also quantitatively measured and compared using Mann-Whitney U test and Bonferroni correction.ResultsAPs and MPs had the same rating for more than 80% of the patients (19 out of 25) in the test group. Both AP and MP achieved PTV coverage criteria for no less than 80% of the patients. For each OAR, the number of APs achieving its criterion was similar to that in the MPs. The AP approach improved planning efficiency by greatly reducing the planning duration to about 17% of the MP (9.85±1.13 min vs. 57.10±6.35 min).ConclusionA robust and effective knowledge-based IMRT treatment planning technique for locally advanced NPC is developed. Patient specific dose objectives can be predicted by trained NN models based on the individual’s OVH and clinical TVH goals. The automated planning scripts can use these dose objectives to efficiently generate APs with largely shortened planning time. These APs had comparable dosimetric qualities when compared to our clinic’s manual plans.

Download Full-text

A knowledge-based intensity-modulated radiation therapy treatment planning technique for locally advanced nasopharyngeal carcinoma radiotherapy

10.21203/rs.3.rs-19262/v2 ◽

2020 ◽

Author(s):

Penggang Bai ◽

Xing Weng ◽

Kerun Quan ◽

Jihong Chen ◽

Yitao Dai ◽

...

Keyword(s):

Radiation Therapy ◽

Nasopharyngeal Carcinoma ◽

Treatment Planning ◽

Locally Advanced ◽

Intensity Modulated Radiation Therapy ◽

Target Volume ◽

Planning System ◽

Intensity Modulated ◽

Knowledge Based ◽

Planning Technique

Abstract BackgroundTo investigate the feasibility of a knowledge-based automated intensity-modulated radiation therapy (IMRT) planning technique for locally advanced nasopharyngeal carcinoma (NPC) radiotherapy.Methods140 NPC patients treated with definitive radiation therapy with the step-and-shoot IMRT techniques were retrospectively selected and seperated into a knowledge library (n=115) and a test library (n=25). For each case, in the knowledge library, the patient’s overlap volume histogram (OVH), target volume histogram (TVH) and dose objectives were extracted from the manually generated plan to train a 3-layer neural network (NN) machine learning model. For patients in the test library, their OVH and TVH were then used by the trained model to predict a corresponding set of dose objectives, which were subsequently used to generate automated plans (APs) in Pinnacle planning system via an in-house developed automated scripting system. All APs were obtained after a single step of optimization. Manual plans (MPs) of the same test patients were generated by an experienced medical physicist strictly following the established clinical protocols. The qualities of the APs and MPs were evaluated by an attending radiation oncologist. The dosimetric parameters for planning target volume (PTV) coverage and the organs-at-risk (OAR) sparing were also quantitatively measured and compared.ResultsAPs and MPs had the same rating for more than 80% of the patients (19 out of 25) in the test group. For greater than 80% of the patients, both AP and MP achieved PTV coverage criteria. For each OAR, the number of APs achieving its criterion was similar to that in the MPs. The AP approach significantly improved planning efficiency by reducing the planning duration to about 17% of the MP (9.73±1.80 min vs. 57.10±6.35 min, P<0.001). ConclusionA robust and effective knowledge-based IMRT treatment planning technique for locally advanced NPC is developed. Patient specific dose objectives can be predicted by a trained NN model based on the individual’s OVH and clinical TVH goals. The automated planning scripts can use these dose objectives to efficiently generate APs with largely shortened planning time. These APs had comparable dosimetric qualities when compared to our clinic’s manual plans.

Download Full-text

A knowledge-based intensity-modulated radiation therapy treatment planning technique for locally advanced nasopharyngeal carcinoma radiotherapy

10.21203/rs.3.rs-19262/v1 ◽

2020 ◽

Author(s):

Penggang Bai ◽

Xing Weng ◽

Kerun Quan ◽

Jihong Chen ◽

Yitao Dai ◽

...

Keyword(s):

Radiation Therapy ◽

Nasopharyngeal Carcinoma ◽

Treatment Planning ◽

Treatment Plan ◽

Locally Advanced ◽

Intensity Modulated Radiation Therapy ◽

Patient Specific ◽

Intensity Modulated ◽

Knowledge Based ◽

Planning Technique

Abstract Background To investigate the feasibility of a knowledge-based automated intensity-modulated radiation therapy (IMRT) planning technique for locally advanced nasopharyngeal carcinoma (NPC) radiotherapy. Methods 140 NPC patients treated with definitive radiation therapy with the step-and-shoot IMRT techniques were retrospectively selected and consisted of a knowledge library (115 patients) and a test library (25 patients). For each patient in the knowledge library, their overlap volume histogram (OVH), target volume histogram (TVH) and dose objectives were extracted from the patient’s manual treatment plan, and these were used to train a 3 layer neural network (NN) model. The OVH and TVH from the test library were input into the trained model to derive patients’ dose objectives which were subsequently used to generate automated plans (APs) by an in-house developed Perl and HotScripts planning scripts with a single iteration optimization. The corresponding manual plans (MPs) of patients in the test library were manually generated by an experienced medical physicist according to clinical protocols. Plan quality was ranked and dosimetric parameters were compared between the APs and MPs. Results Qualitatively, the APs and MPs had the same rank for the majority of the patients (19 of 25). PTV achieved each given criteria in the majority of the patients (greater than 80%) between both the APs and MPs. For each OAR, the number achieving its criterion in the APs was close to that in the MPs. The APs would improve the treatment delivery efficiency by reducing total plan MUs by ~5% (685.24±58.89 vs. 721.36±63.36, P =0.004). AP also significantly improved planning efficiency by reducing the planning duration to ~17% of the MP (9.73±1.80 min vs. 57.10±6.35 min, P <0.001). Conclusion A robust and effective knowledge-based IMRT treatment planning technique for locally advanced NPC is developed. Patient specific TDV can be predicted by a trained NN model based on the individual’s OVH and clinical TVH goals. The automated planning scripts can use these TDV to generate APs with largely shortened planning time with comparable or improved dosimetric qualities compared to our clinic’s manual plans.

Download Full-text