Real-time Position Monitoring Research Articles

Machine learning for predicting accuracy of lung and liver tumor motion tracking using radiomic features.

Internal tumor motion is commonly predicted using external respiratory signals. However, the internal/external correlation is complex and patient-specific. The purpose of this study was to develop various models based on the radiomic features of computed tomography (CT) images to predict the accuracy of tumor motion tracking using external surrogates and to find accurate and reliable tracking algorithms. Images obtained from a total of 108 and 71 patients pathologically diagnosed with lung and liver cancers, respectively, were examined. Real-time position monitoring motion was fitted to tumor motion, and samples with fitting errors greater than 2 mm were considered positive. Radiomic features were extracted from internal target volumes of average intensity projections, and cross-validation least absolute shrinkage and selection operator (LassoCV) was used to conduct feature selection. Based on the radiomic features, a total of 26 separate models (13 for the lung and 13 for the liver) were trained and tested. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to assess performance. Relative standard deviation was used to assess stability. Thirty-three and 22 radiomic features were selected for the lung and liver, respectively. For the lung, the AUC varied from 0.848 (decision tree) to 0.941 [support vector classifier (SVC), logistic regression]; sensitivity varied from 0.723 (extreme gradient boosting) to 0.848 [linear support vector classifier (linearSVC)]; specificity varied from 0.834 (gaussian naive bayes) to 0.936 [multilayer perceptron (MLP), wide and deep (W&D)]; and MLP and W&D had better performance and stability than the median. For the liver, the AUC varied from 0.677 [light gradient boosting machine (Light)] to 0.892 (logistic regression); sensitivity varied from 0.717 (W&D) to 0.862 (MLP); specificity varied from 0.566 (Light) to 0.829 (linearSVC); and logistic regression, MLP, and SVC had better performance and stability than the median. Respiratory-sensitive radiomic features extracted from CT images of lung and liver tumors were proved to contain sufficient information to establish an external/internal motion relationship. We developed a rapid and accurate method based on radiomics to classify the accuracy of monitoring a patient's external surface for lung and liver tumor tracking. Several machine learning algorithms-in particular, MLP-demonstrated excellent classification performance and stability.

The delivered dose assessment in pancreas SBRT with the target position determined using an in-house position monitoring system.

This study assessed the delivered dose accuracy in pancreas SBRT by incorporating the real-time target position determined using an in-house position monitoring system. An online image-based position monitoring system, SeedTracker, was developed to monitor radiopaque marker positions using monoscopic x-ray images, available from the Elekta XVI imaging system. This system was applied to patients receiving SBRT for pancreatic cancer on the MASTERPLAN Pilot trial (ACTRN 12617001642370). All patients were implanted pre-treatment with at least three peri-tumoral radiopaque markers for target localisation. During treatment delivery, marker positions were compared to expected positions delineated from the planning CT. The position tolerance of ±3mm from the expected position of the markers was set to trigger a gating event (GE) during treatment. The dosimetric impact of position deviations and actual dose delivered with position corrections was assessed by convolving the plan control point dose matrices with temporal target positions determined during treatment. Eight patients were treated within this study. At least one GE was observed in 38% of the treatment fractions and more than one GE was observed in 10% of the fractions. The position deviations resulted in the mean(range) difference of -0.1(-1.1 - 0.4)Gy in minimum dose to tumour and 1.9(-0.1- 4.6)Gy increase to Dmax to duodenum compared to planned dose. In actual treatment delivery with the patient realignment, the mean difference of tumour min dose and duodenal Dmax was reduced to 0.1(-1.0 - 1.1)Gy and 1.1 (-0.7 - 3.3)Gy respectively compared to the planned dose. The in-house real-time position monitoring system improved the treatment accuracy of pancreatic SBRT in a general-purpose linac and enabled assessment of delivered dose by incorporating the temporal target position during delivery. The intrafraction motion impacts the dose to tumour even if target position is maintained within a 3mm position tolerance.

The first real-time intrafraction target position monitoring in pancreas SBRT on an Elekta linear accelerator.

To perform implanted fiducial based real-time target position monitoring in pancreas stereotactic body radiotherapy (SBRT) using the x-ray imaging system available in a Elekta linear accelerator. An in-house system was developed and clinically utilised for real-time target position monitoring of pancreas SBRT delivery. The developed system was used for the target position monitoring of a pancreas cancer patient treated in free breathing treatment within the study entitled 'Mfolfirinox And STEreotactic Radiotherapy for Patients with Locally Advanced paNcreas cancer (MASTERPLAN): a feasibility study' (ACTRN 12617001642370) consisting of five treatment fractions. The clinical efficacy of the system was studied by performing a retrospective cumulative dose assessment of delivered dose using observed position deviations. The developed system identified two events of baseline shifts in target position that exceeded the accepted tolerance level of ± 3 mm from reference planned position. The retrospective dose assessment study showed that if the position deviations were not detected and corrected for, the maximum dose to duodenum would have increased from 34.6 to 38.8 Gy. The first real-time position monitoring in pancreas SBRT on an Elekta linear accelerator was successfully performed. The developed system was shown to improve the safety and accuracy of SBRT delivery.

Validation of a gyroscope-based wearable device for real-time position monitoring of patients in a hospital.

Monitoring patients' position is important, but there have been few studies related to validation. The objective of this study was to assess the validity of position monitoring measured using a wearable device by comparing the device's measurements to a patient's actual position. We constructed a wearable device with a three-axis gyroscope and applied it to 10 patients who were unable to change their position independently. We compared the actual angle of the position and the angle transmitted from the wearable device using a Bland-Altman plot and a receiver operating characteristic curve. We compared the actual angle of the position and the angle transmitted from the wearable device using a Bland-Altman plot, but it was difficult to observe statistical similarity. The angles transmitted from the wearable device in the lateral and supine positions showed significant differences. The cutoff value separating the lateral and supine positions was found to be 27.1∘ (sensitivity = 100%, specificity = 99.9%). Through our method, the measured values from the gyroscope-based wearable device did not accurately reflect the patient's actual position. However, the wearable device was able to distinguish the lateral position from the supine position.

Assessment of Intra-Fraction Prostate Motion and Delivered Dose Accuracy in Prostate SBRT Using an in-House Real-Time Position Monitoring System

Assessment of Intra-Fraction Prostate Motion and Delivered Dose Accuracy in Prostate SBRT Using an in-House Real-Time Position Monitoring System

The Virtual-Spine Platform-Acquiring, visualizing, and analyzing individual sitting behavior.

Back pain is a serious medical problem especially for those people sitting over long periods during their daily work. Here we present a system to help users monitoring and examining their sitting behavior. The Virtual-Spine Platform (VSP) is an integrated system consisting of a real-time body position monitoring module and a data visualization module to provide individualized, immediate, and accurate sitting behavior support. It provides a comprehensive spine movement analysis as well as accumulated data visualization to demonstrate behavior patterns within a certain period. The two modules are discussed in detail focusing on the design of the VSP system with adequate capacity for continuous monitoring and a web-based interactive data analysis method to visualize and compare the sitting behavior of different persons. The data was collected in an experiment with a small group of subjects. Using this method, the behavior of five subjects was evaluated over a working day, enabling inferences and suggestions for sitting improvements. The results from the accumulated data module were used to elucidate the basic function of body position recognition of the VSP. Finally, an expert user study was conducted to evaluate VSP and support future developments.

A Low-Cost Tracking System for Running Race Applications Based on Bluetooth Low Energy Technology.

Timing points used in running races and other competition events are generally based on radio-frequency identification (RFID) technology. Athletes’ times are calculated via passive RFID tags and reader kits. Specifically, the reader infrastructure needed is complex and requires the deployment of a mat or ramps which hide the receiver antennae under them. Moreover, with the employed tags, it is not possible to transmit additional and dynamic information such as pulse or oximetry monitoring, alarms, etc. In this paper we present a system based on two low complex schemes allowed in Bluetooth Low Energy (BLE): the non-connectable undirected advertisement process and a modified version of scannable undirected advertisement process using the new capabilities present in Bluetooth 5. After fully describing the system architecture, which allows full real-time position monitoring of the runners using mobile phones on the organizer side and BLE sensors on the participants’ side, we derive the mobility patterns of runners and capacity requirements, which are determinant for evaluating the performance of the proposed system. They have been obtained from the analysis of the real data measured in the last Barcelona Marathon. By means of simulations, we demonstrate that, even under disadvantageous conditions (50% error ratio), both schemes perform reliably and are able to detect the 100% of the participants in all the cases. The cell coverage of the system needs to be adjusted when non-connectable process is considered. Nevertheless, through simulation and experimental, we show that the proposed scheme based on the new events available in Bluetooth 5 is clearly the best implementation alternative for all the cases, no matter the coverage area and the runner speed. The proposal widely exceeds the detection requirements of the real scenario, surpassing the measured peaks of 20 sensors per second incoming in the coverage area, moving at speeds that range from 1.5 m/s to 6.25 m/s. The designed real test-bed shows that the scheme is able to detect 72 sensors below 600 ms, fulfilling comfortably the requirements determined for the intended application. The main disadvantage of this system would be that the sensors are active, but we have proved that its consumption can be so low (9.5 µA) that, with a typical button cell, the sensor battery life would be over 10,000 h of use.

EP-1623: SeedTracker: Enabling real time position monitoring with a conventional linacs for prostate SBRT

EP-1623: SeedTracker: Enabling real time position monitoring with a conventional linacs for prostate SBRT

Variable angle stereo imaging for rapid patient position correction in an in-house real-time position monitoring system

PurposeTo develop and validate a variable angle stereo image based position correction methodology in an X-ray based in-house online position monitoring system. Materials and methodsA stereo imaging module that enables 3D position determination and couch correction of the patient based on images acquired at any arbitrary angle and arbitrary angular separation was developed and incorporated to the in-house SeedTracker real-time position monitoring system. The accuracy of the developed system was studied by imaging an anthropomorphic phantom implanted with radiopaque markers set to known offset positions from its reference position in an Elekta linear accelerator (LA) and associated XVI imaging system. The accuracy of the system was further validated using CBCT data set from 10 prostate SBRT patients. The time gains achieved with the stereo image based position correction was compared with the manual matching of seed positions in Digitally Reconstructed Radiographs (DRRs) and kV images in the Mosaiq record and verify system. ResultsBased on phantom and patient CBCT dataset study stereo imaging module implemented in the SeedTracker shown to have an accuracy of 0.1(σ=0.5)mm in detecting the 3D position offset. The time comparison study showed that stereo image based methodology implemented in SeedTracker was a minimum of 80(4)s faster than the manual method implemented in Mosaiq R&V system with a maximum time saving of 146(6)s. ConclusionThe variable angle stereo image based position correction method was shown to be accurate and faster than the standard manual DRR–kV image based correction approach, leading to more efficient treatment.

Clinical validation of a real-time applicator position monitoring system for gynecologic intracavitary brachytherapy

Purpose: To validate the clinical feasibility and efficacy of a real-time applicator position monitoring system (RAPS) through a phantom study and a prospective clinical trial. Methods and materials: The RAPS measures the brachytherapy applicator displacement in real-time by computing the relative displacement between two infrared reflective targets, one attached to the applicator and the other to the patient’s skin. A phantom study was performed to compare RAPS measurements with the ground truth. Six cervical cancer patients were enrolled in the clinical trial using MRI-based high-dose-rate brachytherapy with a Tandem-and-Ovoids applicator. The results from the RAPS are compared with the clinical method. Results: In the phantom study, an average difference between RAPS measurements and known displacements was 0.02 ± 0.01 mm in the superior-inferior direction, 0.02 ± 0.02 mm in the lateral direction, and 0.11 ± 0.06 mm in the anterior-posterior direction. In the clinical trial, the absolute difference in applicator displacement between the RAPS and the clinical method was 1.46 ± 1.13 mm. In all patient cases, a maximum applicator displacement of 6.66 mm (2.0 ± 1.5 mm) was observed using the RAPS. Conclusions: This work demonstrates the clinical efficacy of RAPS to measure applicator displacement.

Model-Based Angular Scan Error Correction of an Electrothermally-Actuated MEMS Mirror.

In this paper, the actuation behavior of a two-axis electrothermal MEMS (Microelectromechanical Systems) mirror typically used in miniature optical scanning probes and optical switches is investigated. The MEMS mirror consists of four thermal bimorph actuators symmetrically located at the four sides of a central mirror plate. Experiments show that an actuation characteristics difference of as much as 4.0% exists among the four actuators due to process variations, which leads to an average angular scan error of 0.03°. A mathematical model between the actuator input voltage and the mirror-plate position has been developed to predict the actuation behavior of the mirror. It is a four-input, four-output model that takes into account the thermal-mechanical coupling and the differences among the four actuators; the vertical positions of the ends of the four actuators are also monitored. Based on this model, an open-loop control method is established to achieve accurate angular scanning. This model-based open loop control has been experimentally verified and is useful for the accurate control of the mirror. With this control method, the precise actuation of the mirror solely depends on the model prediction and does not need the real-time mirror position monitoring and feedback, greatly simplifying the MEMS control system.

Real-Time Precise IOT Positioning System Based on Chirp Spread Spectrum Technology for Transportation

Real-time precise IOT positioning system based on CSS uses the way of bus or Ethernet with wireless CSS network. With its unique CSS technology to measure the distance, the console can achieve real-time communication and position monitoring of personnel and goods. The system has better comprehensive performance than active RFID, Zigbee and WiFi positioning systems and other technology on the market currently. The functional independence of its communication system makes it easy for the practical application and extensions, and the system has wide application value.

The Research on Real-Time Position Monitoring System in Slurry Pipeline Grading Transmission

Due to the existence of different grades of mineral in the mining area, the technology of slurry pipeline grade transmission is developing. However, the process of grade transmission will appear the phenomenon that the slurry of different grades are mixed. Therefore, before cutting into different grade slurry, we added special medium in pipeline to separate different grades of slurry. We worked out slurry pipeline transport model by mathematical modeling and building slurry pipeline transport real-time location tracking system. The system was able to distinguish different grades of slurry by adding special medium and tracking the actual location of the head and tail of slurry, realizing real-time position monitoring of different grade slurry.

Real-Time Position and Speed Monitoring of Trains Using Phase-Sensitive OTDR

It is essential to carry out real-time position and speed monitoring of trains to ensure the safety of railway operation. In this letter, the phase-sensitive optical-time-domain reflectometry (Φ-OTDR) is demonstrated to overcome the drawbacks of the track circuit, which is widely used and has the risk to be damaged by lightening, for the first time. A long sensing cable, buried nearby to two parallel railways, is used to detect the vibration signals generated by trains with Φ-OTDR and the real-time vibration signals of the trains are extracted/quantified by wavelet denoising. With the edge detection by normalized sliding variance, the operation status of two nearby trains, including their relative position and speed, are obtained over a 10.2-km measurement length in real time. This letter offers a new passive way for safety monitoring of railway operation.

Implementation of GaN based real-time source position monitoring in HDR brachytherapy

For High Dose Rate (HDR) brachytherapy Quality Assurance (QA), we propose a seed position verification method by using two GaN dosimetric probes integrated in an instrumented phantom. As dose rate measured from one probe is related to the seed distance, seed position inside catheter can be determined. The proposed two-probe method extends measureable range and improves accuracy in comparison with the use of one single probe. One instrumented phantom implementing the method was designed, fabricated and tested in clinical conditions. The measureable range of the two-probe method is 105 mm compared to 36 mm using single probe. The validity of the method was experimentally verified.

A real-time applicator position monitoring system for gynecologic intracavitary brachytherapy

To develop a real-time applicator position monitoring system (RAPS) for intracavitary brachytherapy using an infrared camera and reflective markers. 3D image-guided brachytherapy requires high accuracy of applicator localization; however, applicator displacement can happen during patient transfer for imaging and treatment delivery. No continuous applicator position monitoring system is currently available. The RAPS system was developed for real-time applicator position monitoring without additional radiation dose to patients. It includes an infrared camera, reflective markers, an infrared illuminator, and image processing software. After reflective markers are firmly attached to the applicator and the patient body, applicator displacement can be measured by computing the relative change in distance between the markers. The reflective markers are magnetic resonance imaging (MRI) compatible, which is suitable for MRI-guided HDR brachytherapy paradigm. In our prototype, a Microsoft Kinect sensor with a resolution of 640 by 480 pixels is used as an infrared camera. A phantom study was carried out to compare RAPS' measurements with known displacements ranging from -15 to +15 mm. A reproducibility test was also conducted. The RAPS can achieve 4 frames/s using a laptop with Intel(®) Core™2 Duo processor. When the pixel size is 0.95 mm, the difference between RAPS' measurements and known shift values varied from 0 to 0.8 mm with the mean value of 0.1 mm and a standard deviation of 0.44 mm. The system reproducibility was within 0.6 mm after ten reposition trials. This work demonstrates the feasibility of a real-time infrared camera based gynecologic intracavitary brachytherapy applicator monitoring system. Less than 1 mm accuracy is achieved when using an off-the-shelf infrared camera.

Positional accuracy and transmitter orientation of the 3D electromagnetic tracking system

This research investigates the positional accuracy and effects of transmitter orientation of a 3D electromagnetic tracking (EMT) system. EMT systems, capable of real-time position and orientation monitoring, are commonly used in computer-aided surgical navigation and path monitoring. In this study, positional information is evaluated for accuracy by comparing the EMT system against laser interferometer measurements in three orthogonal axes with step sizes between 0.1 and 0.5 mm. The effect of transmitter orientation is evaluated by placing the transmitter with either the front or the side facing the magnetic sensor. Gauge repeatability and reproducibility results demonstrate that the EMT system can accurately measure the motion with a tolerance of 0.2 mm with 0.5 s measurement time. The transmitter oriented with the front facing the sensor has a higher positional accuracy than that of the side transmitter orientation. High accuracy of the EMT system combined with the knowledge of transmitter orientation information presents the potential for accurate navigation and path monitoring in medical procedures.

Sonographic position monitoring of central venous catheters by microbubble-injection in real time: development of a new procedure

Background: Central venous catheter placement requires immediate confirmation of its orthotopic position in the central vena cava to avoid arterial or paravasal infusions which could potentially harm the patient. The correct position is usually verified by chest x-ray or ECG. Echocardiography might offer a fast and safe alternative. This study aimed at developing a new non-invasive, ultrasound-based procedure to determine the orthotopic position of Central Venous Catheters (CVC) position by sonographic detection of injected Microbubbles (MB) Methods: With approval of the local ethics committee 95 patients with 98 CVC and 5 peripheral catheters were examined. The appearance of MB in the right heart can be observed after injection of agitated saline in the subcostal four-chamber view. MB are hyperechogenic and dissolve later. The time from injection at the catheter to the appearance in the RV can be recorded by M-Mode registration allowing an exact measurement independent of user interaction and reaction. To determine positioning, time of appearance of MB was recorded. The transit time was correlated with radiologically confirmed catheter positions. A self assessment of the procedures practicability was performed on a linear anologue scale (0-100%). Results: We report a total of 95 Patients (65.0±18 years, 57% male). The jugular vein was used in 85 and the subclavian vein in 13 cases. The main diagnoses requiring catheter placement were pneumonia (n=49), aortic valve reconstruction (n=9), myocardial infarction (n=6), heart failure (n=6), gastrointestinal bleeding (n=6) or others (n=9). In 95 patients with radiologically confirmed orthotopic position MB appeared in the right heart in less than one second after injection with a reliability of 100%. In 3 patients, quantitatively measured appearance of MB was even below 0.5 seconds. In 5 patients with peripheral venous catheters MB were observed later than 2 sec after injection and in one patient with a heterotopic CVC within 1-2 seconds after injection. Duration of the procedure was 146 (111) seconds (mean, (SD)). Physician's impression of practicalness was 91% (linear analogue self assessment). Conclusion: Ultrasound-based real-time position monitoring of CVC by this new description of the MB-injection technique has the potential to serve as an alternative to X-ray or other methods. Appearance of MBs in less than one second can be assumed to predict orthotopic position with high accuracy. Appearance of MB in more than two seconds indicates malpositioning. Transit times between one and two seconds require additional X-ray confirmation.

Experimental investigation of a general real-time 3D target localization method using sequential kV imaging combined with respiratory monitoring

The goal of this work was to experimentally quantify the geometric accuracy of a novel real-time 3D target localization method using sequential kV imaging combined with respiratory monitoring for clinically realistic arc and static field treatment delivery and target motion conditions. A general method for real-time target localization using kV imaging and respiratory monitoring was developed. Each dimension of internal target motion T(x, y, z; t) was estimated from the external respiratory signal R(t) through the correlation between R(ti) and the projected marker positions p(xp, yp; ti) on kV images by a state-augmented linear model: T(x, y, z; t) = aR(t) + bR(t − τ) + c. The model parameters, a, b, c, were determined by minimizing the squared fitting error ∑‖p(xp, yp; ti) − P(θi) · (aR(ti) + bR(ti − τ) + c)‖2 with the projection operator P(θi). The model parameters were first initialized based on acquired kV arc images prior to MV beam delivery. This method was implemented on a trilogy linear accelerator consisting of an OBI x-ray imager (operating at 1 Hz) and real-time position monitoring (RPM) system (30 Hz). Arc and static field plans were delivered to a moving phantom programmed with measured lung tumour motion from ten patients. During delivery, the localization method determined the target position and the beam was adjusted in real time via dynamic multileaf collimator (DMLC) adaptation. The beam-target alignment error was quantified by segmenting the beam aperture and a phantom-embedded fiducial marker on MV images and analysing their relative position. With the localization method, the root-mean-squared errors of the ten lung tumour traces ranged from 0.7–1.3 mm and 0.8–1.4 mm during the single arc and five-field static beam delivery, respectively. Without the localization method, these errors ranged from 3.1–7.3 mm. In summary, a general method for real-time target localization using kV imaging and respiratory monitoring has been experimentally investigated for arc and static field delivery. The average beam-target error was 1 mm.

SU-E-I-92: Accuracy Evaluation of Depth Data in Microsoft Kinect.

Microsoft Kinect has potential for use in real-time patient position monitoring in diagnostic radiology and radiotherapy. We evaluated the accuracy of depth image data and the device-to-device variation in various conditions simulating clinical applications in a hospital. Kinect sensor consists of infrared-ray depth camera and RGB camera. We developed a computer program using OpenNI and OpenCV for measuring quantitative distance data. The program displays depth image obtained from Kinect sensor on the screen, and the cartesian coordinates at an arbitrary point selected by mouse-clicking can be measured. A rectangular box without luster (300 × 198 × 50 mm3 ) was used as a measuring object. The object was placed on the floor at various distances ranging from 0 to 400 cm in increments of 10 cm from the sensor, and depth data were measured for 10 points on the planar surface of the box. The measured distance data were calibrated by using the least square method. The device-to-device variations were evaluated using five Kinect sensors. There was almost linear relationship between true and measured values. Kinect sensor was unable to measure at a distance of less than 50 cm from the sensor. It was found that distance data calibration was necessary for each sensor. The device-to-device variation error for five Kinect sensors was within 0.46% at the distance range from 50 cm to 2 m from the sensor. The maximum deviation of the distance data after calibration was 1.1 mm at a distance from 50 to 150 cm. The overall average error of five Kinect sensors was 0.18 mm at a distance range of 50 to 150 cm. Kinect sensor has distance accuracy of about 1 mm if each device is properly calibrated. This sensor will be useable for positioning of patients in diagnostic radiology and radiotherapy.