• Home
• Preliminary Program
• Invited Speakers
• Speaker/Poster Information
• Conference Topics
• Programme Information
• Research Market/Beer & Wine Party
• Awards
• Company Market
• Important Dates
• Registration and Venue
• Authors Instructions
• Organizing Board
• Contact

13:00   Imaging Ultrasound II 13:00 15 mins SPATIAL CALIBRATION OF 2D/3D ULTRASOUND WITH THE PLUS FRAMEWORK AND ELECTRO-MAGNETIC TRACKING Erwin Vast, Pierre Ambrosini, Theo van Walsum Abstract: Introduction: Ultrasound imaging is an easy and cheap modality for imaging patients for diagnosis or during interventions. Fusion of pre-operative MR and CT imaging data may improve ultrasound-guided interventions. Adequate fusion imaging currently requires spatial tracking by e.g. attaching a sensor to the ultrasound transducer. In this case, spatial calibration is required; this determines the transformation matrix to convert B-scan image coordinates to the tracking coordinate space. The purpose of this work is to evaluate a free-hand calibration technique for spatial calibration of 2D and 3D transducers. Method: We use the Plus framework [1][2] with the accompanied phantom for 2D calibration. An NDI electromagnetic tracking system is used with a tabletop field generator and sensors. The sensors are attached to the ultrasound transducer and the phantom. Live imaging data from our Philips iU22 ultrasound system is send over a proprietary link to our workstation, which sends the imaging data over the OpenIGTLink protocol (implemented in medical visualization software MeVisLab [3]) to the Plus framework. Ultrasound scans are performed in saline water to match the speed of sound in human tissue. For 3D transducers we apply the 2D calibration using the middle slice of the ultrasound volume. The calibration precision is determined by performing the calibration multiple times and comparing the spatial calibration matrices. The accuracy is calculated by comparing the position of a fiducial marker, as measured with the tracker system, with the same position determined from ultrasound imaging data. We used a reflective ball of 1.5 mm as our fiducial which is scanned from multiple angles. Results: Preliminary results show for 2D transducers a mean precision of 1.1 mm and a mean accuracy of 2.3 mm. The accuracy could be affected by the transducer interfering with the tracker, imprecise localization of the ball on the ultrasound scans and the large distance between the transducer sensor and the tabletop field generator. Furthermore, the accuracy test includes images at wide angles as a worst-case scenario. Conclusion & future work: We conclude that spatial calibration of 2D and 3D transducers is feasible with the Plus framework, yielding accuracies of a few mm. The focus of current work is to improve the accuracy and to evaluate the Plus framework for 3D calibration. REFERENCES [1] Andras Lasso, Tamas Heffter, Adam Rankin, Csaba Pinter, Tamas Ungi, and Gabor Fichtinger, "PLUS: open-source toolkit for ultrasound-guided intervention systems", IEEE Trans Biomed Eng, 2014 October; 61(10): 2527-2537. [2] Guillermo Carbajal, Andras Lasso, Álvaro Gómez, Gabor Fichtinger, “Improving N-wire phantom-based freehand ultrasound calibration”, Int J Comput Assist Radiol Surg. 2013 July; 8(6): 1063–1072. [3] Egger J, Tokuda J, Chauvin L, Freisleben B, Nimsky C, Kapur T, Wells W. “Integration of the OpenIGTLink Network Protocol for image-guided therapy with the medical platform MeVisLab”, Int J Med Robot. 2012 Sep; 8(3):282-290. 13:15 15 mins WALL STRESS ANALYSIS OF ABDOMINAL AORTIC ANEURYSMS USING THREE-DIMENSIONAL ULTRASOUND Annette Kok, V. Lai Nguyen, Peter Brands, Geert-Willem Schurink, Frans van de Vosse, Richard Lopata Abstract: An abdominal aortic aneurysm (AAA) is a local dilation of the aorta that will eventually rupture. Monitoring is already performed by assessing the maximum diameter with 2D ultrasound. Currently, the decision for intervention is based on the maximum diameter (> 5.5 cm). However, the need for a more reliable criterion is often sought for in the mechanics of the aneurysmal wall. Multiple studies on Computed tomography (CT) and Magnetic Resonance Imaging based wall stress analysis are found in literature. The rise of 3D ultrasound (US) enables US-based wall stress analysis, but will suffer from limited contrast, field-of-view (FOV) and penetration depth. However, if feasible, the use of US will enable monitoring, studies on growth and without the use of ionizing radiation or contrast agent. In this study, the feasibility of 3D US-based wall stress analysis is examined. A total of 12 patients in the age of 55 – 83, with a AAA of 55 – 90 mm in diameter, were imaged 1 to 12 weeks prior to elective surgery. A MyLab 70 (Esaote, NL), equipped with a mechanical 3D array (BC431, fc = 3.5 MHz) and an RF-interface was used to image the AAA. One to three 3D volumes were acquired, depending on the length of the aneurysm. The sweep direction was orthogonal to the vessel axis. Computed tomography angiography (CTA) was performed for intervention planning using a Somatom CT system (Samsung, KR) with a pixel size of 0.7 - 0.8 mm. All US data were manually segmented and different volumes were registered. Next, a radial smoothing was performed to reduce irregularities in the mesh. The CT-data were segmented using Mimics (Materialize, BE). The 3D volumes were compared in Matlab (Mathworks, USA) and the similarity index (SI) and Hausdorff distance (HD) were calculated. Next, for 8 patients the 3D US volumes were converted into volume meshes (wall thickness of 2 mm), consisting of 10 node quadratic tetrahedral elements. Wall stresses were simulated in ABAQUS (Dassault Systems, FR) using the two-parameter constitutive material model proposed by Raghavan & Vorp. The initial stresses at mean arterial pressure were reconstructed using a backward incremental method, after which the AAA was inflated to 140 mmHg. The SI of US vs CT was 0.75 – 0.91 (n = 12), with a median HD of 5 – 15 mm, with the higher values found at the proximal and distal sides of the AAA. Wall stresses were in accordance with literature and a good agreement was found between US and CT-based median stresses and inter-quartile stresses (n = 8), which was confirmed by Bland-Altman and regression analysis. Wall stresses based on US were typically higher (+23% on average) caused by geometrical irregularities due to the registration of several 3-D volumes and manual segmentation. In future work, an automated US registration and segmentation approach is the essential point of improvement before pursuing large-scale patient studies. This study is a first step towards US based wall stress analysis. US will enable monitoring of wall stress development over time, since no ionizing radiation and contrast are involved. 13:30 15 mins 2D ULTRASOUND STRAIN IMAGING OF THE SUPPORTED LEFT VENTRICLE Niels Petterson, Marcel Rutten, Sjoerd van Tuijl, Frans van de Vosse, Richard Lopata Abstract: Heart failure (HF) is a common disease which is characterized as the inability of the heart to provide adequate pump function to supply sufficient blood to the body. HF can have all sorts of pathophysiological causes, such as coronary artery disease, heart attacks, heart valve disease and high blood pressure. The treatment of mild HF can be managed with medicine, however, severe cases of HF can only be treated by replacement of the heart. With the limited availability of donor hearts, mechanical assist devices form a temporary solution to keep the patient alive until a donor heart is available. Some patients supported by an assist device show recovery of heart function. This leads to the idea that it is possible to create a ’nurturing’ state using an assist device to induce heart recovery. In this study an existing 2D echography strain imaging technique is used to measure local strain and strain rate in ex vivo supported hearts, which provides a first step towards the assessment of mechanical parameters in vivo. Ex vivo beating porcine hearts1 (LifeTec group, NL) fitted with HeartMate II (Thoratec, US) left ventricular assist devices were imaged. A Mylab 70 (Esaote, NL), equipped with a curved array transducer (fc = 2.7 MHz) and an open radio frequency (RF) interface was used to image the left ventricle. Several heartbeats in three short axes views, ranging from apical to mitral, were acquired. Pump speeds were varied from 6 to 10 krpm in increments of 1 krpm. In addition, an unsupported case was analyzed where the pump was turned off. Pre-shock (CO = 4.5 L/min; MAP = 75 mmHg) and shock (CO = 3 L/min; MAP = 60 mmHg) conditions were acquired by natural degeneration of the heart. The cardiac wall in the short axis views was segmented and a coarse-to-fine strain imaging algorithm2 was used on RF data to acquire strain and strain rate images of the left ventricular myocardial wall. Mean radial strain and strain rates in the left ventricle at apical and mid views decreased with higher pump speeds. Mitral views showed no significant change in strain or strain rate with varying pump speeds. The contraction pattern of the cardiac wall showed very dyssynchronous characteristics, possibly due to implantation trauma, and strain in the lateral wall was underestimated. Future work should include long term monitoring of patients with mechanical support and an improved strain imaging algorithm to reduce underestimation of strain. This will enable the investigation of the nurturing state in the supported heart. 13:45 15 mins USING AN INJECTION SIGNAL TO REDUCE THE EFFECT OF CAPACITANCE CHANGES IN CAPACITIVE ECG RECORDINGS Aline Serteyn, Rik Vullings, Mohammed Meftah, Jan Bergmans Abstract: Capacitive sensors allows to record the electrocardiogram (ECG) of patients in a more comfortable and unobtrusive manner than the conventional adhesive electrodes; they allow the recording of an ECG through insulating materials, e.g. textiles, and can thus be embedded in everyday objects like beds or car seats. Capacitive ECG sensing however still suffers from motion artifacts. More specifically, any distance change at the body-sensor interface (e.g. due to motion) causes a change of the coupling capacitance, which directly affects the recording. The motion artifact consists of a multiplicative and an additive artefact component. The latter dominates since it is proportional to the DC voltage across the coupling capacitor and often masks the features of interest of the ECG signal. Our on-going work consists in injecting a known signal through the recording system to track and compensate for the changes of the coupling capacitance and therefore reduce the motion artifact. A first method was proposed based-on a single-frequency injection signal. An identification scheme derived from a model of the capacitive sensing system was used to estimate the additive artefact and subtract it from the recording. An alternative method consists in reconstructing the ECG with a filter whose parameters are estimated based on a multi-frequency injection signal. The two proposed methods assume that the body-sensor interface is purely capacitive and that the motion causes a significant capacitance change. Some preliminary tests in simulation and lab environment showed that both methods lead to a significant improvement of the quality of the ECG recordings. The developed methods are currently further evaluated and compared in terms of performance and robustness for different recording scenarios. Motion artifacts due to capacitance changes are not the only interferences corrupting capacitive biopotential recordings. However, a proper understanding and reduction of these artifacts is a first step towards a robust capacitive system enabling unobtrusive ECG recordings on dressed patients at the hospital or at home. 14:00 15 mins AUTONOMOUS WIRELESS SENSOR FOR ECG MONITORING Andre Luis Mansano, Yongjia Li, Wouter Serdijn Abstract: Currently, the electronics market is hoping to see great technical developments on wirelessly connected sensors that can be used in many applications, in particular for biomedical signals monitoring that is foreseen to represent a large portion of this market in the near future. The role of researchers and industry is to fulfil the high expectation of the biomedical market by bringing new and smart solutions. Current solutions, such as the ones described in [1]-[4], implement promising wireless ExG monitoring systems using low power signal processing and low power synchronous data transmission. However, for such an application there is a need to reduce the power consumption even further while maintaining performance. In this work, an asynchronous way to monitor ECG signals is proposed to reduce the power consumption of the entire wireless sensor. The implementation consists of an asynchronous Analog to Digital Converter (ADC) combined with a Low Noise Amplifier (LNA) and a Programmable Voltage to Current Converter (PVCC) to process ECG signals. In addition, a Radio Frequency (RF) energy harvester is designed to supply the sensor from an RF supply at 13.56 MHz. The ECG signal processed and converted in the sensor is transmitted by means of a passive transmitter that makes use of the Medical Implant Communication Service (MICS) 402 MHz frequency band. The autonomous wireless sensor has been designed and tested to verify its functionality and performance. It offers an 8 bit signal conversion resolution, 90 kbps data rate, which represents 35% of data compression compared with [1] and [2], when powered by a -13 dBm RF signal at 13.56 MHz. An ECG signal of 8mVpp has been applied to the sensor input and the RF data transmitted through the MICS band has been reconstructed. As a final result, the power consumption of the proposed design is 40% lower than the state of the art. 14:15 15 mins DETECTION OF FETAL MOVEMENTS FROM SINGLE LEAD ABDOMINAL ECG RECORDINGS Michiel Rooijakkers, Chiara Rabotti, Massimo Mischi Abstract: Fetal motility is a valuable indicator of fetal health, as a decrease can be seen as a precursor to fetal death, sometimes by as much as several days [1]. Various systems for automatic detection of fetal movements exist, the most reliable of which is continuous ultrasound imaging with manual identification by a medical expert and, therefore, not suitable for long-term observation. Automatic methods based on abdominal ECG measurements are suitable for long-term ambulatory use, which can greatly increase their predictive value. State-of-the-art methods, like the vectorcardiographic loop alignment based method by Vullings et al. [2], are currently to computationally complex for ambulatory use. Therefore, we propose a new method for fetal motion detection, based on variations in the fetal QRS complex morphology. Like Vullings et al., the proposed method is based on the premise that any movement of the fetal cardiac vector results in changes in both QRS amplitude and shape, which are related to translational and rotational movements with respect to the abdominal electrodes, respectively [3]. Therefore, the changes in the amplitude and the correlation coefficient are extracted from cleaned fetal QRS complexes, obtained from a single abdominal lead, and used as two independent features MT and MR. Linear regression is used to obtain a threshold in the two-dimensional space spanned by MT and MR to classify between fetal movement and fetal rest. The method’s classification quality was validated using four 30 minute abdominal ECG measurements on pregnant women between the 22nd and 27th week of gestation. Two abdominal electrodes were placed near the center and lower right of the abdomen, respectively. Simultaneous ultrasound recordings were performed using an Aloka SSD1100 ultrasound device for classification between episodes with and without fetal movement. Classification was done by visual inspection of the ultrasound recordings by a medical expert. When applying the algorithm, an overall sensitivity of 0.67 and specificity of 0.90 is obtained on the presented dataset. This compares favorably to the results of Vullings et al. [2], which obtains a sensitivity and specificity of 0.47 and 0.87, respectively, and is comparable to that of maternal perception. Because the method operates on only a single bipolar abdominal channel it is well suited for long-term ambulatory registration of fetal motility. REFERENCES [1] A. D. Bocking, "Assessment of fetal heart rate and fetal movements in detecting oxygen deprivation in-utero," Eur J Obstet Gynecol Reprod Biol, vol. 110, pp. S108-S112, 2003. [2] R. Vullings, M. Mischi, S. G. Oei and J. W. M. Bergmans, "Novel Bayesian vectorcardiographic loop alignment for improved monitoring of ECG and fetal movement," IEEE Trans. Biomed. Eng., vol. 60, no. 6, pp. 1580-1588, Jun. 2013. [3] M. J. Rooijakkers et al., "Influence of Electrode Placement on Signal Quality for Ambulatory Pregnancy Monitoring," Comput Math Methods Med., vol. 2014, p. 12, 2014.