Dataset Information

Development of a Machine Learning Model for Optimal Applicator Selection in High-Dose-Rate Cervical Brachytherapy.

ABSTRACT:

Purpose

To develop and validate a preliminary machine learning (ML) model aiding in the selection of intracavitary (IC) versus hybrid interstitial (IS) applicators for high-dose-rate (HDR) cervical brachytherapy.

Methods

From a dataset of 233 treatments using IC or IS applicators, a set of geometric features of the structure set were extracted, including the volumes of OARs (bladder, rectum, sigmoid colon) and HR-CTV, proximity of OARs to the HR-CTV, mean and maximum lateral and vertical HR-CTV extent, and offset of the HR-CTV centre-of-mass from the applicator tandem axis. Feature selection using an ANOVA F-test and mutual information removed uninformative features from this set. Twelve classification algorithms were trained and tested over 100 iterations to determine the highest performing individual models through nested 5-fold cross-validation. Three models with the highest accuracy were combined using soft voting to form the final model. This model was trained and tested over 1,000 iterations, during which the relative importance of each feature in the applicator selection process was determined.

Results

Feature selection indicated that the mean and maximum lateral and vertical extent, volume, and axis offset of the HR-CTV were the most informative features and were thus provided to the ML models. Relative feature importances indicated that the HR-CTV volume and mean lateral extent were most important for applicator selection. From the comparison of the individual classification algorithms, it was found that the highest performing algorithms were tree-based ensemble methods - AdaBoost Classifier (ABC), Gradient Boosting Classifier (GBC), and Random Forest Classifier (RFC). The accuracy of the individual models was compared to the voting model for 100 iterations (ABC = 91.6 ± 3.1%, GBC = 90.4 ± 4.1%, RFC = 89.5 ± 4.0%, Voting Model = 92.2 ± 1.8%) and the voting model was found to have superior accuracy. Over the final 1,000 evaluation iterations, the final voting model demonstrated a high predictive accuracy (91.5 ± 0.9%) and F1 Score (90.6 ± 1.1%).

Conclusion

The presented model demonstrates high discriminative performance, highlighting the potential for utilization in informing applicator selection prospectively following further clinical validation.

SUBMITTER: Stenhouse K

PROVIDER: S-EPMC7973285 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposeHigh-dose-rate (HDR) prostate brachytherapy involves treatment catheter placement, which is currently empirical and physician dependent. The lack of proper catheter placement guidance during the procedure has left the physicians to rely on a heuristic thinking-while-doing technique, which may cause large catheter placement variation and increased plan quality uncertainty. Therefore, the achievable dose distribution could not be quantified prior to the catheter placement. To overcome this challenge, we proposed a learning-based method to provide HDR catheter placement guidance for prostate cancer patients undergoing HDR brachytherapy.MethodsThe proposed framework consists of deformable registration via registration network (Reg-Net), multi-atlas ranking, and catheter regression. To model the global spatial relationship among multiple organs, binary masks of the prostate and organs-at-risk are transformed into distance maps, which describe the distance of each local voxel to the organ surfaces. For a new patient, the generated distance map is used as fixed image. Reg-Net is utilized to deformably register the distance maps from multi-atlas set to match this patient's distance map and then bring catheter maps from multi-atlas to this patient via spatial transformation. Several criteria, namely prostate volume similarity, multi-organ semantic image similarity, and catheter position criteria (far from the urethra and within the partial prostate), are used for multi-atlas ranking. The top-ranked atlas' deformed catheter positions are selected as the predicted catheter positions for this patient. Finally, catheter regression is used to refine the final catheter positions. A retrospective study on 90 patients with a fivefold cross-validation scheme was used to evaluate the proposed method's feasibility. In order to investigate the impact of plan quality from the predicted catheter pattern, we optimized the source dwell position and time for both the clinical catheter pattern and predicted catheter pattern with the same optimization settings. Comparisons of clinically relevant dose volume histogram (DVH) metrics were completed.ResultsFor all patients, on average, both the clinical plan dose and predicted plan dose meet the common dose constraints when prostate dose coverage is kept at V100 = 95%. The plans from the predicted catheter pattern have slightly higher hotspot in terms of V150 by 5.0% and V200 by 2.9% on average. For bladder V75, rectum V75, and urethra V125, the average difference is close to zero, and the range of most patients is within ±1 cc.ConclusionWe developed a new catheter placement prediction method for HDR prostate brachytherapy based on a deep-learning-based multi-atlas registration algorithm. It has great clinical potential since it can provide catheter location estimation prior to catheter placement, which could reduce the dependence on physicians' experience in catheter implantation and improve the quality of prostate HDR treatment plans. This approach merits further clinical evaluation and validation as a method of quality control for HDR prostate brachytherapy.

Project description:To compare the outcomes of localized prostate cancer treatment with high-dose-rate brachytherapy (HDR-BT) and low-dose-rate brachytherapy (LDR-BT), we examined 924 patients treated with HDR-BT + external beam radiotherapy (EBRT) and 500 patients treated with LDR-BT ± EBRT using multi-institutional retrospective data. The HDR-BT treated advanced disease with more hormonal therapy than LDR-BT. To reduce background selection bias, we performed inverse probability of treatment weighting (IPTW) analysis using propensity scores and excluded patients with T3b-4 disease/ initial prostate-specific antigen (PSA) levels > 50 ng/ml. The actuarial 5-year biochemical control rates (5y-bNED) were 96.3% and 95.7% in the HDR-BT and LDR-BT groups, respectively. The corresponding values were 100% and 96.5% in the low-risk group; 97.4% and 97.1% in the intermediate-risk group (97.2% and 97% in the higher titer group and 97.5% and 94.6% in the lower titer group, respectively); and 95.7% and 94.9% in the selected high-risk group, respectively. IPTW correction indicated no significant difference among the groups. The 5y-bNED in the HDR-BT + EBRT, LDR-BT + EBRT, and LDR-BT alone groups were 96.3%, 95.5%, and 97%, respectively (P = 0.3011). The corresponding values were 97.4%, 94.7%, and 96.6% (P = 0.1004) in the intermediate-risk group (97.5%, 100%, and 94.5% in the lower titer group [P = 0.122] and 97.2%, 96.2%, and 100% [P = 0.664] in the higher titer group, respectively) and 95.7%, 95.5%, and 100% (P = 0.859) in the high-risk group, respectively. The HDR-BT group showed a lower incidence of acute grade ≥ 2 genitourinary toxicities; the incidence of other early and late grade ≥ 2 toxicities were similar between the HDR-BT and LDR-BT groups. Acute genitourinary toxicity predicted the occurrence of late genitourinary toxicity. EBRT increased the risk of grade ≥ 2 gastrointestinal toxicity. HDR-BT + EBRT is a good alternative to LDR-BT ± EBRT for low-, intermediate-, and selected high-risk patients.

Project description:BACKGROUND:The objective of focal brachytherapy (BT) is to provide effective prostate cancer control for low-risk disease but with reduced genitourinary, gastrointestinal and sexual side effects in a cost-effective way. OBJECTIVE:The aim of this study is to describe a phase II study examining technical and dosimetric feasibility and toxicity, quality of life changes, and local control with post-treatment biopsy outcomes in men with early stage low volume prostate cancer treated with focal iodine-125 seed BT. METHODS:The study design is a prospective, multicenter trial with a planned sample size of 20 patients including men with a minimum age of 60 years, a life expectancy estimated to be greater than 10 years, with low or low-tier intermediate risk prostate cancer, unilateral disease on the biopsy, and a Gleason score of ?3+4 and <25% cores involved. The investigations specific for the study are multi-parametric magnetic resonance imaging (Mp-MRI) baseline, at 20 and 36 months to rule out high grade disease and a transperineal mapping biopsy (baseline and at 36 months) for more accurate patient selection. The hemigland region will receive 144 Gy. Standard normal tissue constraints will be considered as for a whole gland (WG) implant. Dosimetric parameters will be evaluated at day 30 after the implant. Toxicity and quality of life will be evaluated with international validated questionnaires focusing on urinary, rectal, sexual domain, and general health-related quality of life. The patients will complete this assessment at baseline and then approximately every 6 months after the implant up to 10 years. RESULTS:To date, one patient is involved in the trial. He underwent the pre-implant investigations which found bilateral disease. Therefore, a standard seed implant was performed. If the results from this trial provide evidence that the treatment is safe, feasible, and improves toxicity, funding will be sought to conduct a large, multicenter, randomized controlled trial (RCT). CONCLUSIONS:This protocol is designed to show feasibility in delivering hemigland focal therapy with seed BT. It may answer crucial questions and obtain data which will enable downstream decisions on focal low dose rate (LDR) prostate BT. CLINICALTRIAL:Clinicaltrial.gov NCT02643511; https://www.clinicaltrials.gov/ct2/show/NCT02643511 (Archived by Webcite at http://www.webcitation.org/6ghLCzIhY).