Fiducial Detection Research Articles

SU‐C‐BRA‐04: Determination of the Optimal X‐Ray Monitoring Angle for Creating a Correlation Model in Dynamic Tumor Tracking Irradiation with Vero4DRT(MHI‐TM2000)

Purpose: The Vero4DRT(MHI‐Tm2000) can perform dynamic tumor tracking (DTT) irradiation using two special features; orthogonal kV x‐ray imaging subsystems and a gimbaled x‐ray head. Prior to DTT irradiation a correlation model between the tumor position estimated from implanted fiducials and IR marker position on the abdomen is required. The purpose of this study was to estimate an optimal x‐ray monitoring angle for creating the correlation model utilizing CT images. Methods: A water equivalent path length (WEPL) on CT images was employed as an indicator to estimate the optimal angles. First, the relationship between the WEPL on 4D‐CT images at the end‐exhalation phase (4D‐CT EE) and an intensity ratio of the fiducial to its surroundings on fluoroscopic images was examined to assess the validity of our proposed methods. Subsequently, the optimal angles were calculated on 4D‐CT EE images for 11 patients with 3 to 5 implanted fiducials as follows; (1) detection of fiducials on CT images, (2) exclusion of angles at which a fiducial was overlapped with other fiducials, (3) calculation of WEPLs on the orthogonal lines passing the fiducials, and (4) determination of the monitoring angles by minimizing the sum of the WEPLs. The determined optimal angles were also applied to 4D‐CT images at the end‐inhalation phase and other 4D‐CT sets. Results: The WEPL was highly correlated with the logarithm of the intensity ratio (R=−0.89 [95%CI: −0.86‐0.91], p<0.01). The optimal gantry and ring angles depended on the geometric arrangement between fiducials and anatomical structures. The inter‐ and intrafractional variations in the intensity ratios estimated from WEPLs were observed; however, the intensity ratios exceeded a threshold level to detect fiducials. Conclusions: We demonstrated that the optimal x‐ ray monitoring angles for the correlation model were determined by minimizing the WEPLs on CT images. This research was supported by the Japan Society for the Promotion of Science (JSPS) through its Funding Program for World‐Leading Innovation R&D on Science and Technology (FIRST Program). Research sponsored in part by Mitsubishi Heavy Industries, Ltd.

WE-G-213CD-05: Beam's-Eye-View Prostate Motion Detection during Arc Radiotherapy.

To track prostate motion during clinical arc radiotherapy using a novel statistical motion tracking algorithm in both phantom and patient systems. For the phantom study, a dynamic phantom with a custom tumor model and radiopaque fiducials was used. For the patient study, the clinical data from two prostate patients were acquired during volumetric modulated arc therapy (VMAT). For each patient, two gold fiducials (X-Mark, 1cm and 2cm in length) were implanted in the prostate. A total of 35 fractions with 2-field full arcs were delivered. Beam's-eye- view (BEV) images were acquired in cine mode every five fractions at a frame rate of 2 fps.The prostate motion was measured by two sequential steps: fiducial detection and motion estimation. In the first step, a wavelet- based block matching filter was used to suppress the noise and enhance the fiducials in the BEV images. A trajectory of detected fiducials was generated along the projection angles. Sequentially, the prostate motion was tracked using a statistical maximum a posteriori (MAP) algorithm. The 3-D fiducial position was estimated by maximizing the probability of displacement of 2-D projections. A total of 8,400 BEV images were processed for this study. The phantom experiments showed that the baseline accuracy of estimation is 0.69+/-0.56 mm. For patient 1, intra- fraction prostate motion was measured to be 0.98+/-0.51 mm for all fractions, and individual fraction motion less than 1 mm. For patient 2, intra- fraction motion was 1.12+/-0.65 mm for all fractions. However, an exceptionally large displacement was measured as 4.9+/-0.74mm in one of the fractions. Prostate motion during arc-therapy can be rigorously tracked to an accuracy of one millimeter. This technique is well suited for in-treatment room target tracking and post-therapy evaluation. Varian Medical Systems Inc.

SU-E-J-138: Fast 2-D Fiducial Marker Detection on Sequential MV Projections in Arc Therapy

To automatically detect intrafraction motion during arc radiotherapy for prostate cancer patients by tracking fiducial markers in two-dimensional MV images acquired using the treatment beam, in order to adjust radiation dose accordingly. Four fiducial gold markers are implanted in a patient's prostate. Patients are irradiated using a Varian Linac 2100 C/D with RapidArc upgrade (Varian Medical Systems, Palo Alto, CA). MV images (1024 × 768 pixels, 0.392 × 0.392 mm2 pixel size) acquired during a 360 degree gantry rotation at a one second interval (5 degrees) are preprocessed by subtracting a smoothed version of the image to retain only high image frequencies. Edge detection is then applied, followed by a one pixel wide dilation and erosion to transform the edges into contiguous regions. Next, our method searches the centers of visible markers (i.e. not covered by the MLC), constrained by marker estimates from the planning CT. This is done by finding all contiguous regions and maximizing a marker-region distance criterion for every visible marker. A two-dimensional estimate correction over consecutive projections is also implemented to improve marker estimates during gantry rotation. We applied our method on four treatment fractions of the same patient. As such, a total of 191 projections with manually indicated marker ends as ground truth were used as validation. Markers were indicated twice on all images, to include observer errors. Results show a mean detection error of less than 0.5 mm in the projection image (standard deviation 0.6 mm), with an execution time of less than one second per image in matlab. Undetected markers and false positives mostly occurred at moving leaf boundaries, where marker visibility was determined by the observer. Preliminary findings demonstrate that this method can be used to detect intrafraction motion during arc radiotherapy by only using projected MV images. Research sponsored by Varian Medical Systems, Palo Alto, CA.

SU-E-J-140: Initial Clinical Assessment of a Gimbaled Linac Tumor Tracking System in a Patient Simulation Study.

A simulation study was conducted on patients to evaluate the workflow and quantify the performance of the BrainLab/MHI Vero dynamic tumor tracking system in clinical circumstances. The gimbals hold the linac-MLC assembly which enables tracking of moving tumors. Two kV imaging systems are attached at ±45° from the 6MV beam allowing simultaneous X-rays. A simulation study was conducted on 5 lung-liver patients. The procedure involved quantification of tumor motion based on localization of Visicoil gold markers implanted in the tumor. Except for switching on the treatment beam, the entire tumor tracking workflow was executed involving patient positioning, synchronized acquisition of skin marker motion and X-ray images, fiducial marker detection, external-internal correlation model calculation, skin marker surrogate guided tracking and monitoring imaging. Tracking error was calculated from gimbals log-files and the acquired monitoring X-rays. Imaging dose was measured with TLD on phantoms and on the patients. Imaging for correlation model building resulted in 17.6mGy skin dose. Taking the treatment duration of a 3×20Gy lung SBRT treatment, depending on the treatment fields orientation an additional maximal exposure of 28.8mGy was estimated for acquiring 1Hz X-ray monitoring during tracking. A mean absolute tracking error of 1.1 mm was measured, with a 90% percentile of 2.1 mm. The average time to set up the patient entering the room to the first MV beam-on was 9min. From the acquisition of the modeling images sequence up to beam-on took 3min. A clinical version of the Vero tumor tracking system has been installed, including automatic detection of fiducial markers implanted in the tumor. An initial assessment has shown that the tracking system is functional and its performance adequate to move forward to final commissioning and initiation of patient treatments. This collaborative work was supported by the Flemish government through the Hercules foundation and the “Fonds voor Wetenschappelijk Onderzoek - Vlaanderen†grants G.0486.06 and G.0412.08, and corporate funding from BrainLab AG. There are no other conflicts of interest.

WE-G-213CD-03: A Dual Complementary Verification Method for Dynamic Tumor Tracking on Vero SBRT.

to use complementary cine EPID and gimbals log file analysis for in-vivo tracking accuracy monitoring. A clinical prototype of dynamic tracking (DT) was installed on the Vero SBRT system. This prototype version allowed tumor tracking by gimballed linac rotations using an internal-external correspondence model. The DT prototype software allowed the detailed logging of all applied gimbals rotations during tracking. The integration of an EPID on the vero system allowed the acquisition of cine EPID images during DT. We quantified the tracking error on cine EPID (E-EPID) by subtracting the target center (fiducial marker detection) and the field centroid. Dynamic gimbals log file information was combined with orthogonal x-ray verification images to calculate the in-vivo tracking error (E-kVLog). The correlation between E-kVLog and E-EPID was calculated for validation of the gimbals log file. Further, we investigated the sensitivity of the log file tracking error by introducing predefined systematic tracking errors. As an application we calculate gimbals log file tracking error for dynamic hidden target tests to investigate gravity effects and decoupled gimbals rotation from gantry rotation. Finally, calculating complementary cine EPID and log file tracking errors evaluated the clinical accuracy of dynamic tracking. A strong correlation was found between log file and cine EPID tracking error distribution during concurrent measurements (R=0.98). We found sensitivity in the gimbals log files to detect a systematic tracking error up to 0.5 mm. Dynamic hidden target tests showed no gravity influence on tracking performance and high degree of decoupled gimbals and gantry rotation during dynamic arc dynamic tracking. A submillimetric agreement between clinical complementary tracking error measurements was found. Redundancy of the internal gimbals log file with x-ray verification images with complementary independent cine EPID images was implemented to monitor the accuracy of gimballed tumor tracking on Vero SBRT. Research was financially supported by the Flemish government (FWO), Hercules Foundation and BrainLAB AG.

SU‐C‐BRA‐06: Monitoring Tumor Motion Using Beam‐Eye‐View EPID Imaging during IMRT: A Gradient‐Based Fiducial Detection Algorithm

Purpose: Intrafractional tumor motion introduces significant dose uncertainty for radiotherapy. Using EPID imaging during IMRT delivery will provide a beam‐eye‐view (BEV) tumor motion observation and it also provides a cost‐efficient tumor tracking method, avoiding additional imaging system. The purpose of this study is to develop a fiducial detection algorithm for tumor motion monitoring using beam‐eye‐view (BEV) EPID imaging during external beam IMRT treatment. Methods: A gradient‐based fiducial detection algorithm was developed which includes steps: (1) determine IMRT segment; (2) define a search region; (3) Image gradient, grad(I), within the search region was calculated and a threshold was applied to find candidates; (4) the size and orientation (obtained using radon transform) of each candidate was determined, and compared to expected values (from treatment plan) to find the fiducial. To test this algorithm, external beam breast IMRT fields were delivered to a breast phantom (size: ∼13×10cm) with a gold fiducial (1×5 mm) on the Varian Trilogy. Portal images were acquired using Varian aSi1000 EPID system with the following settings: continuous mode, 3 averages per image, image size 1024×768 with resolution 0.392mm per pixel, EPID at 140cm SID (source‐imager‐ distance). Results: Twenty portal images acquired at different time during IMRT delivery, of which 10 images were with couch shift (4mm), have been tested and initial results show that the algorithm is able to detect the fiducial successfully (100%). The gradient‐based method, which basically uses the edge information, has the advantage over intensity‐based method for MV portal images since generally MV images have lower contrast compared with KV images. Conclusion: A gradient‐based fiducial algorithm has been developed and tested on breast IMRT treatment using a breast phantom. Further studies are warranted for other tumor sites such as prostate IMRT treatment.

TU-E-BRA-03: Real-Time Fiducial Detection and Prostate Movement Assessment with Cine MV Images in RapidArc Treatments.

To develop an algorithm for detection of metallic fiducial markers in cine MV images, and to assess the prostate movement during RapidArc treatment. A Varian TrueBeam linear accelerator (LINAC) was used to deliver RapidArc treatment for prostate patients. Cine images were acquired with the onboard electronic portal imaging device (EPID) using the MV therapeutic beam. Three metallic fiducial markers were implanted inside the prostate. To detect the fiducial position, we explicitly account for the possible marker blockage by MLC during beam modulation. If the marker is not blocked, we employ the planning coordinates of the marker centroids projected onto the cine MV images and perform template matching in the vicinity of its projection to localize the actual position of the marker. Displacements of the fiducial markers are assessed by comparing the actual and planned positions. We analyzed ∼280 cine MV images acquired during a 55-sec RapidArc treatment for a prostate patient. The three markers were visible in about 46%, 52%, and 48% of the images, and at least one fiducial was visible during almost entire treatment (97% of the time). The marker detection algorithm agrees well with manual detection (< 0.2 mm). The mean displacement for each fiducial was 0.40 ± 0.42, 0.27 ± 0.29, and 0.46 ± 0.34 mm. The maximum displacement was 2.33, 1.75, and 2.23 mm. An algorithm for automatic detection of fiducial markers in cine MV images has been developed. The prostate movement during a RapidArc treatment has been analyzed for a patient with implanted markers. Accurate target positioning is achieved at all times during treatment. In light of the random nature of intrafraction prostate motion, this work represents an important step toward real-time image-guided prostate radiation therapy.

Unveiling the uncertainty principle in the QRS complex offset detection on high resolution electrocardiography

The accuracy of high resolution electrocardiographic (HRECG) methods for stratifying the risk of malignant ventricular arrhythmia depends on the fidelity of QRS fiducial points detection. This study aims at examining the effect of acquisition and processing variables in HRECG on the variability of QRS complex offset (QRS offset) detection in simulated and biological signals, as well as investigating the factors related to the so called uncertainty principle applied to HRECG. Successive QRS offset locations were calculated in different signals configurations including HRECG data from patients with and without ventricular late potentials and simulated data using linear and exponential functions. The expected error in QRS offset detection was assessed as a function of: i) signal characteristics (Simulated or Biological); ii) Sampling Frequency (SF); iii) Residual Noise Level (RNL); iv) QRS maximum amplitudes. The uncertainty principle was related to HRECG and a given exponential signals, and increasing RNL up to 0.5 µV. SF and RNL are outstanding factors influencing QRS offset variability. Thus, HRECG related uncertainty principle is a deterministic phenomenon associated with both HRECG signal and mathematical formulation of the terminal decay of the QRS complex to the fusion with the ST segment.

Packaging Technology of Multi Deflection Arrays for Multi-Shaped Beam Lithography

Multi-Shaped electron beam lithography is considered a promising approach for high throughput mask and direct writing. Providing multiple apertures and individually controlled electrodes it allows for massive parallelization of exposure shots, thus significantly decreasing write time. A silicon-based micro-structured MEMS multi-beam deflection array (MDA) featuring 8×8 apertures is presented. The hybrid integration of MDA devices in ceramic system carriers utilizing a laser-based Solderjet Bumping process is demonstrated. This flux-free soldering process provides adhesive-free, long term stable and vacuum compatible joints and is used for both mechanical fixation and electrical connection. Electron beam deflection in two perpendicular directions requires the highly accurate placement of two crossed MDA devices, which is carried out by three degrees of freedom alignment procedures and solder joining. Electrical signal routing within the electron optical column using flexible printed circuit boards and flux-free soldering is also reported. The precision adjustment of two carriers is accomplished by fiducial mark detection using image processing. Results on alignment accuracy in the sub-micron range, mechanical and electrical testing of such assemblies are reported.

TU‐A‐204B‐03: 3‐D Fiducial Localization with MV EPID Imaging during a RapidArc Therapy

Purpose: Fiducials can be used to aid organ localization during RapidArc delivery. Specifically, the 3‐D tumor position may be found through the tracking of fiducials during beam delivery. A promising technique is the detection of fiducials in 2‐D EPID images, and then reconstruction of the 3‐D position using limited views. Methods and Materials: The CIRS dynamic thorax phantom with a custom tumor model is used in the study. Two fiducials are placed on the surface of the tumor at clinically relevant distances apart from each other. Clinical RapidArc plans are delivered to the phantom based on a single 360° arc. The fiducial detection algorithms are developed based on an image block matching filter and anisotropic diffusion techniques. A wavelet‐based block matching filter is developed to suppress the noise and texture of images while preserving the edges of fiducials. Furthermore, the anisotropic diffusion technique removes redundant edges and trivial structures of the results of the previous filtering. Experiments show that the algorithms are robust to image contrast and aperture variations during the RapidArc delivery. With detected 2‐D locations, 3‐D spatial locations of fiducials are reconstructed using the inverse 3‐D rotation matrix. Results: The position accuracy in 2D EPID is 0.77+0.58 mm by comparing with the manual outlines. The 3D fiducial positions are reconstructed using the 2D EPID images. The displacement from the references in the 3D CBCT image is within one millimeter. Conclusion: Novel filtering and detection algorithms are developed to identify the fiducials in 2D EPID images. The results show that limited angles of full rotation can rigorously reconstruct the locations of each of fiducials in 3D to within one millimeter. Conflict of Interest: Varian Medical Systems, Inc.

Multimodal information fusion for selected multimedia applications

The effective interpretation and integration of multiple information content are important for the efficacious utilisation of multimedia in a wide variety of application context. The major challenge in multimodal information fusion lies in the difficulty of identifying the complementary and discriminatory representations from individual channels, and the efficient fusion of the resulting information for the targeted application problem. This paper outlines several multimedia systems that utilise a multimodal approach, and provides a comprehensive review of the state-of-the-art in related areas, including emotion recognition, image annotation and retrieval, and biometrics. Data collected from diverse sources or sensors are employed to improve the recognition or classification accuracy. It is shown that the combination of multimodality information is capable of providing a more complete and effective description of the intrinsic characteristics of the specific pattern, and producing improved system performance than single modality only. In addition, we present a facial fiducial point detection and a gesture recognition system, which can be incorporated into a multimodal framework. The issues and challenges in the research and development of multimodal systems are discussed, and a cutting-edge application of multimodal information fusion for intelligent robotic system is presented.

Robust automatic detection and removal of fiducial projections in fluoroscopy images: An integrated solution

Automatic detection and removal of fiducial projections in fluoroscopy images is an essential prerequisite for C-arm calibration. This paper presents an integrated solution to fulfill this task. A custom-designed calibration cage with a two-plane pattern of fiducials is utilized in our solution. The cage is attached to the C-arm image intensifier and acquired images are calibrated automatically by a three-step on-line calibration algorithm including fiducial projection detection, image calibration, and fiducial projection removal. A sequence of carefully designed image processing operations consisting of image binarization, connected-component labeling, region classification, adaptive template matching, and shape analysis, are developed for an accurate and robust localization of fiducial projections. A similarity measure that is proposed previously for image-based 2D–3D registration is employed in the adaptive template matching to improve the detection accuracy. Shape analysis based on the design information of the calibration cage is used to further improve the robustness of the detection. Thin-plate spline based vector transforms are used to correct the image distortion. The detected fiducial projections are then removed by an image inpainting technique based on the fast marching method for level set applications. Our in vitro experiments show on average 4 s execution time on a Pentium IV machine, a zero false-detection rate, a miss-detection rate of 1.6 ± 2.3%, and a sub-pixel localization error. Using a custom-made tool for checking accuracy, a forward projection error of 1.0 ± 0.4 pixels and a backward projection error of 0.3 ± 0.1 mm were found. We are confident that our solution is fast, robust, and accurate enough for image-guided interventional applications.

ECG biometric analysis in cardiac irregularity conditions

Biometric traits offer direct solutions to the critical security concerns involved in identity authentication systems. In this paper, a systematic analysis of the electrocardiogram (ECG) signal for application in human recognition is reported, suggesting that cardiac electrical activity is highly personalized in a population. Features extracted from the autocorrelation of healthy ECG signals embed considerable diacritical power, and render fiducial detection unnecessary. The central consideration of this paper is the evaluation of an identification system that is robust to common cardiac irregularities such as premature ventricular contraction (PVC) and atrial premature contraction (APC). Criteria concerning the power distribution and complexity of ECG signals are defined to bring to light abnormal ECG recordings, which are not employable for identification. Experimental results indicate a recognition rate of 96.2% and render identification based on ECG signals rather promising.

A fiducial detection algorithm for real-time image guided IMRT based on simultaneous MV and kV imaging

The advantage of highly conformal dose techniques such as 3DCRT and IMRT is limited by intrafraction organ motion. A new approach to gain near real-time 3D positions of internally implanted fiducial markers is to analyze simultaneous onboard kV beam and treatment MV beam images (from fluoroscopic or electronic portal image devices). Before we can use this real-time image guidance for clinical 3DCRT and IMRT treatments, four outstanding issues need to be addressed. (1) How will fiducial motion blur the image and hinder tracking fiducials? kV and MV images are acquired while the tumor is moving at various speeds. We find that a fiducial can be successfully detected at a maximum linear speed of 1.6 cm/s. (2) How does MV beam scattering affect kV imaging? We investigate this by varying MV field size and kV source to imager distance, and find that common treatment MV beams do not hinder fiducial detection in simultaneous kV images. (3) How can one detect fiducials on images from 3DCRT and IMRT treatment beams when the MV fields are modified by a multileaf collimator (MLC)? The presented analysis is capable of segmenting a MV field from the blocking MLC and detecting visible fiducials. This enables the calculation of nearly real-time 3D positions of markers during a real treatment. (4) Is the analysis fast enough to track fiducials in nearly real time? Multiple methods are adopted to predict marker positions and reduce search regions. The average detection time per frame for three markers in a 1024 x 768 image was reduced to 0.1 s or less. Solving these four issues paves the way to tracking moving fiducial markers throughout a 3DCRT or IMRT treatment. Altogether, these four studies demonstrate that our algorithm can track fiducials in real time, on degraded kV images (MV scatter), in rapidly moving tumors (fiducial blurring), and even provide useful information in the case when some fiducials are blocked from view by the MLC. This technique can provide a gating signal or be used for intra-fractional tumor tracking on a Linac equipped with a kV imaging system. Any motion exceeding a preset threshold can warn the therapist to suspend a treatment session and reposition the patient.

WE-C-351-06: Tracking Multiple Moving Fiducials During Treatment Based On Simultaneous Onboard KV and Treatment MV Imaging

Purpose: Intra‐fractional organ motion can limit the advantage of highly conformal dose techniques as IMRT due to target position uncertainty. A new strategy has been proposed and investigated to track implanted fiducial markers during treatment on a LINAC with an onboard kV imaging system by using simultaneous kV and treatment MV beam imaging.Method and Materials: A pelvic phantom implanted with three gold cylindrical seeds was tested on a moving platform on a Varian Trilogy. Two phantom verification plans for 3DCRT and IMRTtreatments from real patients were delivered on this phantom while the platform was moving at different speeds. A new pattern matching algorithm has been developed to locate multiple cylindrical fiducials on MV and kV images. Their 3D positions were then calculated from simultaneous orthogonal MV and kV imaging results. Results: Four issues were studied towards the clinical application of this method. 1) Multiple methods were adopted to predict marker positions and reduce search regions to achieve an analysis speed of 10 frame/second. 2) The kV and MV images were acquired while the phantom was moving at various speeds and fiducials could be successfully detected at a linear speed of 1.6 cm/s or less. 3) Varying MV field size and kV source to imager distance indicated that the detection of fiducials on kV images was not affected by the scattering from simultaneous treatment MV beams. 4) Prediction based on relative positions of fiducials was used to closely locate positions of fiducials blocked by a multileaf collimator.Conclusion: These four studies pave the way to automatically track nearly real‐time 3D positions of multiple moving fiducial markers during treatment based on simultaneous kV and MV imaging. It does not require extra hardware such as stereo kV imaging systems and utilizes the treatment beam to reduce dose.

TH‐C‐AUD C‐06: Nearly Real‐Time Tumor‐Position Monitoring During Arc Therapy with Combined MV and KV Imaging

Purpose: To examine the feasibility and accuracy of using treatment MV beam and on‐board kV imaging for monitoring of the positions of implanted fiducials during arc therapy. Method and Materials: A Varian Trilogy LINAC with onboard kV imager was used for the study. A phantom with 13 ball bearings at known locations was used to calibrate the hybrid MV/kV imaging system to determine the spatial transformation matrix from the pixel coordinates to the radiation‐source‐centered coordinates. The feasibility and accuracy of the fiducial tracking system was examined using a 4D motion phantom capable of moving in accordance with a pre‐programmed trajectory. During an arc delivery, MV images acquired by the EPID and cine kV images were obtained simultaneously using a two‐channel frame‐grabber for real‐time analysis. A fast fiducial detection algorithm was developed to extract fiducial coordinates. Tracking results are compared with the preset trajectories. The accuracy of the tracking system was evaluated for a number of fiducial motion trajectories and for a variety of kV beam sampling rates ranging 15fps to 0.5fps. Results: The studies showed that it is feasible to use treatment MV beam and the orthogonal kV beam to monitor the fiducial motion in nearly real‐time during arc therapy. A time delay of ∼150ms was observed, which was caused by the imaging electronics and data analysis. This delay does not pose any significant error in fiducial tracking for the motion speed commonly seen in the clinics and can be compensated by a motion prediction algorithm when needed. Overall, the spatial accuracy was found to be better than 1mm. Conclusion: Nearly real‐time monitoring of implanted markers using hybrid MV/kV imaging during arc treatments is achievable. The system requires no hardware modification and delivers much less dose to the patient as compared with the conventional stereoscopic imaging technique.

Method for automatic localization of MR‐visible markers using morphological image processing and conventional pulse sequences: Feasibility for image‐guided procedures

To evaluate the feasibility and accuracy of an automated method to determine the 3D position of MR-visible markers. Inductively coupled RF coils were imaged in a whole-body 1.5T scanner using the body coil and two conventional gradient echo sequences (FLASH and TrueFISP) and large imaging volumes up to (300 mm(3)). To minimize background signals, a flip angle of approximately 1 degrees was used. Morphological 2D image processing in orthogonal scan planes was used to determine the 3D positions of a configuration of three fiducial markers (FMC). The accuracies of the marker positions and of the orientation of the plane defined by the FMC were evaluated at various distances r(M) from the isocenter. Fiducial marker detection with conventional equipment (pulse sequences, imaging coils) was very reliable and highly reproducible over a wide range of experimental conditions. For r(M) </= 100 mm, the estimated maximum errors in 3D position and angular orientation were 1.7 mm and 0.33 degrees , respectively. For r(M) </= 175 mm, the respective values were 2.9 mm and 0.44 degrees . Detection and localization of MR-visible markers by morphological image processing is feasible, simple, and very accurate. In combination with safe wireless markers, the method is found to be useful for image-guided procedures.

Analysis of Human Electrocardiogram for Biometric Recognition

Security concerns increase as the technology for falsification advances. There are strong evidences that a difficult to falsify biometric trait, the human heartbeat, can be used for identity recognition. Existing solutions for biometric recognition from electrocardiogram (ECG) signals are based on temporal and amplitude distances between detected fiducial points. Such methods rely heavily on the accuracy of fiducial detection, which is still an open problem due to the difficulty in exact localization of wave boundaries. This paper presents a systematic analysis for human identification from ECG data. A fiducial-detection-based framework that incorporates analytic and appearance attributes is first introduced. The appearance-based approach needs detection of one fiducial point only. Further, to completely relax the detection of fiducial points, a new approach based on autocorrelation (AC) in conjunction with discrete cosine transform (DCT) is proposed. Experimentation demonstrates that the AC/DCT method produces comparable recognition accuracy with the fiducial-detection-based approach.

Automated Fiducial Marker Detection for Patient Registration in Image-Guided Neurosurgery

Objective: The registration of applied fiducial markers within the preoperative data is often left to the surgeon, who has to identify and tag the center of each marker. This is both time-consuming and a potential source of error. For this reason, the development of an automated procedure was desirable. In this study, we have investigated the accuracy of a software algorithm for detecting fiducial markers within the navigation data set. The influence of adjustable values for accuracy and threshold on the sensitivity and specificity of the detection process, as well as the time gain, was investigated.Patients and Methods: One hundred MP-RAGE MRI data sets of patients with different pathologies who were scheduled for image-guided surgery were used in this study. A total of 591 applied fiducial markers were to be detected using the algorithm of the software VVPlanning 1.3 (BrainLAB, Heimstetten, Germany) on a Pentium II standard PC. The size value of a marker in the y-direction is called “accuracy” and depends on the slice thickness. “Threshold” describes the gray level above which the algorithm starts searching for pixel clusters. The threshold value was changed stepwise on the basis of a constant “accuracy” value. The “ccuracy” value was changed on the basis of that threshold value at which all markers were detected correctly.Results: The time needed for automatic detection varied between 12 s and 25 s. An optimum value for adjustable marker size was found to be 1.1 mm, with 8 undetected markers (1.35%) and 7 additionally detected structures (1.18%) out of 591. The mean gray level (Threshold) for all data sets above which marker detection was correct was 248.9. The automatic detection of markers was good for higher gray levels, with 11 missed markers (1.86%). Starting the algorithm at lower gray levels led to a decreased incidence of missed markers (0.17%), but increased the incidence of additionally detected structures to 27.92%.Conclusion: The automatic marker-detection algorithm is a robust, fast and objective instrument for reliable fiducial marker registration when used with optimum settings for both threshold and accuracy.

Multiring Fiducial Systems for Scalable Fiducial-Tracking Augmented Reality

In augmented reality (AR), a user can see a virtual world as well as the real world. To avoid registration problems between the virtual world and the real world, the user's viewing pose in both worlds should be kept the same. Fiducial-tracking AR is an attractive approach to the registration problem. However, most of the developed fiducialtracking AR systems have restricted workspaces. To provide a wide range of workspaces (from a small-scale desktop space to a large-scale space) and a wide range of views (from far views to detailed views), an AR system should have scalability. In this paper, we present multiring color fiducial systems and a real-time fiducial detection method for scalable fiducial-tracking AR. We analyze the optimal ring width and develop formulas to obtain the optimal fiducial set with applicationspecific inputs. We develop a real-time ring-detection method that converts the five-DOF ellipse-detection problem to a series of simple steps with a 1-D segmentfilter and multithreshold segmentation. The results lead to a simple and inexpensive means of achieving scalable-area tracking for AR and an approach that is suitable as an optical tracking method for VR as well.