Implantable RF telemetry for cardiac monitoring in the murine heart: a tutorial review
- 7.1k Downloads
Research and development of implantable RF telemetry systems intended specifically to enable and support cardiac monitoring of genetically engineered small animal subjects, rats and mice in particular, has already gained significant momentum. This article presents the state of the art review of experimental cardiac monitoring telemetry systems, with strong accent on the systems designed to work with a dual pressure–volume conductance-based catheter sensor. These commercially available devices are already small enough to fit inside a left-ventricle of a mouse heart. However, if the complete system is to be fully implanted and the subject allowed to freely move inside a cage, the mouse’s small body size sets harsh constrains on the size and power consumption of the required electronics. Consequently, significant portion of the research efforts is directed towards the development of low-volume and -power electronics, as well as RF energy harvesting systems that are required to serve as the energy source to the implanted telemetry instead of the relatively very bulky batteries.
KeywordsRing Electrode Cardiac Monitoring Conductance Catheter Sensor Interface Power Supply Rejection Ratio
Consequently, today’s biomedical researchers still rely on small, heavily distorted samples of PV data collected by the external data-collection unit during an open heart operation while the subject is fully anesthetized. Thus, the lack of micro-sized implantable telemetry systems for cardiac monitoring remains an important problem. Therefore, it is extremely important to develop a fully implantable cardiac telemetry microsystem that can safely fit into small mouse’s body and, eventually, be permanently implanted into a human heart . Then, the researchers will be able to continuously monitor the subject’s heart condition by collecting real-time data over longer periods of time while, in the case of the small animal subjects, the subject is fully conscious and freely moving inside a cage specifically designed for that purpose, i.e. inside a cage with the embedded communication antennas and RF energy harvesting infrastructure.
RF telemetry has already become an indispensable tool in clinical applications. A “radio pill” first reported in , over the time, has evolved into a fully functional “video pill”  that is now almost routinely used to monitor pH, temperature, and the human digestive system from the inside. The contemporary research on cardiac RF telemetry systems for a murine heart monitoring, however, may be broadly grouped under the following four themes:
Conductance catheter model
Accurate modelling of the relationship between the blood conductance and the beating heart’s volume is a very complex problem, which forces the researchers to rely on onerous use of numerical tools to develop practical working models . Today, Baan’s linear equation  and Wei’s nonlinear equation  form the core of the conductance volumetric catheter model that is currently accepted by the researchers.
However, first apparent limitation of the conductance catheter based methodology is due to the random finite resistance of a heart’s muscle, which causes dynamic fluctuations and the current leakage through the heart walls [14, 15, 16]. Second drawback of this volumetric method is its relativity, thus the need for the calibration methodology [15, 17].
At the same time, the alternative noninvasive reported methods for the volumetric measurements are mostly based on ultrasound scanning , and three-dimensional (3D) microcomputed tomography (CT) imaging techniques . However, these methods require a full size scanning equipment located in the laboratory, and numerically intensive post-processing of the data. In this article, we focus specifically on conductance catheter based invasive cardiac monitoring methods, while the detailed review of the noninvasive methods is beyond the scope of this article and will be addressed in another publication.
Wireless energy transfer
Although the reported energy harvesting techniques exploit various physical principles , due to the power consumption level and system size constrains specific to the mouse cardiac implant application, choice of the applicable power sources is drastically limited. Hence, the inductive resonance-based coupling techniques are emerging as the leading method , with various optimization proposals recently reported in [22, 23, 24]. In this article we focus specifically on inductive resonance-based coupling technique for RF energy harvesting.
Design of a biocompatible package and antenna intended for RF telemetry systems is, by all measures, very involved phase of the overall design process. Occasionally, design details of implantable micro-telemetry package itself are reported within the context of the overall system [24, 25, 26], while most of the published studies related to design of a package intended for various implantable systems focus only on the specific technological steps . It is interesting, however, to observe that most current implantable designs, for instance a commercially available Cochlear implant, RF telemetry system in , and an active RFID micro tag , follow the same traditional system-level design approach. That is, the system package and the antenna are usually designed and manufactured as the two separate entities. Recently, the researchers are reporting more creative approach to the antenna/package assembly design, as is partially reported in , which resulted in the novel use of a well established medical stent device as structural support of the implant as well as an antenna for simultaneous wireless telemetry and powering. Thus, the electro-mechanical design process of a biocompatible package and the accompanying antenna for RF power and data transfer is still as much an art as it is engineering.
Conductance based catheter
Baan’s linear model
As a material, blood has both conductive and dielectric properties, hence, each inner segment of the heart volume is considered as the equivalent resistor in parallel with a capacitor, whose height is determined by the inter-electrode pair distance L and time-varying median cross-sectional area A(t) (Figure 3).
In deriving his linear model Baan et al.  makes the following four crude assumptions: (a) the intracardiac electric field distribution is uniform; (b) the ventricular wall is perfectly insulated from the cavity blood, i.e. the total measured conductivity is strictly due to the blood and not to the heart’s muscle; (c) a heart cavity has a cylindrical shape; and (d) the catheter is stationary and always perfectly centred along the cylinder’s axis.
where, each volume section Δ V n (t) is calculated as in (5). Additionally, this method produces relative measurement that is function of the blood conductance σ, therefore the absolute value of the conductance must be determined beforehand by using an independent measurement methods.
where, k is an empirical slope correction factor, and V c is the linear offset empirical correction factor. The two empirical linear correction factors are determined by comparing values for the volume V(t) as calculated from the analytical expression (7) with the experimentally determined value after using an electromagnetic flow probe or Doppler ultrasound volume measurement method. Nevertheless, we conclude that the first order linear approximation model is overly simplified.
Wei’s nonlinear model
In order to compensate for the intrinsic error caused by the nonuniform internal electric field, while still keeping the other assumptions used in the linear model, Wei et al.  suggested the following nonlinear model.
where I is current (A), V is voltage (V), is electric field intensity (V m), is current density (A m2), a is a surface enclosing the source electrode, l is the path length for potential calculation, and σ is the blood conductivity (the reciprocal of blood resistivity ρ).
Consequently, the performance of the analytic approximation (11) is not fully satisfying either, since the analytic approximation was derived under the assumption that the electrodes are placed in a large medium. Further experimental empirical corrections should be applied again .
The issue of parasitic conductance of the heart muscle, which from electrical perspective is parallel to the blood conductance, is addressed in [14, 32, 33]. Due to this parasitic conductance, the traditional conductance based volume measurement method fails to accurately correct for the parallel conductance contributed by the myocardium, resulting in overestimation of blood volumes.
and, Δ σ=(σ m −σ b ), Δ ε=(ε m −ε b ), and β=f(S V,Y) is an empirical calibration factor that is meant to compensate for the error due to the use of cylindrical model. Muscle admittance (Y m ) with boundary condition at R=∞ is Yinf.
The calibration issue
Wireless energy transfer
Commonly, mobile devices depend on a battery as the energy source. However, a battery’s energy density does not scale at the same rate as an IC’s physical size. Consequently, the battery size has become major bottleneck on the path to further miniaturization of mobile systems, which is especially important parameter for the implantable telemetry.
Overcoming this issue requires adaptation of alternative energy sources. For instance, recently, new developments and fabrication technologies have resulted in micro fuel cells  that are becoming competitive with the state of the art batteries. However, a holy grail of bioelectronics is to engineer biologically implantable systems that can be embedded without disturbing their local environments while harvesting from their surroundings all of the power they require. In particular, most important question is whether the implantable electronics can be powered by drawing the required energy from their surrounding tissues .
<1 μ W/cm2
A few mW with a short distance inductive coupled systems 
100 mW/cm2 (direct sunlight)
Assuming common polycrystalline solar cells at 16–17%
100 μ W/cm2 (office light)
efficiency, while standard monocrystalline cells approach 20% at Δ T=5°C; typical thermoelectric generators ≤1% efficient for Δ T<40°C
60 μ W/cm2
at Δ T=5°C; typical thermoelectric generators ≤1% efficient for Δ T<40°C.
4 μ W/cm3 (human)
Predictions for 1 cm3 generators
800 μ W/cm3 (machine)
Demonstrated in microelectromechanical turbine at 30 liters/min
50 μ J/N
MIT Media Lab Device
Nissho Engineering’s Tug Power
10−800 μ W
7 W potentially available (1 cm deflection at 70 kg per 1 Hz walk)
The basic idea of wireless energy transfer based on Tesla’s coils, that is currently being exploited in many forms and modifications, is now a century old . In the first approximation, the power density at the receiving RF antenna is produced through E/Z0 relation, where Z0=377 Ω is the radiation resistance of free space, and E is the local electric field strength. An electric field of E=1 V m thus yields about 0.26 μ W/cm2. This crude analysis illustrates that the freely available RF energy is very limited, unless the radiation levels are set to dangerously high level, or the receiving antenna is very close to the transmitter. Which further imposes the upper limit of the total power that can be safely transmitted through a living tissue. Currently, as a “rule of a thumb”, most researchers arbitrary set the upper total power budget for the implantable telemetry to a few hundred μ W, i.e. less than approximately 300−500 μ W.
Thus, being able to control intensity and direction of the electro-magnetic filed vector is very important, because if energy being transmitted through a living tissue has too high radiation density level, then it can permanently damage the cells it passes through. In other words, the design of a mouse cage with the embedded RF energy harvesting system, Figure 8, is not much different from a microwave oven design.
The reason for ambiguity related to the accepted power levels is that, despite a large body of publicly available literature related to the topic of mobile phone radiation and health, there is no consensus on what energy level exactly is considered to be dangerous for the living tissue. Discussion on that topic is beyond the scope of this article, hence, in our research we take conservative approach and keep the transmitted energy levels significantly below those generated by modern cellphones.
Considering that the RF energy transmission efficiency is the main design parameter, the bottleneck of the remote powering link is generally at the inductive link because the coupling factor between the coils of the inductive link is usually very small. Therefore, these coils should be designed properly to achieve high power transfer efficiency [21, 22].
Inductance coupling background
Since the maximum power transfer can only be achieved when the external and implanted inductors are perfectly aligned, the challenge is to design a powering system that would have low sensitivity to the coil orientation and distance [21, 42]. Such designs, which are mainly focused on the generation of constant minimum power level inside the subject’s cage, have been investigated in .
while denotation of all variables in (17) to (19) follows .
Most modern IC rectifier/charge-pump topologies are derived from the conventional multistage Dickson circuit , with modern variants based on self- Vth-cancellation (SVC) methodology published , where each two-diodes–two-capacitors stage acts as a voltage doubler, therefore contributing the voltage overdrive to the output, where is the peak input voltage while V D is the diode turn-on voltage, and the diodes are implemented by MOS devices in CMOS technology. Hence, V D is equivalent to a MOS threshold voltage, |Vth|.
Utilizing energy collected at the receiver coil requires some sort of feedback control circuit that includes a rectifier, regulator and a bandgap voltage reference to deliver a stable, load-independent voltage to the circuit .
The widely fluctuating voltage generated by the charge-pump is regulated by a series regulator, which by itself has to work with very low power and low-voltage , if it is to be useful for implantable medical telemetry systems. Schematic block diagram of our regulator architecture, Figure 15, shows that a compromise between the voltage drop across M00 and overall PSRR is made based on the specific application conditions.
The voltage reference Vre f is then routed back to the input of the loop. The loop is designed to be stable with wide range of load impedances, with full load set up to R L =250 Ω and C L =50 p F. The maximum load corresponds to the maximum current drawn from VPW R, which is designed to be IPWR(max)=4 mA, which is far above the needs of our implantable telemetry electronics. By powering the reference circuit (BG) from VPW R, the overall PSRR of the Vre f is further improved. In this design, the regulated supply voltage is set to VPWR=0.985 V while the complete regulator consumes less than 11.6 μ A current (typically). The reference voltage also serves as a load to the regulator loop, which helps to keep the loop stable when its load is at the minimum.
Power supply rejection ratio (PSRR)
RF telemetry system
In the previous sections we have seen that a batteryless low-power implant electronic system is critical for realizing the RF implantable wireless telemetry [51, 52]. Once the harvesting energy system designed, the remaining part of the implantable telemetry consists of the sensor interface, RF transmitter, and the controlling logic (Figure 9).
PV sensor interface
Biocompatible package and antenna
A living organism presents an extremely hostile environment for electronic circuits, which puts very harsh requirements on the overall system design, including constrains on specifications of the implantable package. For instance, while the specific shape and size of an eye predetermine possible options in terms of the system packaging and the antenna design [54, 55], the overall eye volume provides relatively comfortable volume for implanting the modern HF antennas . In addition, the design methodologies of implantable flexible antennas suitable for biomedical research remain vigorously pursued topic .
For instance, a typical system level assembly includes an active integrated circuit connected to the external RF coil, and then the assembly is sealed with a biocompatible material . Consequently, each section of the system appears to be designed separately using different design tools, which are then integrated at the system’s top level.
On the other hand, when a larger volume is available inside the patient to place the complete telemetry/antenna assembly, then there is also a room for a more creative approach and more integrated overall package design. As an example, Chow et al.  presented their package/antenna analysis within the content of the medical stent application. In our design, we aim to derive the minimum volume system that includes the electronics, the package, and the antenna.
A number of research groups around the world have already focused their respective efforts on development of implantable telemetry technologies. Thus, when the core of an RF telemetry system for cardiac monitoring is reduced to (2 × 2 × 2) mm3 volume or less and, more importantly, when it is interfaced with various sensors it will drastically expand the list of potential industrial and scientific applications. Eventually, a wireless sensor network will be embedded, for instance, into crop fields, various constructions, and into a human body. That will enable real-time monitoring of growing crops, bridge integrity, or human health. This will then open up a wide range of other possible new applications for this potentially disruptive technology.
The author would like to express sincere gratitude to Scisense Inc, OCE, NSERC, CFI, and CMC Microsystems for providing support for our research. Additionally, I acknowledge contribution of Mr. Sorin Popa who created some of the plots and verified data used in this article. Also, I acknowledge cited contributions from the referenced sources used in this review.
- 1.Statistics CanadaMortality, Summary List of Causes. Catalogue no. 84F0209X, Table 1–1: 14, 2008Google Scholar
- 6.Scisense Inc. . Accessed 2012/11/02 http://www.scisense.com/
- 7.Millar Inc. . Accessed 2012/11/02 http://www.millar.com/
- 8.Herrera MC, Olivera JM, Gómez López MdlA: Implantable hemodynamic monitors: can be conductance catheter system successfully implemented? In IEEE 32nd International Conference EMBC. Buenos Aires, Argentina; 2010:3549-3552.Google Scholar
- 9.Zworykin VK, Farrar JT: A ‘radio pill’. Nature 1957, 179: 898. [A note]Google Scholar
- 11.Carson JP, Kuprat AP, Jiao X, del Pin F, Einstein DR: An anisotropic fluid–solid model of the mouse heart. Comput. Cardiol 2009, 36: 377-380.Google Scholar
- 14.Kottam A, Dubois J, McElligott A, Henderson HK: Novel approach to admittance to volume conversion for ventricular volume measurement. In IEEE 33rd International Conference EMBS1. Boston MA, USA; 2011:2514-2517.Google Scholar
- 28.Schuettler M, Huegle M, Ordonez JS, Wilde J, Stieglitz F: A device for vacuum drying, inert gas backfilling and solder sealing of hermetic implant packages. In IEEE 32nd International Conference EMBC. Buenos Aires, Argentina; 2010:1577-1580.Google Scholar
- 29.Lin TY, Ha D, de Vries WN, Kim B, Chlebowski A, John SWM, Irazoqui PP, Chappell WJ: Ultra-thin tag fabrication and sensing technique using third harmonic for implantable wireless sensors. In IEEE MTT-S International Microwave Symposium Digest (MTT). Baltimore, MD USA; 2011:1-4.Google Scholar
- 30.Dekker AL, Geskes GG, Cramers AA, Dassen WR, Maessen JG, Prenger KB, van der Veen FH: Right ventricular support for off-pump coronary artery bypass grafting studied with bi-ventricular pressure-volume loops in sheep. Eur. J. Cardio-thoracic Surg 2001, 19: 179-184. 10.1016/S1010-7940(00)00635-7CrossRefGoogle Scholar
- 32.Wu CC, Skalak TC, Schwenk TR, Mahler CM, Anne A, Finnerty PW, Haber HL, Weikle RMI, Feldman MD: Accuracy of the conductance catheter for measurement of ventricular volumes seen clinically: effects of electric field homogeneity and parallel conductance. IEEE Trans. Biomed. Eng 1997,44(4):266-277. 10.1109/10.563296CrossRefGoogle Scholar
- 33.Porterfield JE, Kottam ATG, Raghavan K, Escobedo D, Jenkins JT, Larson ER, Treviño RJ, Valvano JW, Pearce JA, Feldman MD: Dynamic correction for parallel conductance, GP, and gain factor, α, in invasive murine left ventricular volume measurements. J. Appl. Physiol 2009, 107: 1693-1703. 10.1152/japplphysiol.91322.2008CrossRefGoogle Scholar
- 34.Popa S, Fricke K, Sobot R: Calibration and finite element numerical analysis of an admittance based volume sensor for the murine heart. In IEEE International Conference for Upcoming Engineers, ICUE. Toronto ON, Canada; 2012:12-13.Google Scholar
- 39.Tesla N: Apparatus for transmitting electrical energy. Patent number:1,119,732, Filing date: 18 Jan, 1902,Issue date: 1 Dec, 1914Google Scholar
- 42.Luo L, De Gannes K, Fricke K, Senjuti S, Sobot R: Low-power CMOS voltage regulator architecture for implantable RF circuits. In The 4th International EURASIP Workshop on RFID Technology, RFID 2012. Italy: Torino; 2012.Google Scholar
- 44.Senjuti S, Fricke K, Dounavis A, Sobot R: Misalignment analysis for resonance-based wireless power transfer to biomedical implants. In IEEE Canadian Conference on Electrical and Computer Engineering, CCECE2012. Canada: Montreal QC; 2012:1-5.Google Scholar
- 49.Zargham M, Gulak P: High-efficiency CMOS rectifier for fully integrated mW wireless power transfer. In IEEE International Symposium on Circuits and Systems (ISCAS). Seoul, Korea: Zarg12; 2012:2869-2872.Google Scholar
- 52.Fricke K: Wireless telemetry system for implantable sensors. Master’s thesis, Western University, Canada, 2012Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.