Predicting which patients may benefit from the hybrid intracavitary+interstitial needle (IC/IS) applicator for advanced cervical cancer: A dosimetric comparison and toxicity benefit analysis.

PURPOSE: The purpose of this study is to compare the predicted rate of local control and bladder and rectum toxicity rates for image-guided adaptive brachytherapy plans using a tandem and ovoid (T/O) applicator versus using a simulated hybrid intracavitary/interstitial tandem and ring applicator with needles (T/R + N) for patients with locally advanced cervical cancer (LACC). METHODS AND MATERIALS: Patients with ≥ FIGO Stage IIB locally advanced cervical cancer treated with T/O from a single institution were included. Simulated treatment plans were created with a T/R + N applicator for the best high-risk clinical target volume (CTV) coverage and minimal dose to organs at risk. Three-year local control rate was estimated using published dose-volume effect relationships. Next, the high-risk CTV EQD2 D90 of T/R + N plans were calculated, and bladder and rectum toxicity rates were estimated. Analysis was performed in subpatient groups defined based on tumor volume and ratio of maximal and minimal tumor radii (RR) that reflects tumor shape asymmetry. RESULTS: Improvements in predicted local control rate for the T/R + N were 0.8, 4.1, 1.6, and 3.9% for groups with tumor volume <35 cc, ≥35 cc, RR < 2.0, and ≥2.0, respectively, with the latter three being statistically significant. Predicted reductions in Grade 2-4 toxicity rates of bladder and rectum were significant in all groups except bladder toxicity in tumor volume <35 cc, when T/R + N plans were normalized to the same CTV coverage as the T/O plans. Comparing unnormalized T/R + N plans and T/O plans, predicted toxicity reductions were significant in all groups except rectum toxicity in RR ≥ 2.0. Predicted reduction of toxicity rate was larger for patients with large tumor or large tumor RR, although some reductions were relatively small. CONCLUSIONS: Cases with large tumor (volume ≥35 cc) or large tumor asymmetry (RR ≥ 2.0) would probably benefit more from the use of hybrid applicators.

Patch-based generative adversarial neural network models for head and neck MR-only planning.

PURPOSE: To evaluate pix2pix and CycleGAN and to assess the effects of multiple combination strategies on accuracy for patch-based synthetic computed tomography (sCT) generation for magnetic resonance (MR)-only treatment planning in head and neck (HN) cancer patients. MATERIALS AND METHODS: Twenty-three deformably registered pairs of CT and mDixon FFE MR datasets from HN cancer patients treated at our institution were retrospectively analyzed to evaluate patch-based sCT accuracy via the pix2pix and CycleGAN models. To test effects of overlapping sCT patches on estimations, we (a) trained the models for three orthogonal views to observe the effects of spatial context, (b) we increased effective set size by using per-epoch data augmentation, and (c) we evaluated the performance of three different approaches for combining overlapping Hounsfield unit (HU) estimations for varied patch overlap parameters. Twelve of twenty-three cases corresponded to a curated dataset previously used for atlas-based sCT generation and were used for training with leave-two-out cross-validation. Eight cases were used for independent testing and included previously unseen image features such as fused vertebrae, a small protruding bone, and tumors large enough to deform normal body contours. We analyzed the impact of MR image preprocessing including histogram standardization and intensity clipping on sCT generation accuracy. Effects of mDixon contrast (in-phase vs water) differences were tested with three additional cases. The sCT generation accuracy was evaluated using mean absolute error (MAE) and mean error (ME) in HU between the plan CT and sCT images. Dosimetric accuracy was evaluated for all clinically relevant structures in the independent testing set and digitally reconstructed radiographs (DRRs) were evaluated with respect to the plan CT images. RESULTS: The cross-validated MAEs for the whole-HN region using pix2pix and CycleGAN were 66.9 ± 7.3 vs 82.3 ± 6.4 HU, respectively. On the independent testing set with additional artifacts and previously unseen image features, whole-HN region MAEs were 94.0 ± 10.6 and 102.9 ± 14.7 HU for pix2pix and CycleGAN, respectively. For patients with different tissue contrast (water mDixon MR images), the MAEs increased to 122.1 ± 6.3 and 132.8 ± 5.5 HU for pix2pix and CycleGAN, respectively. Our results suggest that combining overlapping sCT estimations at each voxel reduced both MAE and ME compared to single-view non-overlapping patch results. Absolute percent mean/max dose errors were 2% or less for the PTV and all clinically relevant structures in our independent testing set, including structures with image artifacts. Quantitative DRR comparison between planning CTs and sCTs showed agreement of bony region positions to <1 mm. CONCLUSIONS: The dosimetric and MAE based accuracy, along with the similarity between DRRs from sCTs, indicate that pix2pix and CycleGAN are promising methods for MR-only treatment planning for HN cancer. Our methods investigated for overlapping patch-based HU estimations also indicate that combining transformation estimations of overlapping patches is a potential method to reduce generation errors while also providing a tool to potentially estimate the MR to CT aleatoric model transformation uncertainty. However, because of small patient sample sizes, further studies are required.

Deep-learning-assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer.

PURPOSE: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning-assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. METHODS AND MATERIALS: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. RESULTS: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. CONCLUSIONS: We have developed a deep-learning-assisted applicator digitization tool for 3D CT image-based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.

Intelligent inverse treatment planning via deep reinforcement learning, a proof-of-principle study in high dose-rate brachytherapy for cervical cancer.

Inverse treatment planning in radiation therapy is formulated as solving optimization problems. The objective function and constraints consist of multiple terms designed for different clinical and practical considerations. Weighting factors of these terms are needed to define the optimization problem. While a treatment planning optimization engine can solve the optimization problem with given weights, adjusting the weights to yield a high-quality plan is typically performed by a human planner. Yet the weight-tuning task is labor intensive, time consuming, and it critically affects the final plan quality. An automatic weight-tuning approach is strongly desired. The procedure of weight adjustment to improve the plan quality is essentially a decision-making problem. Motivated by the tremendous success in deep learning for decision making with human-level intelligence, we propose a novel framework to adjust the weights in a human-like manner. This study used inverse treatment planning in high-dose-rate brachytherapy (HDRBT) for cervical cancer as an example. We developed a weight-tuning policy network (WTPN) that observes dose volume histograms of a plan and outputs an action to adjust organ weighting factors, similar to the behaviors of a human planner. We trained the WTPN via end-to-end deep reinforcement learning. Experience replay was performed with the epsilon greedy algorithm. After training was completed, we applied the trained WTPN to guide treatment planning of five testing patient cases. It was found that the trained WTPN successfully learnt the treatment planning goals and was able to guide the weight tuning process. On average, the quality score of plans generated under the WTPN's guidance was improved by ~8.5% compared to the initial plan with arbitrarily set weights, and by 10.7% compared to the plans generated by human planners. To our knowledge, this was the first time that a tool was developed to adjust organ weights for the treatment planning optimization problem in a human-like fashion based on intelligence learnt from a training process, which was different from existing strategies based on pre-defined rules. The study demonstrated potential feasibility to develop intelligent treatment planning approaches via deep reinforcement learning.

Automated high-dose rate brachytherapy treatment planning for a single-channel vaginal cylinder applicator.

High dose rate (HDR) brachytherapy treatment planning is conventionally performed manually and/or with aids of preplanned templates. In general, the standard of care would be elevated by conducting an automated process to improve treatment planning efficiency, eliminate human error, and reduce plan quality variations. Thus, our group is developing AutoBrachy, an automated HDR brachytherapy planning suite of modules used to augment a clinical treatment planning system. This paper describes our proof-of-concept module for vaginal cylinder HDR planning that has been fully developed. After a patient CT scan is acquired, the cylinder applicator is automatically segmented using image-processing techniques. The target CTV is generated based on physician-specified treatment depth and length. Locations of the dose calculation point, apex point and vaginal surface point, as well as the central applicator channel coordinates, and the corresponding dwell positions are determined according to their geometric relationship with the applicator and written to a structure file. Dwell times are computed through iterative quadratic optimization techniques. The planning information is then transferred to the treatment planning system through a DICOM-RT interface. The entire process was tested for nine patients. The AutoBrachy cylindrical applicator module was able to generate treatment plans for these cases with clinical grade quality. Computation times varied between 1 and 3 min on an Intel Xeon CPU E3-1226 v3 processor. All geometric components in the automated treatment plans were generated accurately. The applicator channel tip positions agreed with the manually identified positions with submillimeter deviations and the channel orientations between the plans agreed within less than 1 degree. The automatically generated plans obtained clinically acceptable quality.

Digital in-line holographic microscopy.

We first briefly review the state of the art of digital in-line holographic microscopy (DIHM) with numerical reconstruction and then discuss some technical issues, such as lateral and depth resolution, depth of field, twin image, four-dimensional tracking, and reconstruction algorithm. We then present a host of examples from microfluidics and biology of tracking the motion of spheres, algae, and bacteria. Finally, we introduce an underwater version of DIHM that is suitable for in situ studies in an ocean environment that show the motion of various plankton species.

Influence of pressure on the lengths of chemical bonds.

An expression to describe the force that a chemical bond exerts on its terminal atoms is proposed, and is used to derive expressions for the bond force constant and bond compressibility. The unknown parameter in this model, the effective charge on the atoms that form the bond, is determined by comparing the derived force constants with those obtained spectroscopically. The resultant bond compressibilities are shown to generally agree well with those determined from high-pressure structure determinations and from the bulk moduli of high-symmetry structures. Bond valences can be corrected for pressure by recognizing that the bond-valence parameter, R(0), changes with pressure according to the equation dR(0)/dP = 10(-4) R(04)/(1/B-2/R(0) AA GPa (-1).