Fully Automatic Treatment Planning for External-Beam Radiation Therapy of Locally Advanced Cervical Cancer: A Tool for Low-Resource Clinics

Purpose

The purpose of this study was to validate a fully automatic treatment planning system for conventional radiotherapy of cervical cancer. This system was developed to mitigate staff shortages in low-resource clinics.

Methods

In collaboration with hospitals in South Africa and the United States, we have developed the Radiation Planning Assistant (RPA), which includes algorithms for automating every step of planning: delineating the body contour, detecting the marked isocenter, designing the treatment-beam apertures, and optimizing the beam weights to minimize dose heterogeneity. First, we validated the RPA retrospectively on 150 planning computed tomography (CT) scans. We then tested it remotely on 14 planning CT scans at two South African hospitals. Finally, automatically planned treatment beams were clinically deployed at our institution.

Results

The automatically and manually delineated body contours agreed well (median mean surface distance, 0.6 mm; range, 0.4 to 1.9 mm). The automatically and manually detected marked isocenters agreed well (mean difference, 1.1 mm; range, 0.1 to 2.9 mm). In validating the automatically designed beam apertures, two physicians, one from our institution and one from a South African partner institution, rated 91% and 88% of plans acceptable for treatment, respectively. The use of automatically optimized beam weights reduced the maximum dose significantly (median, −1.9%; P < .001). Of the 14 plans from South Africa, 100% were rated clinically acceptable. Automatically planned treatment beams have been used for 24 patients with cervical cancer by physicians at our institution, with edits as needed, and its use is ongoing.

Conclusion

We found that fully automatic treatment planning is effective for cervical cancer radiotherapy and may provide a reliable option for low-resource clinics. Prospective studies are ongoing in the United States and are planned with partner clinics.

Automatic detection of contouring errors using convolutional neural networks

PURPOSE

To develop a head and neck normal structures autocontouring tool that could be used to automatically detect the errors in autocontours from a clinically validated autocontouring tool.

METHODS

An autocontouring tool based on convolutional neural networks (CNN) was developed for 16 normal structures of the head and neck and tested to identify the contour errors from a clinically validated multiatlas-based autocontouring system (MACS). The computed tomography (CT) scans and clinical contours from 3495 patients were semiautomatically curated and used to train and validate the CNN-based autocontouring tool. The final accuracy of the tool was evaluated by calculating the Sørensen-Dice similarity coefficients (DSC) and Hausdorff distances between the automatically generated contours and physician-drawn contours on 174 internal and 24 external CT scans. Lastly, the CNN-based tool was evaluated on 60 patients' CT scans to investigate the possibility to detect contouring failures. The contouring failures on these patients were classified as either minor or major errors. The criteria to detect contouring errors were determined by analyzing the DSC between the CNN- and MACS-based contours under two independent scenarios: (a) contours with minor errors are clinically acceptable and (b) contours with minor errors are clinically unacceptable.

RESULTS

The average DSC and Hausdorff distance of our CNN-based tool was 98.4%/1.23 cm for brain, 89.1%/0.42 cm for eyes, 86.8%/1.28 cm for mandible, 86.4%/0.88 cm for brainstem, 83.4%/0.71 cm for spinal cord, 82.7%/1.37 cm for parotids, 80.7%/1.08 cm for esophagus, 71.7%/0.39 cm for lenses, 68.6%/0.72 for optic nerves, 66.4%/0.46 cm for cochleas, and 40.7%/0.96 cm for optic chiasm. With the error detection tool, the proportions of the clinically unacceptable MACS contours that were correctly detected were 0.99/0.80 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable, respectively. The proportions of the clinically acceptable MACS contours that were correctly detected were 0.81/0.60 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable, respectively.

CONCLUSION

Our CNN-based autocontouring tool performed well on both the publically available and the internal datasets. Furthermore, our results show that CNN-based algorithms are able to identify ill-defined contours from a clinically validated and used multiatlas-based autocontouring tool. Therefore, our CNN-based tool can effectively perform automatic verification of MACS contours.

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System

The Radiation Planning Assistant (RPA) is a system developed for the fully automated creation of radiotherapy treatment plans, including volume-modulated arc therapy (VMAT) plans for patients with head/neck cancer and 4-field box plans for patients with cervical cancer. It is a combination of specially developed in-house software that uses an application programming interface to communicate with a commercial radiotherapy treatment planning system. It also interfaces with a commercial secondary dose verification software. The necessary inputs to the system are a Treatment Plan Order, approved by the radiation oncologist, and a simulation computed tomography (CT) image, approved by the radiographer. The RPA then generates a complete radiotherapy treatment plan. For the cervical cancer treatment plans, no additional user intervention is necessary until the plan is complete. For head/neck treatment plans, after the normal tissue and some of the target structures are automatically delineated on the CT image, the radiation oncologist must review the contours, making edits if necessary. They also delineate the gross tumor volume. The RPA then completes the treatment planning process, creating a VMAT plan. Finally, the completed plan must be reviewed by qualified clinical staff.

Model for Estimating Power and Downtime Effects on Teletherapy Units in Low-Resource Settings

Purpose

More than 6,500 megavoltage teletherapy units are needed worldwide, many in low-resource settings. Cobalt-60 units or linear accelerators (linacs) can fill this need. We have evaluated machine performance on the basis of patient throughput to provide insight into machine viability under various conditions in such a way that conclusions can be generalized to a vast array of clinical scenarios.

Materials and Methods

Data from patient treatment plans, peer-reviewed studies, and international organizations were combined to assess the relative patient throughput of linacs and cobalt-60 units that deliver radiotherapy with standard techniques under various power and maintenance support conditions. Data concerning the frequency and duration of power outages and downtime characteristics of the machines were used to model teletherapy operation in low-resource settings.

Results

Modeled average daily throughput was decreased for linacs because of lack of power infrastructure and for cobalt-60 units because of limited and decaying source strength. For conformal radiotherapy delivered with multileaf collimators, average daily patient throughput over 8 years of operation was equal for cobalt-60 units and linacs when an average of 1.83 hours of power outage occurred per 10-hour working day. Relative to conformal treatments delivered with multileaf collimators on the respective machines, the use of advanced techniques on linacs decreased throughput between 20% and 32% and, for cobalt machines, the need to manually place blocks reduced throughput up to 37%.

Conclusion

Our patient throughput data indicate that cobalt-60 units are generally best suited for implementation when machine operation might be 70% or less of total operable time because of power outages or mechanical repair. However, each implementation scenario is unique and requires consideration of all variables affecting implementation.

Multichannel scattering calculations using absorbing potentials and mapped grids

The authors investigate the use of absorbing potentials and discrete variable representation grid methods in multichannel time-independent scattering calculations. An exactly solvable, coupled-two-channel problem involving square-well potentials is used to assess the quality of numerical results. Special emphasis is given to the description of scattering resonances and near-threshold regions. Numerical treatment of close vicinities of thresholds requires the introduction of nonequidistant grids through a mapping procedure of the scattering coordinate.

Calculations of partial cross sections for photofragmentation processes using complex absorbing potentials

We investigate the use of complex absorbing potentials for the calculation of partial cross sections in multichannel photofragmentation processes. An exactly solvable, coupled-two-channel problem involving square-well potentials is used to compare the performance of various types of absorbing potentials. Special emphasis is given to the near-threshold regions and the conditions under which the numerical results are able to reproduce the Wigner threshold laws. It was found that singular, transmission-free absorbing potentials perform better than those of power or polynomial form.

Calculations of near-threshold cross sections for photodissociation of CH+ using the Lanczos algorithm

We combine the Lanczos algorithm with the absorbing-potential method, implemented in a discrete variable representation to calculate the near-threshold photodissociation cross sections of CH+. The method is iterative, based on a continued fraction representation of the Green function and avoids any explicit matrix diagonalization. A very good agreement is found with experiment and close-coupling calculations.

Requirement of the spindle checkpoint for proper chromosome segregation in budding yeast meiosis

The spindle checkpoint was characterized in meiosis of budding yeast. In the absence of the checkpoint, the frequency of meiosis I missegregation increased with increasing chromosome length, reaching 19% for the longest chromosome. Meiosis I nondisjunction in spindle checkpoint mutants could be prevented by delaying the onset of anaphase. In a recombination-defective mutant (spo11Delta), the checkpoint delays the biochemical events of anaphase I, suggesting that chromosomes that are attached to microtubules but are not under tension can activate the spindle checkpoint. Spindle checkpoint mutants reduce the accuracy of chromosome segregation in meiosis I much more than that in meiosis II, suggesting that checkpoint defects may contribute to Down syndrome.

Direct cloning of DNA sequences from the common fragile site region at chromosome band 3p14.2

Despite several lines of evidence suggesting that common chromosomal fragile sites are biologically important as hot spots for recombination, their structure remains unknown. We showed previously that the plasmid pSV2neo preferentially integrates into bands containing fragile sites in cells transfected under conditions of fragile site induction. Here we report the isolation and characterization of the DNA sequences from two such independent integrations into 3p14.2, a common fragile site (FRA3B). These FRA3B region sequences were shown to lie within a 1330-kb YAC, 850A6, approximately 350 kb telomeric of the breakpoint of t(3;8), a constitutional rearrangement. The two integration sites are 10 kb apart, but each integration is associated with a deletion. We have constructed a partial genomic contig of the integration sites and deleted regions spanning approximately 85 kb. Analysis of the DNA sequences immediately surrounding the plasmid integrations revealed no known coding sequences or repeat structures resembling the (CGG)n motif characteristic of the rare fragile sites. In addition, by Southern blotting analysis, none of the phage clones isolated from the FRA3B region were found to contain CGG repeats. Fluorescence in situ hybridization analysis of genomic clones from this contig to metaphase cells induced to express breaks demonstrated hybridization adjoining the chromosome breaks, and occasionally the hybridization signal spanned the break. The results imply that breakage occurs at variable positions within a large region (at least on the order of 85 kb). Together, these data suggest that the structure of FRA3B differs from that of rare fragile sites.