A SENSORIZED PLATFORM TO MONITOR AND REGULATE VAD

FUNCTIONALITIES

Wearable and Implantable Devices to Improve VAD Performances and Extend its

Field of Application

N. Taccini, M. Nannizzi, C. Ciaponi, P. Valdastri and P. Dario

CRIM-Lab, Scuola Superiore Sant’Anna, Pisa, Italy

Keywords:

Biomedical instrumentation, Implantable electronics, Electrical bio-impedance, Hemodynamics monitoring,

Health monitoring devices.

Abstract:

A novel system to control Ventricular Assist Device (VAD) is under study in the frame of the project SEN-

SORART (FP7-ICT-248763, funded by the European Commission). The SensorART platform will be open,

interoperable, extendable and VAD-independent and it will incorporate different hardware and software com-

ponents in order to improve both the quality of the patients’ treatment and the workﬂow of the specialists.

In the frame of this project a set of implantable sensors will be identiﬁed, developed and/or adapted for long

term implant, in order to gather all the necessary information to control the VAD functionalities by a wearable

hardware unit. Wearable sensors will be studied to assist the decision algorithms by gathering physiological

parameters using non invasive approach. Purpose of this paper is to give a brief description of the system under

study and describe the preliminary activity performed during the developing of the implantable and wearable

sensors. In particular, the evaluation and calibration of a novel catheter pressure sensor is reported and the

results of the feasibility of a tonometry wearable system is described.

1 INTRODUCTION

Heart failure (HF) is the most increasing cause of

death in Western Countries. For that reason, together

with the difﬁculty of having a sufﬁcient number of

donor organs, it is recognized that the device-based

therapeutic approaches will assume an increasingly

important role in treating the growing number of pa-

tients with advanced heart failure, not only as bridge

to transplant, but also as destination therapy. In this

scenario the project SENSORART (FP7-ICT-248763,

funded by the European Commission) leads to the de-

velopment of a complete system composed by hard-

ware and software devices able to control the Ventric-

ular Assist Device. Scope of this paper is to describe

the performed activity in the study of the sensor mod-

ule.

2 MATERIALS AND METHODS

2.1 System Overview

The proposed platform will be able to vary the func-

tional parameter of the VAD to support patients with

chronic heart failure. It will also provide informa-

tion to healthcare professionals, allowing patients to

be treated at home without renouncing to accessing

high medical expertise. The main research and de-

velopment key modules of the integrated platform

will be: Sensor Module, Signal Acquisition Mod-

ule, Hardware Controller and Remote Control Frame-

work. Figure 1 depicts the described platform in prin-

ciple.

Most VAD controllers derive now all their infor-

mation by the pump’s power consumption, using this

information predominantly as monitoring (to detect

suction, low ﬂow conditions, pump stoppage) and

to trigger alarms. The SensorART platform would

take advantage also from physiological signals ac-

quired by a mix of implantable and wearable sensors.

All the implantable and wearable sensors will wire-

lessly communicate with a wearable device. This unit

315

Taccini N., Nannizzi M., Ciaponi C., Valdastri P. and Dario P..

A SENSORIZED PLATFORM TO MONITOR AND REGULATE VAD FUNCTIONALITIES - Wearable and Implantable Devices to Improve VAD

Performances and Extend its Field of Application.

DOI: 10.5220/0003132103150318

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 315-318

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 1: SensorART platform.

will allow to auto-adjust the blood ﬂow provided by

the VAD to the patient’s heart according to signals

coming from physical and physiological sensors, by

means of a wireless link to the VAD actuators. More-

over, this unit will monitor the energy consumption,

as well as the VAD functionality, thus generating the

appropriate crucial and vital alert messages. Real-

time adjustable thresholds and control parameters will

be a peculiar feature of the proposed platform. The

auto-regulation control algorithm will be fully ad-

justable to the patient’s condition and will incorpo-

rate context-aware techniques in order to meet the pa-

tient’s needs according to his/her status. Physiologi-

cal sensors are usually invasive, with a complex pack-

age, and not reliable in long term implants. Within

the SensorART project, wearable approach will be ex-

ploited together with the implantable ones in order to

minimize risk of failures. This paper will describe

the actual stage of the project regarding the develop-

ment of implantable (see 2.2) and wearable sensors

(see 2.3).

2.2 Implantable Sensors

Main physiological sensors described in literature as

suitable for LVADs monitoring are: oxygen satura-

tion sensor (Nakamura et al., 2000), pressure sensor

(Bullister et al., 2001), ﬂow rate sensor (Waters et al.,

1999), and acceleration (Maeda et al., 1988) sensors.

The ﬁrst stage of the development of the implantable

devices consists in a screening of the cited parame-

ters during tests on laboratory mock-up and in-vivo,

in implantable or wearable version. After these trials,

the long term implant issue will be faced and a proper

package will be studied. Thanks to the collaboration

with clinical experts it has been possible to determine

a set of variables to be monitored by implantable sen-

sors in order to monitor the residual heart functional-

ities, the VAD function and the patient’s reaction to

the treatment. These parameters are: left atrial pres-

sure, aortic blood ﬂow rate, aortic pressure, blood

pressure and ﬂow in the VAD cannula, pulmonary

artery or right ventricular pressure. A previous work

of the authors (Valdastri et al., 2008) describes a wire-

less implantable platform for in vivo monitoring. A

similar approach has been used to integrate an inno-

vative pressure sensor made by STMicroelectronics,

that is still in evaluation phase (P30PCB). After the

necessary characterization phase, this sensor will be

encapsulated in a custom made catheter to perform

the ﬁrst session of tests. An electronic front-end has

been developed in order to acquire the signal derived

by the sensor inserted in a controlled pressure cham-

ber. The output of the sensor needs a differential read-

ing that is performed by an instrumentation ampliﬁer

(AD623N). The pre-ampliﬁed signal is then low pass

ﬁltered by an RC ﬁlter (band pass at 20 Hz) and am-

pliﬁed by a non inverting ampliﬁer, realized with a

CMOS operational ampliﬁer (LMC6482).

Figure 2 shows the system realized for the cali-

bration using a reference sensor (MPX5050 GP by

Freescale semiconductor). The acquisition circuit,

Figure 2: Calibration system for the pressure system.

based on a wireless microcontroller (CC2430 by

Texas Instruments) can communicate wirelessly (zig-

bee ready) or by USB connection to a PC. A graphi-

cal interface, acquires and stores data with a sampling

rate of 25 Hz by both test and reference sensors. The

pressure in the chamber has been varied step by step

in the range from 8 kPa to 35 kPa. The range has been

chosen taking into account the ﬁnal ﬁeld of applica-

tion, that is the radial artery pressure measurement.

The measures have been performed on both growing

and decreasing pressures, to evaluate the hysteresis of

the signal. Tests have been performed on 3 sensors

to assure the reproducibility. Three complete cycles

have been acquired for each sensor with pressure set

as following: 8 kPa, 16 kPa, 24 kPa, 32 kPa and 35

kPa. The calibration curves have been approximated

by polynomial ﬁtting by using Matlab software. Also

the stability of each sample has been tested, by means

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

316

of long acquisition (30 min) at stable pressure (8 kPa,

24 kPa and 35 kPa).

2.3 Wearable Monitoring Devices

One of the most relevant physiological signals to be

monitored is arterial blood pressure. This can be mon-

itored by either implantable or wearable sensors. In

particular, radial tonometry (Chemla et al., 2008), has

been explored to derive the central arterial pressure

by means of non-invasive technique. An example of

such type of device is the BPro by HealthSTATS

. In

the frame of the project a wearable system with wire-

less communication will be developed. In order to

monitor the central pressure 24/7 and pulse by pulse,

the system will be designed to assure the stability of

the signal in case of patient’s arm movements. Wear-

ability, ease of use and ergonomics will be carefully

considered. An ad hoc support has been produced by

fast prototyping with a 3D printer (Invision SI2 by

Inition), to investigate this technique. Figure 3 shows

the support and its placement on the wrist for the pre-

liminar tests.

Figure 3: P30PCB STM pressure sensor and the support for

the tonometry device (left), placement of the system on the

radial artery (right).

The support is designed in order to assure the arte-

rial applanation, while reducing the discomfort. Ap-

propriate slots allow to use an elastic ribbon to ﬁx

the support on the wrist. It is possible to regulate

the applied force on the skin by the protrusion with

the inserted sensor simply by varying the tension of

the elastic band. The pressure sensor is the STM

P30PCB described in 2.2. An additional sensing sys-

tem, that will be developed, regards impedance car-

diography (ICG). Also in this case either implanted

or wearable strategy can be pursued. ICG can moni-

tor the ﬂuid content of the trunk, thus acting as heart

failure detector (Bour and Kellett, 2008), or hemody-

namics variables such as: stroke volume (SV), stroke

index (SI), thoracic ﬂuid content (TFC), systemic vas-

cular resistance (SVR), systemic vascular resistance

index (SVRI), acceleration index (ACI), velocity in-

www.healthstats.com/en/bpro.html

dex (VI), pre-ejection period (PEP), left ventricular

ejection time (LVET) and systolic time ratio (STR).

The under study wearable system performs a tho-

racic bioimpedance measurement by means of four

electrodes placed on the thorax at the diaphragm level.

The development is in an early stage and in vivo tests

are scheduled to evaluate innovative approaches.

3 RESULTS AND DISCUSSION

3.1 Characterization of the Pressure

Sensor

Tests have been performed on 3 samples of the

P30PCB by using the system described in 2.2, in or-

der to calibrate and evaluate the pressure sensor. The

ﬁgures 4 and 5 show the results of the tests in term of

hysteresis and reproducibility.

Figure 4: Typical hysteresis plot (dotted line for decreasing

pressure, full line for increasing pressure).

The average hysteresis has been evaluated by cal-

culating the maximum difference, in term of percent-

age of the full scale of the test, between the curves rel-

ative to growing and decreasing pressure. The mean

among all tests of so calculated hysteresis is 0.33%.

Figure 5: Experimental values (triangles, circles and

squares) and ﬁtting lines for the three tested pressure sen-

sors.

The table 1 collects the slope, the zero value and

the coefﬁcient of determination (R

) for each calibra-

tion curve, calculated by linear polynomial ﬁtting us-

ing Matlab software.

The coefﬁcient of determination is very close to

one, so the sensors can be considered linear. The

differences in the zero values among the calibration

A SENSORIZED PLATFORM TO MONITOR AND REGULATE VAD FUNCTIONALITIES - Wearable and

Implantable Devices to Improve VAD Performances and Extend its Field of Application

317

Table 1: Coefﬁcients for the calibration curve of the three

sensors.

Slope Zero value R

S1 0,0004 0,0706 0,9987

S2 0,0004 0,0725 0,9989

S3 0,0004 0,0711 0,9874

curves of the samples make necessary the calibration

of each implantable device. This is not a problem for

the ﬁeld of application of the SensoART platform.

As example of the stability test, ﬁgure 6 represents

a typical sensor output at 24 kPa for 30 minutes. All

the sensors showed a high stability.

Figure 6: Typical sensor output at 24 kPa.

3.2 Tonometry

Some preliminary acquisitions have been performed

with the wireless tonometry device described in 2.3,

placed as in ﬁgure 3. The gathered signal depicts the

pressure wave of the radial artery of the wrist, as de-

picted in ﬁgure 7.

3.3 Conclusions

SensorART project aims to develop a disrupting sys-

tem that will change substantially the application of

VAD in heart failure treatment. The activity presented

in this paper is a preliminary step towards this chal-

lenging result. The evaluation of the P30PCB sensor

as candidate for the pressure implantable catheter and

the tonometry system gave positive results. The lin-

ear calibration curve and the reproducibility of this

sensor, together with its small dimensions allow to in-

tegrate it in a catheter to start the evaluation in labora-

tory mock-up and in-vivo test. Next steps will be the

design and production of a custom catheter and the

study for the package for a long term implant. The

preliminary tonometric signal, acquired by the proto-

type system, demonstrated the feasibility of this ap-

proach. Next development of this system will lead to

the improvement of the reliability during movement

and to make easier the wearing process.

Figure 7: Pressure signal of the radial artery gathered by the

tonometry device.

ACKNOWLEDGEMENTS

The research leading to these results has received

funding from the European Community’s Seventh

Framework Programme (FP7/2007-2013) under grant

agreement num. 248763 (SensorART Project).

The pressure sensors under test have been kindly

provided ahead of commercialization by STMicro-

electronics.

REFERENCES

Bour, J. and Kellett, J. (2008). Impedance cardiography:

a rapid and cost-effective screening tool for cardiac

disease. In Eur J Intern Med., volume Epub 2008 Feb

11.

Bullister, E., Reich, S., and Entremont, P. (2001). A blood

pressure sensor for long-term implantation. artiﬁcial

organs. In Artiﬁcial Organs, volume 25(5), pages

376–379.

Chemla, D., Teboul, J., and Richard, C. (2008). Non inva-

sive assessment of arterial pressure. In Curr Opin Crit

Care, volume 14(3), pages 317–321.

Maeda, K., Chinzei, T., Imachi, K., and et al. (1988). Pre-

dictive control by physical activity rate of a total ar-

tiﬁcial heart during exercise. In ASAIO Transaction,

volume 34, pages 480–484.

Nakamura, M., Homma, A., Tatsumi, E., and et al. (2000).

Mixed venous oxygen saturation as a promising pa-

rameter for the physiologic control of the artiﬁcial

heart. In ASAIO J, pages 761–766.

Valdastri, P., Rossi, S., Menciassi, A., Lionetti, V., Bernini,

F., Recchia, F., and Dario, P. (2008). An implantable

zigbee ready telemetric platform for in vivo monitor-

ing of physiological parameters. Sensors and Actua-

tors, Volume 142:369–378.

Waters, T., Allaire, P., Tao, G., and et al. (1999). Motor

feedback physiological control for a continuous ﬂow

ventricular assist device. In Artiﬁcial Organs, volume

23(6), pages 480–487.

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

318