Search

Purpose The purpose of this study is to propose algorithms and methods for achieving high accuracy in tracking and interception irradiation technology for tumors that move by respiration using MR-linac (MRIdian®, ViewRay Inc.) and to use deep learning to predict the movement of moving tumors in real time during radiation therapy and reconstruct cine magnetic resonance imaging (cine-MRI) into four-dimensional (4D) movies. Methods In this study, we propose a reconstruction algorithm using 4DCT for treatment planning taken before irradiation as training data in consideration of the actual treatment flow. In the algorithm, two neural networks made before treatment are used to reconstruct 4D movies that predict tumor movement in real time during treatment. Cycle GAN (generative adversarial network) was used to convert MR images to CT images, and long short-term memory was used to convert cine-MRI to 4D movies and predict tumor movement. Results We succeeded in predicting the time including the imaging time of the MR images, the lag until irradiation, and the calculation time in the algorithm. In addition, the conversion and prediction results at each phase of reconstruction were generally good so that they could be clinically applied. Conclusions The reconstruction algorithm proposed in this study enables high-precision radiotherapy while predicting the volume information of the tumor and the actual tumor position, which could not be obtained during radiotherapy.

To identify functions and facilities that are critical for ensuring safety in advanced reactors, for which only a limited usage history is available, it is necessary to consider the characteristics of each reactor type by utilizing more risk-related information. In this study, we compare the response capability during an accident for light water reactors and fast reactors to quantitatively evaluate the impact of the reactor type on resilience. A resilience evaluation procedure is developed for assessing the response margin and response reliability of a nuclear power plant during an accident; this procedure was then applied to a simple pressurized-water reactor and a sodium-cooled fast reactor plant model using an accident scenario involving an earthquake that generated a leakage of coolant from the primary heat-transport system. Our research indicated a relatively large difference between the response reliability with respect to time for the two reactor types.

Anomaly detection in vibration signals is an important technique for fault diagnosis, monitoring, and maintenance in nuclear power plants. Therefore, various signal-analysis methods that apply stat...

Installation of surveillance cameras in nuclear power plants is critical to protecting the facilities against terrorist attacks or monitoring the reactor operator. This has led to large amounts of video surveillance data, creating a demand for automatic detection of anomalies or suspicious movements. Tracking human motion from video sequences is a notable technique used for detecting anomalies in human behavior and is currently achieved with the use of a depth camera. However, depth cameras require a complicated camera system and their field of view is limited. To overcome this problem, there is a need for recognizing human motion in wide-angle images – a view that often causes distortion. In this study, we devised a method for tracking human motion through wideangle image distortion. The main contribution of this study is a methodology that automatically estimates the transformation parameters needed to improve the accuracy of motion recognition; these parameters are applied to a distorted wide-angle image in every frame. We propose a new multi-layered convolutional neural architecture for estimating the locations of human joints in images and transformation parameters simultaneously. When applied to distorted wide-angle images, the robustness of our method is demonstrated through a quantitative evaluation of human joint location prediction. In addition, we compare our method with a motion tracking system and an infrared-camera-based motion capture system to demonstrate its ability to handle wide-angle and close-range images.

Tracking human motion from video sequences is a well-known video surveillance technique, and many commercially available motion capture devices can now recognize human poses using a depth camera. However, depth camera systems are complicated and have limited optical fields of view. To overcome this problem, it is necessary to develop techniques for recognizing human motion in wide-angle images. In this study, we devised a method for tracking human motion that is robust to wide-angle image distortion. To do so, we developed a new multilayered convolutional neural network architecture for estimating the locations of human body parts in images along with associated transformation parameters that can be applied to a distorted wide-angle image on a frame-by-frame basis. The proposed method was applied to distorted wide-angle images, and its robustness was demonstrated via a quantitative evaluation of human joint prediction and a comparative analysis with a commercially available depth camera-based motion capture system.

When a heat transfer tube of the steam generator of a pressurized water reactor fails, the primary cooling water leaks quickly into the secondary system. Moreover, if this leakage is large, the nuc...

Decommissioning of the Fukushima Daiichi nuclear power station (NPS) is an unprecedented effort in which the entire plant was contaminated with radioactive materials from the accident and radiation dose levels were not low. The decommissioning workers in these sites are indicated to wear personal protective equipment (PPE) properly for radiation protection. In response to the difficulties of on-site PPE management in decommissioning site of Fukushima Daiichi NPS, this paper presents a graph-based scene understanding approach that automatically identifies whether workers at a decommissioning site are using PPE properly. The experimental results demonstrate that the high precision and high recall of the model can effectively detect decommissioning workers’ failure use of PPE and can facilitate improved safety inspection and supervision.

In recent years, deep learning has been very successful in various fields. In particular, since image data contains a lot of information, it is expected that deep learning will be applied to nuclear security technology for detecting the malicious action. However, most human action recognition using deep learning only sets OK / NO in advance for the combination of object identification result, so it is difficult to apply them to complicated nuclear security situation. On the other hand, in most nuclear facility, the nuclear security rules are determined as a detailed text. So, it is desirable to detect malicious actions in captured images according to these rule text. If it is possible to flexibly determine whether the captured image scene violates or matches the nuclear security rules, it is necessary to greatly improve its practicality. However, this requires different deep learning interfaces for image identification and natural language processing. Most of the deep learning research currently being conducted is closed only within each field, and there is no mutual access, that is, an interface. The human brain, on the other hand, usually links and processes all kinds of cognitive information, including language, sight, speech, and touch. Like the brain, deep learning abilities should develop anew when different types of deep learning are interconnected. One of the reasons this is difficult is that there is no common data format between different types of deep learning. In this study, it was determined that the logical expression is the most suitable as a common data format. The logical expression is a method of expressing the correlation between an object and a person. In this research, a deep learning interface method between image recognition and natural language processing was developed via the logical expression. This allows flexible judgment of wrong actions according to the rule text.

Decommissioning of the Fukushima Daiichi nuclear power station is challenging due to industrial and chemical hazards as well as radiological ones. The decommissioning workers in these sites are indicated to wear Personal Protective Equipment (PPE) appropriately for radiation protection. In response to the difficulties of on-site PPE management in decommissioning site of Fukushima Daiichi NPS, this paper proposes the combination of deep learning-based object detection and individual detection using geometry relationships analysis to automatically facilitate the safety monitoring task of decommissioning workers to ensure hardhats, full-face masks and dust masks are correctly used for each zone. The experimental results demonstrate that the high precision, high recall and fast speed of the proposed approach can effectively detect decommissioning workers’ incorrect use of PPE.

Following the Fukushima Daiichi nuclear accident, fears have arisen that terrorists can cause a similar accident by acts of sabotage against nuclear facilities; as a result, the importance of nucle...

Respiratory motion management is a huge challenge in radiation therapy. Respiratory motion induces temporal anatomic changes that distort the tumor volume and its position. In this study, a markerless tumor-tracking algorithm was investigated by performing phase recognition during stereotactic body radiation therapy (SBRT) using four-dimensional cone-beam computer tomography (4D-CBCT) obtained at patient registration, and in-treatment cone-beam projection images. The data for 20 treatment sessions (five lung cancer patients) were selected for this study. Three of the patients were treated with conventional flattening filter (FF) beams, and the other two were treated with flattening filter-free (FFF) beams. Prior to treatment, 4D-CBCT was acquired to create the template projection images for 10 phases. In-treatment images were obtained at near real time during treatment. Template-based phase recognition was performed for 4D-CBCT re-projected templates using prior 4D-CBCT based phase recognition algorithm and was compared with the results generated by the Amsterdam Shroud (AS) technique. Visual verification technique was used for the verification of the phase recognition and AS technique at certain tumor-visible angles. Offline template matching analysis using the cross-correlation indicated that phase recognition performed using the prior 4D-CBCT and visual verification matched up to 97.5% in the case of FFF, and 95% in the case of FF, whereas the AS technique matched 83.5% with visual verification for FFF and 93% for FF. Markerless tumor tracking based on phase recognition using prior 4D-CBCT has been developed successfully. This is the first study that reports on the use of prior 4D-CBCT based on normalized cross-correlation technique for phase recognition.

During the radiotherapy treatment of patients with lung cancer, the radiation delivered to healthy tissue around the tumor needs to be minimized, which is difficult because of respiratory motion and the latency of linear accelerator systems. In the proposed study, we first use the Lucas-Kanade pyramidal optical flow algorithm to perform deformable image registration of chest computed tomography scan images of four patients with lung cancer. We then track three internal points close to the lung tumor based on the previously computed deformation field and predict their position with a recurrent neural network (RNN) trained using real-time recurrent learning (RTRL) and gradient clipping. The breathing data is quite regular, sampled at approximately 2.5Hz, and includes artificial drift in the spine direction. The amplitude of the motion of the tracked points ranged from 12.0mm to 22.7mm. Finally, we propose a simple method for recovering and predicting 3D tumor images from the tracked points and the initial tumor image based on a linear correspondence model and Nadaraya-Watson non-linear regression. The root-mean-square error, maximum error, and jitter corresponding to the RNN prediction on the test set were smaller than the same performance measures obtained with linear prediction and least mean squares (LMS). In particular, the maximum prediction error associated with the RNN, equal to 1.51mm, is respectively 16.1% and 5.0% lower than the maximum error associated with linear prediction and LMS. The average prediction time per time step with RTRL is equal to 119ms, which is less than the 400ms marker position sampling time. The tumor position in the predicted images appears visually correct, which is confirmed by the high mean cross-correlation between the original and predicted images, equal to 0.955., Comment: 21 pages, 24 figures; accepted manuscript version

Bearings are one of the main components of rotating machinery, and their failure is one of the most common cause of mechanical failure. Therefore, many fault detection methods based on artificial intelligence, such as machine learning and deep learning, have been proposed. Particularly, with recent advances in deep learning, many anomaly detection methods based on deep neural networks (DNN) have been proposed. DNNs provide high-performance recognition and are easy to implement; however, optimizing DNNs require large annotated datasets. Additionally, the annotation of time-series data, such as abnormal vibration signals, is time consuming. To solve these problems, we proposed a method to automatically extract features from abnormal vibration signals from the time-series data. In this research, we propose a new DNN training method and fault detection method inspired by multi-instance learning. Additionally, we propose a new loss function for optimizing the DNN model that identifies anomalies from a time-series data. Furthermore, to evaluate the feasibility of automatic feature extraction from vibration signal data using the proposed method, we conducted experiments to determine whether anomalies could be detected, identified, and localized in published datasets.

Construction sites are one of the most perilous environments where many potential hazards may occur. Even though workers are trained to stay away from potential dangers, there are still many types of risks that can occur within only a few minutes of carelessness. Personal Protective Equipment (PPE) is an important safety measure used to protect construction workers from accidents. However, PPE usage is not strictly enforced among workers due to all kinds of reasons. This paper proposes the combination of deep learning-based object detection and individual detection using geometry relationships analysis to automatically identify non-PPE-use (NPU); i.e., if a worker is wearing hardhat, eye protection visors, dust masks, or both, to help to facilitate the safety monitoring work of construction workers to ensure PPE are appropriately used. The experimental results demonstrate that the approach was capable of detecting NPU workers with high precision (84.13%) and recall rate (93.10%) while ensuring real-time performance (7.95 FPS on average).

In this work, a fiber-guided laser array spots thermography (LAST) system with good flexibility is proposed for the inspection of surface cracks in metallic structural objects with difficult access. To generate an array of laser spots on the target surface, an optical head with fiber delivery is designed and fabricated at first, which has good adaptability and robustness. Second, a numerical simulation model is developed to optimize the spatial distribution of the laser spots. In addition, an improved image processing method with use of multiple background free images is developed to enhance the crack detectability from the LAST images. Finally, cracks of different types in stainless steel specimens are successfully detected from the thermal images obtained using the fiber-guided LAST system.

Phase map cross-correlation detection and quantification may produce highlighted signal at superparamagnetic iron oxide nanoparticles, and distinguish them from other hypointensities. The method may quantify susceptibility change by performing least squares analysis between a theoretically generated magnetic field template and an experimentally scanned phase image. Because characteristic phase recognition requires the removal of phase wrap and phase background, additional steps of phase unwrapping and filtering may increase the chance of computing error and enlarge the inconsistence among algorithms. To solve problem, phase gradient cross-correlation and quantification method is developed by recognizing characteristic phase gradient pattern instead of phase image because phase gradient operation inherently includes unwrapping and filtering functions. However, few studies have mentioned the detectable limit of currently used phase gradient calculation algorithms. The limit may lead to an underestimation of large magnetic susceptibility change caused by high-concentrated iron accumulation. In this study, mathematical derivation points out the value of maximum detectable phase gradient calculated by differential chain algorithm in both spatial and Fourier domain. To break through the limit, a modified quantification method is proposed by using unwrapped forward differentiation for phase gradient generation. The method enlarges the detectable range of phase gradient measurement and avoids the underestimation of magnetic susceptibility. Simulation and phantom experiments were used to quantitatively compare different methods. In vivo application performs MRI scanning on nude mice implanted by iron-labeled human cancer cells. Results validate the limit of detectable phase gradient and the consequent susceptibility underestimation. Results also demonstrate the advantage of unwrapped forward differentiation compared with differential chain algorithms for susceptibility quantification at high-concentrated iron accumulation.

The detection of the presence of a structural abnormality in rotating machines with parts without gap, such as rigid couplings, by using the conventional method of vibration RMS, is difficult. This work investigates the application of Singular Spectral Analysis (SSA) to change detection in rotating machines. The performance of this technique is quantitatively evaluated in the scenario of misalignment in a turbopump assembly with rigid coupling. It is shown that a statistically significant change appears when using the Change Score calculated using the SSA method. Comparison with special indicators of a structural abnormality is also performed. Since the SSA detects changes in the shape of the signal, it is much more sensitive to changes related to abnormality than conventional methods or the use of special indicators.

Decommissioning of the Fukushima Daiichi nuclear power station (NPS) is challenging due to industrial and chemical hazards as well as radiological ones. The decommissioning workers in these sites are instructed to wear proper Personal Protective Equipment (PPE) for radiation protection. However, workers may not be able to accurately comply with safety regulations at decommissioning sites, even with prior education and training. In response to the difficulties of on-site PPE management, this paper presents a vision-based automated monitoring approach to help to facilitate the occupational safety monitoring task of decommissioning workers to ensure proper use of PPE by the combination of deep learning-based individual detection and object detection using geometric relationships analysis. The performance of the proposed approach was experimentally evaluated, and the experimental results demonstrate that the proposed approach is capable of identifying decommissioning workers&rsquo, improper use of PPE with high precision and recall rate while ensuring real-time performance to meet the industrial requirements.

In this study, we develop new indices that evaluate the response capability — response margin and response reliability — of a nuclear power plant in case of occurrence of a severe accident by considering the ability of the nuclear power plant to recover system safety functions, which may have been affected by the accident, to the minimum safety values. Further, we demonstrate a specific evaluation procedure for our proposed indices. While performing the evaluation, we consider margins with respect to a minimum safety function level and a time constraint along with their temporal changes. Additionally, for a simplified fast reactor (FR) plant model, we conduct a trial assessment of the recovery capacity of the system to ensure system safety based on the concept of new indices. Further, the applicability of the response reliability evaluation method to FR plants is discussed based on the viewpoints of reflecting the characteristics of each reactor type.

Respiratory motion induces the limit in delivery accuracy due to the lack of the consideration of the anatomy motion in the treatment planning. Therefore, image-guided radiation therapy (IGRT) system plays an essential role in respiratory motion management and real-time tumor tracking in external beam radiation therapy. The objective of this research is the prediction of dynamic time-series images considering the motion and the deformation of the tumor and to compensate the delay that occurs between the motion of the tumor and the beam delivery. For this, we propose a prediction algorithm for dynamic time-series images. Prediction is performed using principal component analysis (PCA) and multi-channel singular spectral analysis (MSSA). Using PCA, the motion can be denoted as a vector function and it can be estimated by its principal component which is the linear combination of eigen vectors corresponding to the largest eigen values. Time-series set of 320-detector-row CT images from lung cancer patient and kilovolt (kV) fluoroscopic images from a moving phantom were used for the evaluation of the algorithm, and both image sets were successfully predicted by the proposed algorithm. The accuracy of prediction was quite high, more than 0.999 for CT images, whereas 0.995 for kV fluoroscopic images in cross-correlation coefficient value. This algorithm for image prediction makes it possible to predict the tumor images over the next breathing period with significant accuracy.

In this work, a non-contact ultrasonic method based on laser generation and EMAT (electromagnetic acoustic trans- ducer) detection is proposed for residual stress measurement. A numerical method based on finite element model is established to simulate the laser-generated ultrasound in pre-stressed media to determine the capability and sensitivity of this method. Fi- nally, the laser-EMAT ultrasonic method isstudied to measure the residual stress in bearing balls to assess the fatigue level of the bearings.