Experience of Implementing Deep Learning-Based Automatic Contouring in Breast Radiation Therapy Planning: Insights From Over 2000 Cases.

PURPOSE: This study evaluated the impact and clinical utility of an auto-contouring system for radiation therapy treatments. METHODS AND MATERIALS: The auto-contouring system was implemented in 2019. We evaluated data from 2428 patients who underwent adjuvant breast radiation therapy before and after the system's introduction. We collected the treatment's finalized contours, which were reviewed and revised by a multidisciplinary team. After implementation, the treatment contours underwent a finalization process that involved manual review and adjustment of the initial auto-contours. For the preimplementation group (n = 369), auto-contours were generated retrospectively. We compared the auto-contours and final contours using the Dice similarity coefficient (DSC) and the 95% Hausdorff distance (HD95). RESULTS: We analyzed 22,215 structures from final and corresponding auto-contours. The final contours were generally larger, encompassing more slices in the superior or inferior directions. Among organs at risk (OAR), the heart, esophagus, spinal cord, and contralateral breast demonstrated significantly increased DSC and decreased HD95 postimplementation (all P < .05), except for the lungs, which presented inaccurate segmentation. Among target volumes, CTVn_L2, L3, L4, and the internal mammary node showed increased DSC and decreased HD95 postimplementation (all P < .05), although the increase was less pronounced than the OAR outcomes. The analysis also covered factors contributing to significant differences, pattern identification, and outlier detection. CONCLUSIONS: In our study, the adoption of an auto-contouring system was associated with an increased reliance on automated settings, underscoring its utility and the potential risk of automation bias. Given these findings, we underscore the importance of considering the integration of stringent risk assessments and quality management strategies as a precautionary measure for the optimal use of such systems.

Video-assisted thoracic surgery for spontaneous pneumothorax under two-lung ventilation and single-lumen endotracheal tube intubation.

Background: This study investigated the feasibility of video-assisted thoracic surgery (VATS) performed under two-lung ventilation (TLV) and single-lumen endotracheal tube (SLET) intubation in patients with spontaneous pneumothorax. Methods: From January 2016 to December 2019, 344 patients who underwent VATS with spontaneous pneumothorax, whether primary or secondary, were enrolled. The surgery was performed through TLV using SLET intubation or one-lung ventilation (OLV) using double-lumen endotracheal tube (DLET) intubation. Patient data were collected retrospectively from medical records and compared with an emphasis on the time required for anesthesia and surgery. Results: The average anesthesia time was 72.6±17.8 min for TLV and 89.9±24.3 min for OLV (P<0.001). The average operating time was 42.1±16.2 min for TLV and 54.7±23.8 min for OLV (P<0.001). The average time from the onset of anesthesia to incision was 23.6±7.0 min for TLV and 27.6±9.5 min for OLV (P<0.001). There was no case of conversion to OLV using DLET intubation during surgery with TLV using SLET intubation. Removal of the chest tube took 1.6±1.1 days for the TLV group and 2.3±3.6 days for the OLV group (P=0.017). Patients were discharged at 2.7±1.2 days after surgery for the TLV group and 3.2±2.3 days after surgery for the OLV group (P=0.009). Conclusions: TLV using SLET intubation could shorten the time required for anesthesia-related procedures and surgery. In addition, it can be a beneficial surgical and anesthetic option for pneumothorax.

Automated Computer-Aided Detection of Lung Nodules in Metastatic Colorectal Cancer Patients for the Identification of Pulmonary Oligometastatic Disease.

PURPOSE: This study aimed to explore the possibility and clinical utility of existing artificial intelligence (AI)-based computer-aided detection (CAD) of lung nodules to identify pulmonary oligometastases. PATIENTS AND METHODS: The chest computed tomography (CT) scans of patients with lung metastasis from colorectal cancer between March 2006 and November 2018 were analyzed. The patients were selected from a database of 1395 patients and studied in 2 cohorts. The first cohort included 50 patients, and the CT scans of these patients were independently evaluated for lung-nodule (≥3 mm) detection by a CAD-assisted radiation oncologist (CAD-RO) as well as by an expert radiologist. Interobserver variability by 2 additional radiation oncologists and 2 thoracic surgeons were also measured. In the second cohort of 305 patients, survival outcomes were evaluated based on the number of CAD-RO-detected nodules. RESULTS: In the first cohort, the sensitivity and specificity of the CAD-RO for identifying oligometastatic disease (OMD) from varying criteria by ≤2 nodules, ≤3 nodules, ≤4 nodules, and ≤5 nodules were 71.9% and 88.9%, 82.9% and 93.3%, 97.1% and 73.3%, and 97.5% and 90.0%, respectively. The sensitivity of the CAD-RO in the nodule detection compared with the radiologist was 81.6%. The average (standard deviation) sensitivity in interobserver variability analysis was 80.0% (3.7%). In the second cohort, the 5-year survival rates of patients with 1, 2, 3, 4, or ≥5 metastatic nodules were 75.2%, 52.9%, 45.7%, 29.1%, and 22.7%, respectively. CONCLUSIONS: Proper identification of the pulmonary OMD and the correlation between the number of CAD-RO-detected nodules and survival suggest the potential practicality of AI in OMD recognition. Developing a deep learning-based model specific to the metastatic setting, which enables a quick estimation of disease burden and identification of OMD, is underway.

Evaluation of deep learning-based autosegmentation in breast cancer radiotherapy.

PURPOSE: To study the performance of a proposed deep learning-based autocontouring system in delineating organs at risk (OARs) in breast radiotherapy with a group of experts. METHODS: Eleven experts from two institutions delineated nine OARs in 10 cases of adjuvant radiotherapy after breast-conserving surgery. Autocontours were then provided to the experts for correction. Overall, 110 manual contours, 110 corrected autocontours, and 10 autocontours of each type of OAR were analyzed. The Dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to compare the degree of agreement between the best manual contour (chosen by an independent expert committee) and each autocontour, corrected autocontour, and manual contour. Higher DSCs and lower HDs indicated a better geometric overlap. The amount of time reduction using the autocontouring system was examined. User satisfaction was evaluated using a survey. RESULTS: Manual contours, corrected autocontours, and autocontours had a similar accuracy in the average DSC value (0.88 vs. 0.90 vs. 0.90). The accuracy of autocontours ranked the second place, based on DSCs, and the first place, based on HDs among the manual contours. Interphysician variations among the experts were reduced in corrected autocontours, compared to variations in manual contours (DSC: 0.89-0.90 vs. 0.87-0.90; HD: 4.3-5.8 mm vs. 5.3-7.6 mm). Among the manual delineations, the breast contours had the largest variations, which improved most significantly with the autocontouring system. The total mean times for nine OARs were 37 min for manual contours and 6 min for corrected autocontours. The results of the survey revealed good user satisfaction. CONCLUSIONS: The autocontouring system had a similar performance in OARs as that of the experts' manual contouring. This system can be valuable in improving the quality of breast radiotherapy and reducing interphysician variability in clinical practice.

Clinical feasibility of deep learning-based auto-segmentation of target volumes and organs-at-risk in breast cancer patients after breast-conserving surgery.

BACKGROUND: In breast cancer patients receiving radiotherapy (RT), accurate target delineation and reduction of radiation doses to the nearby normal organs is important. However, manual clinical target volume (CTV) and organs-at-risk (OARs) segmentation for treatment planning increases physicians' workload and inter-physician variability considerably. In this study, we evaluated the potential benefits of deep learning-based auto-segmented contours by comparing them to manually delineated contours for breast cancer patients. METHODS: CTVs for bilateral breasts, regional lymph nodes, and OARs (including the heart, lungs, esophagus, spinal cord, and thyroid) were manually delineated on planning computed tomography scans of 111 breast cancer patients who received breast-conserving surgery. Subsequently, a two-stage convolutional neural network algorithm was used. Quantitative metrics, including the Dice similarity coefficient (DSC) and 95% Hausdorff distance, and qualitative scoring by two panels from 10 institutions were used for analysis. Inter-observer variability and delineation time were assessed; furthermore, dose-volume histograms and dosimetric parameters were also analyzed using another set of patient data. RESULTS: The correlation between the auto-segmented and manual contours was acceptable for OARs, with a mean DSC higher than 0.80 for all OARs. In addition, the CTVs showed favorable results, with mean DSCs higher than 0.70 for all breast and regional lymph node CTVs. Furthermore, qualitative subjective scoring showed that the results were acceptable for all CTVs and OARs, with a median score of at least 8 (possible range: 0-10) for (1) the differences between manual and auto-segmented contours and (2) the extent to which auto-segmentation would assist physicians in clinical practice. The differences in dosimetric parameters between the auto-segmented and manual contours were minimal. CONCLUSIONS: The feasibility of deep learning-based auto-segmentation in breast RT planning was demonstrated. Although deep learning-based auto-segmentation cannot be a substitute for radiation oncologists, it is a useful tool with excellent potential in assisting radiation oncologists in the future. Trial registration Retrospectively registered.

Viscum pleurodesis is as effective as talc pleurodesis and tends to have less adverse effect.

PURPOSE: Many patients diagnosed with advanced cancer have malignant pleural effusion that does not respond to chemotherapy or radiation therapy. These patients often have respiratory symptoms, especially dyspnea. In order to relieve these symptoms, various procedures including chemical pleurodesis have been performed. Although talc is the most widely used and effective sclerosing agent, there it has various adverse effects. The objective of this study was to determine whether Viscum (ABNOVA Viscum® Fraxini Injection, manufactured by ABNOVA GmbH, Germany) could be used as an agent to replace talc in clinical practice. METHODS: Data of 56 patients with malignant pleural effusion who received chemical pleurodesis after tube thoracostomy from January 2003 to December 2017 were retrospectively reviewed to analyze clinical course and response after pleurodesis with each agent. RESULTS: After pleurodesis, changes in numeric rating scale (NRS) was 1.4 ± 1.6 in the talc group and 0.5 ± 1.5 in the Viscum group (p = 0.108). Changes in white blood cell counts after pleurodesis were 4154.8 ± 6710.7 in the talc group and 3487.3 ± 6067.7 in the Viscum group (p = 0.702). Changes in C-reactive protein (CRP) were 9.03 ± 6.86 in the talc group and 6.3 ± 7.5 in the Viscum group (p = 0.366). The success rate of pleurodesis was 93.3% in the talc group and 96% in the Viscum group (p = 0.225). CONCLUSION: Viscum pleurodesis showed comparable treatment results with talc pleurodesis while its adverse effects such as chest pain and fever tended to be relatively weak.

Cardiac Motion Correction for Helical CT Scan With an Ordinary Pitch.

Cardiac X-ray computed tomography (CT) imaging is still challenging due to the cardiac motion during CT scanning, which leads to the presence of motion artifacts in the reconstructed image. In response, many cardiac X-ray CT imaging algorithms have been proposed, based on motion estimation (ME) and motion compensation (MC), to improve the image quality by alleviating the motion artifacts in the reconstructed image. However, these ME/MC algorithms are mainly based on an axial scan or a low-pitch helical scan. In this paper, we propose a ME/MC-based cardiac imaging algorithm for the data set acquired from a helical scan with an ordinary pitch of around 1.0 so as to obtain the whole cardiac image within a single scan of short time without ECG gating. In the proposed algorithm, a sequence of partial angle reconstructed (PAR) images is generated by using consecutive parts of the sinogram, each of which has a small angular span. Subsequently, an initial 4-D motion vector field (MVF) is obtained using multiple pairs of conjugate PAR images. The 4-D MVF is then refined based on an image quality metric so as to improve the quality of the motion-compensated image. Finally, a time-resolved cardiac image is obtained by performing motion-compensated image reconstruction by using the refined 4-D MVF. Using digital XCAT phantom data sets and a human data set commonly obtained via a helical scan with a pitch of 1.0, we demonstrate that the proposed algorithm significantly improves the image quality by alleviating motion artifacts.

Subject-specific Brain Tumor Growth Modelling via An Efficient Bayesian Inference Framework.

An accurate prediction of brain tumor progression is crucial for optimized treatment of the tumors. Gliomas are primarily treated by combining surgery, external beam radiotherapy, and chemotherapy. Among them, radiotherapy is a non-invasive and effective therapy, and an understanding of tumor growth will allow better therapy planning. In particular, estimating parameters associated with tumor growth, such as the diffusion coefficient and proliferation rate, is crucial to accurately characterize physiology of tumor growth and to develop predictive models of tumor infiltration and recurrence. Accurate parameter estimation, however, is a challenging task due to inaccurate tumor boundaries and the approximation of the tumor growth model. Here, we introduce a Bayesian framework for a subject-specific tumor growth model that estimates the tumor parameters effectively. This is achieved by using an improved elliptical slice sampling method based on an adaptive sample region. Experimental results on clinical data demonstrate that the proposed method provides a higher acceptance rate, while preserving the parameter estimation accuracy, compared with other state-of-the-art methods such as Metropolis-Hastings and elliptical slice sampling without any modification. Our approach has the potential to provide a method to individualize therapy, thereby offering an optimized treatment.

A laceration of left internal mammary artery caused by an amateur golfer's swing.

We report the rare case of a patient presenting with a spontaneous laceration of left internal mammary artery (LIMA) after playing golf. The patient had no specific history except for cardiac surgery, and there were no results that caused bleeding on preoperative examination. A computed tomography (CT) scan of the chest demonstrated an anterior mediastinal hematoma and a left hemothorax with active extravasation close to LIMA. Through thoracotomy, hematoma evacuation and clipping for lacerated artery were performed. The patient was discharged in stable condition on the sixteenth postoperative day. This is the first reported case of a spontaneous laceration of internal mammary artery (IMA) after playing golf.

Cardiac Image Reconstruction via Nonlinear Motion Correction Based on Partial Angle Reconstructed Images.

Even though the X-ray Computed Tomography (CT) scan is considered suitable for fast imaging, motion-artifact-free cardiac imaging is still an important issue, because the gantry rotation speed is not fast enough compared with the heart motion. To obtain a heart image with less motion artifacts, a motion estimation (ME) and motion compensation (MC) approach is usually adopted. In this paper, we propose an ME/MC algorithm that can estimate a nonlinear heart motion model from a sinogram with a rotation angle of less than 360°. In this algorithm, we first assume the heart motion to be nonrigid but linear, and thereby estimate an initial 4-D motion vector field (MVF) during a half rotation by using conjugate partial angle reconstructed images, as in our previous ME/MC algorithm. We then refine the MVF to determine a more accurate nonlinear MVF by maximizing the information potential of a motion-compensated image. Finally, MC is performed by incorporating the determined MVF into the image reconstruction process, and a time-resolved heart image is obtained. By using a numerical phantom, a physical cardiac phantom, and an animal data set, we demonstrate that the proposed algorithm can noticeably improve the image quality by reducing motion artifacts throughout the image.

Denoising of polychromatic CT images based on their own noise properties.

PURPOSE: Because of high diagnostic accuracy and fast scan time, computed tomography (CT) has been widely used in various clinical applications. Since the CT scan introduces radiation exposure to patients, however, dose reduction has recently been recognized as an important issue in CT imaging. However, low-dose CT causes an increase of noise in the image and thereby deteriorates the accuracy of diagnosis. In this paper, the authors develop an efficient denoising algorithm for low-dose CT images obtained using a polychromatic x-ray source. The algorithm is based on two steps: (i) estimation of space variant noise statistics, which are uniquely determined according to the system geometry and scanned object, and (ii) subsequent novel conversion of the estimated noise to Gaussian noise so that an existing high performance Gaussian noise filtering algorithm can be directly applied to CT images with non-Gaussian noise. METHODS: For efficient polychromatic CT image denoising, the authors first reconstruct an image with the iterative maximum-likelihood polychromatic algorithm for CT to alleviate the beam-hardening problem. We then estimate the space-variant noise variance distribution on the image domain. Since there are many high performance denoising algorithms available for the Gaussian noise, image denoising can become much more efficient if they can be used. Hence, the authors propose a novel conversion scheme to transform the estimated space-variant noise to near Gaussian noise. In the suggested scheme, the authors first convert the image so that its mean and variance can have a linear relationship, and then produce a Gaussian image via variance stabilizing transform. The authors then apply a block matching 4D algorithm that is optimized for noise reduction of the Gaussian image, and reconvert the result to obtain a final denoised image. To examine the performance of the proposed method, an XCAT phantom simulation and a physical phantom experiment were conducted. RESULTS: Both simulation and experimental results show that, unlike the existing denoising algorithms, the proposed algorithm can effectively reduce the noise over the whole region of CT images while preventing degradation of image resolution. CONCLUSIONS: To effectively denoise polychromatic low-dose CT images, a novel denoising algorithm is proposed. Because this algorithm is based on the noise statistics of a reconstructed polychromatic CT image, the spatially varying noise on the image is effectively reduced so that the denoised image will have homogeneous quality over the image domain. Through a simulation and a real experiment, it is verified that the proposed algorithm can deliver considerably better performance compared to the existing denoising algorithms.

Cardiac motion correction based on partial angle reconstructed images in x-ray CT.

PURPOSE: Cardiac x-ray CT imaging is still challenging due to heart motion, which cannot be ignored even with the current rotation speed of the equipment. In response, many algorithms have been developed to compensate remaining motion artifacts by estimating the motion using projection data or reconstructed images. In these algorithms, accurate motion estimation is critical to the compensated image quality. In addition, since the scan range is directly related to the radiation dose, it is preferable to minimize the scan range in motion estimation. In this paper, the authors propose a novel motion estimation and compensation algorithm using a sinogram with a rotation angle of less than 360°. The algorithm estimates the motion of the whole heart area using two opposite 3D partial angle reconstructed (PAR) images and compensates the motion in the reconstruction process. METHODS: A CT system scans the thoracic area including the heart over an angular range of 180° + α + ß, where α and ß denote the detector fan angle and an additional partial angle, respectively. The obtained cone-beam projection data are converted into cone-parallel geometry via row-wise fan-to-parallel rebinning. Two conjugate 3D PAR images, whose center projection angles are separated by 180°, are then reconstructed with an angular range of ß, which is considerably smaller than a short scan range of 180° + α. Although these images include limited view angle artifacts that disturb accurate motion estimation, they have considerably better temporal resolution than a short scan image. Hence, after preprocessing these artifacts, the authors estimate a motion model during a half rotation for a whole field of view via nonrigid registration between the images. Finally, motion-compensated image reconstruction is performed at a target phase by incorporating the estimated motion model. The target phase is selected as that corresponding to a view angle that is orthogonal to the center view angles of two conjugate PAR images. To evaluate the proposed algorithm, digital XCAT and physical dynamic cardiac phantom datasets are used. The XCAT phantom datasets were generated with heart rates of 70 and 100 bpm, respectively, by assuming a system rotation time of 300 ms. A physical dynamic cardiac phantom was scanned using a slowly rotating XCT system so that the effective heart rate will be 70 bpm for a system rotation speed of 300 ms. RESULTS: In the XCAT phantom experiment, motion-compensated 3D images obtained from the proposed algorithm show coronary arteries with fewer motion artifacts for all phases. Moreover, object boundaries contaminated by motion are well restored. Even though object positions and boundary shapes are still somewhat different from the ground truth in some cases, the authors see that visibilities of coronary arteries are improved noticeably and motion artifacts are reduced considerably. The physical phantom study also shows that the visual quality of motion-compensated images is greatly improved. CONCLUSIONS: The authors propose a novel PAR image-based cardiac motion estimation and compensation algorithm. The algorithm requires an angular scan range of less than 360°. The excellent performance of the proposed algorithm is illustrated by using digital XCAT and physical dynamic cardiac phantom datasets.

Hybrid palliation for right atrial isomerism associated with obstructive total anomalous pulmonary venous drainage.

A twenty-four-day-old girl, who was prematurely born at 36 weeks of gestation, and weighed 2.2 kg, and diagnosed with right atrial isomerism, functionally single ventricle, bilateral superior vena cava (SVC) and obstructive supracardiac total anomalous pulmonary venous drainage (TAPVD) draining to the junction between the right SVC and the right atrium, underwent a hybrid procedure in the operating room, which consisted of pulmonary artery banding, ductus ligation and stenting of the draining vein of TAPVD. Obstruction at the drainage site of TAPVD was initially relieved after stenting, but, one month after the procedure, the distal end of the stent became stenotic and she received bilateral sutureless repair of TAPVD. At postoperative seven months, she underwent bidirectional cavopulmonary shunt uneventfully, and she has been followed-up for two months in a stable state without any problem in the pulmonary venous pathway.