Workflow efficiency for the treatment planning process in CT-guided high-dose-rate brachytherapy for cervical cancer

Neural network dose prediction for cervical brachytherapy: Overcoming data scarcity for applicator-specific models

BACKGROUND

3D neural network dose predictions are useful for automating brachytherapy (BT) treatment planning for cervical cancer. Cervical BT can be delivered with numerous applicators, which necessitates developing models that generalize to multiple applicator types. The variability and scarcity of data for any given applicator type poses challenges for deep learning.

PURPOSE

The goal of this work was to compare three methods of neural network training-a single model trained on all applicator data, fine-tuning the combined model to each applicator, and individual (IDV) applicator models-to determine the optimal method for dose prediction.

METHODS

Models were produced for four applicator types-tandem-and-ovoid (T&O), T&O with 1-7 needles (T&ON), tandem-and-ring (T&R) and T&R with 1-4 needles (T&RN). First, the combined model was trained on 859 treatment plans from 266 cervical cancer patients treated from 2010 onwards. The train/validation/test split was 70%/16%/14%, with approximately 49%/10%/19%/22% T&O/T&ON/T&R/T&RN in each dataset. Inputs included four channels for anatomical masks (high-risk clinical target volume [HRCTV], bladder, rectum, and sigmoid), a mask indicating dwell position locations, and applicator channels for each applicator component. Applicator channels were created by mapping the 3D dose for a single dwell position to each dwell position and summing over each applicator component with uniform dwell time weighting. A 3D Cascade U-Net, which consists of two U-Nets in sequence, and mean squared error loss function were used. The combined model was then fine-tuned to produce four applicator-specific models by freezing the first U-Net and encoding layers of the second and resuming training on applicator-specific data. Finally, four IDV models were trained using only data from each applicator type. Performance of these three model types was compared using the following metrics for the test set: mean error (ME, representing model bias) and mean absolute error (MAE) over all dose voxels and ME of clinical metrics (HRCTV D90% and D2cc of bladder, rectum, and sigmoid), averaged over all patients. A positive ME indicates the clinical dose was higher than predicted. 3D global gamma analysis with the prescription dose as reference value was performed. Dice similarity coefficients (DSC) were computed for each isodose volume.

RESULTS

Fine-tuned and combined models showed better performance than IDV applicator training. Fine-tuning resulted in modest improvements in about half the metrics, compared to the combined model, while the remainder were mostly unchanged. Fine-tuned MAE = 3.98%/2.69%/5.36%/3.80% for T&O/T&R/T&ON/T&RN, and ME over all voxels = -0.08%/-0.89%/-0.59%/1.42%. ME D2cc were bladder = -0.77%/1.00%/-0.66%/-1.53%, rectum = 1.11%/-0.22%/-0.29%/-3.37%, sigmoid = -0.47%/-0.06%/-2.37%/-1.40%, and ME D90 = 2.6%/-4.4%/4.8%/0.0%. Gamma pass rates (3%/3 mm) were 86%/91%/83%/89%. Mean DSCs were 0.92%/0.92%/0.88%/0.91% for isodoses ≤ 150% of prescription.

CONCLUSIONS

3D BT dose was accurately predicted for all applicator types, as indicated by the low MAE and MEs, high gamma scores and high DSCs. Training on all treatment data overcomes challenges with data scarcity in each applicator type, resulting in superior performance than can be achieved by training on IDV applicators alone. This could presumably be explained by the fact that the larger, more diverse dataset allows the neural network to learn underlying trends and characteristics in dose that are common to all treatment applicators. Accurate, applicator-specific dose predictions could enable automated, knowledge-based planning for any cervical brachytherapy treatment.

Automated treatment planning framework for brachytherapy of cervical cancer using 3D dose predictions

Objective. To lay the foundation for automated knowledge-based brachytherapy treatment planning using 3D dose estimations, we describe an optimization framework to convert brachytherapy dose distributions directly into dwell times (DTs).Approach. A dose rate kernelḋ(r,θ,φ)was produced by exporting 3D dose for one dwell position from the treatment planning system and normalizing by DT. By translating and rotating this kernel to each dwell position, scaling by DT and summing over all dwell positions, dose was computed (Dcalc). We used a Python-coded COBYLA optimizer to iteratively determine the DTs that minimize the mean squared error betweenDcalcand reference doseDref, computed using voxels withDref80%-120% of prescription. As validation of the optimization, we showed that the optimizer replicates clinical plans whenDref= clinical dose in 40 patients treated with tandem-and-ovoid (T&O) or tandem-and-ring (T&R) and 0-3 needles. Then we demonstrated automated planning in 10 T&O usingDref= dose predicted from a convolutional neural network developed in past work. Validation and automated plans were compared to clinical plans using mean absolute differences (MAD=1N∑n=1Nabsxn-xn') over all voxels (xn= Dose,N= #voxels) and DTs (xn= DT,N= #dwell positions), mean differences (MD) in organD2ccand high-risk CTV D90 over all patients (where positive indicates higher clinical dose), and mean Dice similarity coefficients (DSC) for 100% isodose contours.Main results. Validation plans agreed well with clinical plans (MADdose= 1.1%, MADDT= 4 s or 0.8% of total plan time,D2ccMD = -0.2% to 0.2% and D90 MD = -0.6%, DSC = 0.99). For automated plans, MADdose= 6.5% and MADDT= 10.3 s (2.1%). The slightly higher clinical metrics in automated plans (D2ccMD = -3.8% to 1.3% and D90 MD = -5.1%) were due to higher neural network dose predictions. The overall shape of the automated dose distributions were similar to clinical doses (DSC = 0.91).Significance. Automated planning with 3D dose predictions could provide significant time savings and standardize treatment planning across practitioners, regardless of experience.

Development of a workflow process mapping protocol to inform the implementation of regional patient navigation programs in breast oncology

BACKGROUND

Implementing city-wide patient navigation processes that support patients across the continuum of cancer care is impeded by a lack of standardized tools to integrate workflows and reduce gaps in care. The authors present an actionable workflow process mapping protocol for navigation process planning and improvement based on methods developed for the Translating Research Into Practice study.

METHODS

Key stakeholders at each study site were identified through existing community partnerships, and data on each site's navigation processes were collected using mixed methods through a series of team meetings. The authors used Health Quality Ontario's Quality Improvement Guide, service design principles, and key stakeholder input to map the collected data onto a template structured according to the case-management model.

RESULTS

Data collection and process mapping exercises resulted in a 10-step protocol that includes: 1) workflow mapping procedures to guide data collection on the series of activities performed by health care personnel that comprise a patient's navigation experience, 2) a site survey to assess program characteristics, 3) a semistructured interview guide to assess care coordination workflows, 4) a site-level swim lane workflow process mapping template, and 5) a regional high-level process mapping template to aggregate data from multiple site-level process maps.

CONCLUSIONS

This iterative, participatory approach to data collection and process mapping can be used by improvement teams to streamline care coordination, ultimately improving the design and delivery of an evidence-based navigation model that spans multiple treatment modalities and multiple health systems in a metropolitan area. This protocol is presented as an actionable toolkit so the work may be replicated to support other quality-improvement initiatives and efforts to design truly patient-centered breast cancer treatment experiences.

LAY SUMMARY

Evidence-based patient navigation in breast cancer care requires the integration of services through each phase of cancer treatment. The Translating Research Into Practice study aims to implement patient navigation for patients with breast cancer who are at risk for delays and are seeking care across 6 health systems in Boston, Massachusetts. The authors designed a 10-step protocol outlining procedures and tools that support a systematic assessment for health systems that want to implement breast cancer patient navigation services for patients who are at risk for treatment delays.

Artificial intelligence in brachytherapy: a summary of recent developments

Artificial intelligence (AI) applications, in the form of machine learning and deep learning, are being incorporated into practice in various aspects of medicine, including radiation oncology. Ample evidence from recent publications explores its utility and future use in external beam radiotherapy. However, the discussion on its role in brachytherapy is sparse. This article summarizes available current literature and discusses potential uses of AI in brachytherapy, including future directions. AI has been applied for brachytherapy procedures during almost all steps, starting from decision-making till treatment completion. AI use has led to improvement in efficiency and accuracy by reducing the human errors and saving time in certain aspects. Apart from direct use in brachytherapy, AI also contributes to contemporary advancements in radiology and associated sciences that can affect brachytherapy decisions and treatment. There is a renewal of interest in brachytherapy as a technique in recent years, contributed largely by the understanding that contemporary advances such as intensity modulated radiotherapy and stereotactic external beam radiotherapy cannot match the geometric gains and conformality of brachytherapy, and the integrated efforts of international brachytherapy societies to promote brachytherapy training and awareness. Use of AI technologies may consolidate it further by reducing human effort and time. Prospective validation over larger studies and incorporation of AI technologies for a larger patient population would help improve the efficiency and acceptance of brachytherapy. The enthusiasm favoring AI needs to be balanced against the short duration and quantum of experience with AI in limited patient subsets, need for constant learning and re-learning to train the AI algorithms, and the inevitability of humans having to take responsibility for the correctness and safety of treatments.

Automated applicator digitization for high-dose-rate cervix brachytherapy using image thresholding and density-based clustering

PURPOSE

The purpose of the study was to develop and evaluate an automated digitization algorithm for high-dose-rate cervix brachytherapy, with the goal of reducing the duration of treatment planning, staff resources, variability, and potential for human error.

METHODS

An automated digitization algorithm was developed and retrospectively evaluated using treatment planning data from 10 patients with cervix cancer who were treated with a titanium tandem and ovoids applicator set. Applicators were segmented, without human interaction, by thresholding CT images to isolate high-density voxels and assigning the voxels to applicator and nonapplicator structures using HDBSCAN, a density-based linkage clustering algorithm. The applicator contours were determined from the centroid of the clustered voxels on each image slice and written to a treatment plan file. Automated contours were evaluated against manual digitization using distance and dosimetric metrics.

RESULTS

A close agreement between automatic and manual digitization was observed. The mean magnitude of contour disagreement for 10 patients equaled 0.3 mm. Hausdorff distances were ≤1.0 mm. The applicator tip coordinates had submillimeter agreement. The median and mean dose volume histogram parameter differences were less than or equal to 1% for high-risk clinical target volume D₉₀, high-risk clinical target volume D₉₅, bladder D_2cc, rectum D_2cc, large bowel D_2cc, and small bowel D_2cc. The average execution time for the automated algorithm was less than 30 s.

CONCLUSION

The digitization of titanium tandem and ovoids applicators for high-dose-rate brachytherapy treatment planning can be automated using straightforward thresholding and clustering algorithms. The adoption of automated digitization is expected to improve the consistency of treatment plans and reduce the duration of treatment planning.

Workflow and efficiency in MRI-based high-dose-rate brachytherapy for cervical cancer in a high-volume brachytherapy center

PURPOSE

We report the clinical workflow and time required for MRI-based image-guided brachytherapy (MR-IGBT) of cervical cancer patients in a high-volume brachytherapy center with 10 years of experiences to provide a practical guideline for implementing MR-IGBT into clinical use.

METHODS AND MATERIALS

We recorded the time and workflow of each procedure step within the 40 consecutive ring and tandem applicator fractions of MR-IGBT by our multidisciplinary team. We divided the entire procedure into four sections based on where the procedure was performed: (1) applicator insertion under sedation, (2) MR imaging, (3) planning, and (4) treatment delivery. In addition, we compared the current procedure time to the initial procedure time when first implementing MR-IGBT in 2007-2008 via a retrospective review.

RESULTS

Mean total procedure time was 149.3 min (SD 17.9, ranges 112-178). The multidisciplinary team included an anesthesia team, radiologist, radiation oncologist, nurses, radiation therapists, MRI technicians, dosimetrists, and physicists. The mean procedure time and ranges for each section (min) were as follows: (1) 56.2 (28.0-103.0), (2) 31.0 (19.0-70.0), (3) 44.3 (21.0-104.0), and (4) 17.8 (9.0-34.0). Under current setting, the combined mean procedure time for MR imaging and planning was 63.2 min. In comparison, the same procedure took 137.7 min in 2007-2008 period, which was significantly longer than the current workflow (p < 0.001).

CONCLUSIONS

A skilled and dedicated multidisciplinary team is required for an efficient clinical workflow and delivery of MR-IGBT. Over the years, we have improved efficiency with clinical experience and continuous efforts in staff education.