Autodelineation of organ at risk in head and neck cancer radiotherapy using artificial intelligence.

Aim: The aim of this study is to check the practical feasibility of artificial intelligence for day-to-day operations and how it generalizes when the data have considerable interobserver variability. Background: Automated delineation of organ at risk (OAR) using a deep learning model is reasonably accurate. This will considerably reduce the medical professional time in manually contouring the OAR and also reduce the interobserver variation among radiation oncologists. It allows for quick radiation planning which helps in adaptive radiotherapy planning. Materials and Methods: Head and neck (HN) computed tomography (CT) scan data of 113 patients were used in this study. CT scan was done as per the institute protocol. Each patient had about 100-300 slices in Dicom format. A total number of 19,240 images were used as the data set. The OARs were delineated by the radiation oncologist in the contouring system. Of the 113 patient records, 13 records were kept aside as test dataset and the remaining 100 records were used for training the UNet 2D model. The study was performed on the spinal cord and left and right parotids as OARs on HN CT images. The model performance was quantified using the Dice similarity coefficient (DSC) score. Results: The trained model is used to predict three OARs, spinal cord and left and right parotids. The DSC score of 84% and above could be achieved using the UNet 2D Convolutional Neural Network. Conclusion: This study showed that the accuracy of predicted organs was within acceptable DSC scores, even when the underlying dataset has significant interobserver variability.

Assessment of Knowledge-Based Planning Model in Combination with Multi-Criteria Optimization in Head-and-Neck Cancers.

Aim: The aim of this study was to build knowledge-based planning model (KBPM) for head-and-neck (HN) cancers using volumetric-modulated arc therapy (VMAT), optimized with multi-criteria optimization (MCO), and to evaluate KBPM plan quality with clinical plan (CP) using in-house developed Python script. Materials and Methods: Two hundred previously treated simultaneously integrated boost (SIB) HN VMAT plans (RapidArc®) were selected for creating KBPM. These plans were further optimized using MCO to strike right trade-off between target and organs at risk (OARs). The script was written using Python V3.7.1 to automatically extract and analyze treatment plan dosimetric parameters through Eclipse Scripting Application Programming Interface (ESAPI). Analyzed plans that met deliverable quality were modeled using regression-based KBPM framework. The trained model is validated with 35 cohorts of HN SIB patients. Results: MCO plans were able to improve the OAR sparing without compromising target coverage compared to user-optimized CPs except for increased heterogeneity. With MCO, spinal cord dose D0.03cc is reduced by 3.2 Gy ± 1.8 Gy, parotid mean dose by 2 Gy ± 1.7 Gy compared to CPs, respectively. MCO-based KBPM plans were comparable to CP with improved sparing for left and right parotids by 11.5% and 7.8%, respectively. Conclusion: MCO-based KBPM plans were superior to user plans in terms of OAR sparing and user need to spend more time to meet the model-based plan outcomes. Created KBPM planning is simple and efficient to generate estimate for OAR sparing to guide entry and intermittent planners to improve their clinical planning skills with lesser planning time. Python ESAPI is a powerful tool to extract plan parameters and quickly evaluate either individual or a cohort of plans.

Treatment planning methods in high dose rate interstitial brachytherapy of carcinoma cervix: a dosimetric and radiobiological analysis.

Treatment planning is a trial and error process that determines optimal dwell times, dose distribution, and loading pattern for high dose rate brachytherapy. Planning systems offer a number of dose calculation methods to either normalize or optimize the radiation dose. Each method has its own characteristics for achieving therapeutic dose to mitigate cancer growth without harming contiguous normal tissues. Our aim is to propose the best suited method for planning interstitial brachytherapy. 40 cervical cancer patients were randomly selected and 5 planning methods were iterated. Graphical optimization was compared with implant geometry and dose point normalization/optimization techniques using dosimetrical and radiobiological plan quality indices retrospectively. Mean tumor control probability was similar in all the methods with no statistical significance. Mean normal tissue complication probability for bladder and rectum is 0.3252 and 0.3126 (P = 0.0001), respectively, in graphical optimized plans compared to other methods. There was no significant correlation found between Conformity Index and tumor control probability when the plans were ranked according to Pearson product moment method (r = -0.120). Graphical optimization can result in maximum sparing of normal tissues.

Use of flattening filter-free photon beams in treating medulloblastoma: a dosimetric evaluation.

Aim. To evaluate the dosimetric benefits of flattening filter-free (FFF) photon beams in intensity modulated radiation therapy (IMRT) and Rapid Arc (RA) over conventional CSI methods. Methods and Materials. Five patients treated with IMRT using static multileaf collimators (MLC) were randomly selected for this retrospective study. Dynamic MLC IMRT, RA, and conformal therapy (3DCRT) were iterated with the same CT data sets with and without flattening filter photons. Total dose prescribed was 28.80 Gy in 16 fractions. Dosimetric parameters such as D max⁡, D min⁡, D mean, V 95%, V 107%, DHI, and CI for PTV and D max⁡, D mean, V 80%, V 50%, V 30%, and V 10% for OARs were extracted from DVHs. Beam on time (BOT) for various plans was also compared. Results. FFF RA therapy (6F_RA) resulted in highly homogeneous and conformal doses throughout the craniospinal axis. 3DCRT resulted in the highest V 107% (SD) 46.97 ± 28.6, whereas flattening filter (FF) and FFF dynamic IMRT had a minimum V 107%. 6F_RA and 6F_DMLC resulted in lesser doses to thyroid, eyes, esophagus, liver, lungs, and kidneys. Conclusion. FFF IMRT and FFF RA for CSI have definite dosimetric advantages over 3DCRT technique in terms of target coverage and OAR sparing. Use of FFF in IMRT resulted in 50% reduction in BOT, thereby increasing the treatment efficiency.

Quantification of uncertainties in conventional plan evaluation methods in Intensity Modulated Radiation Therapy.

PURPOSE: In Intensity Modulated Radiation Therapy (IMRT) dose distributions tend to be more complex and heterogeneous because of the modulated fluences in each beamlet of every single beam. These dose-volume (DV) parameters derived from the dose volume histogram (DVH) are physical quantities, thought to correlate with the biological response of the tissues. The aim of this study was to quantify the uncertainty of physical dose metrics to predict clinical outcomes of radiotherapy. METHODS: The radiobiological estimates such as tumor control probability (TCP) and Normal Tissue Complication Probability (NTCP) were made for a cohort of 40 cancer patients (10 brain;19 head & neck;11 cervix) using the DV parameters. Statistical analysis was performed to determine the correlation of physical plan quality indicators with radiobiological estimates. RESULTS: The correlation between conformity index (CI) and TCP was found to be good and the dosimetric parameters for optic nerves, optic chiasm, brain stem, normal brain and parotids correlated well with the NTCP estimates. A follow up study (median duration 18 months) was also performed. There was no grade 3 or 4 normal tissue complications observed. Local tumor control was found to be higher in brain (90%) and pelvic cases (95%), whereas a decline of 70% was noted with head & neck cancer cases. CONCLUSIONS: The equivalent uniform dose (EUD) concept of radiobiological model used in the software determines TCP and NTCP values which can predict outcomes precisely using DV data in the voxel level. The uncertainty of using physical dose metrics for plan evaluation is quantified with the statistical analysis. Radiobiological evaluation is helpful in ranking the rival treatment plans also.

A brachytherapy plan evaluation tool for interstitial applications.

Radiobiological metrics such as tumor control probability (TCP) and normal tissue complication probability (NTCP) help in assessing the quality of brachytherapy plans. Application of such metrics in clinics as well as research is still inadequate. This study presents the implementation of two indigenously designed plan evaluation modules: Brachy_TCP and Brachy_NTCP. Evaluation tools were constructed to compute TCP and NTCP from dose volume histograms (DVHs) of any interstitial brachytherapy treatment plan. The computation module was employed to estimate probabilities of tumor control and normal tissue complications in ten cervical cancer patients based on biologically effective equivalent uniform dose (BEEUD). The tumor control and normal tissue morbidity were assessed with clinical followup and were scored. The acute toxicity was graded using common terminology criteria for adverse events (CTCAE) version 4.0. Outcome score was found to be correlated with the TCP/NTCP estimates. Thus, the predictive ability of the estimates was quantified with the clinical outcomes. Biologically effective equivalent uniform dose-based formalism was found to be effective in predicting the complexities and disease control.