SEISMOCARDIOGRAPHY: A NOVEL APPLICATION

FOR THE NON-INVASIVE ASSESSMENT OF THE FIRST

MAXIMAL DERIVATIVE OF LEFT VENTRICULAR PRESSURE

Melonie Burrows

, Graeme Jahns

, Geoffrey Houlton

, Berry van Gelder

and Frank Marcus

Heart Force Medical Inc., Vancouver, British Columbia, Canada

Catharina Hospital, Eindhoven, Netherlands

The University Medical Centre, Tucson, Arizona, U.S.A.

Keywords: Cardiac Resynchronization Therapy, Seismocardiography, Ballistocardiography, Biventricular pacing, Heart

failure.

Abstract: Cardiac resynchronization therapy (CRT) results in improved clinical status in patients with heart failure

and left ventricular dyssynchrony. One third of CRT patients fail to respond due to the inability to 1)

identify non-responders prior to treatment, 2) optimize coronary sinus lead placement for left ventricular

pacing and 3) optimize the atrio-ventricular (A-V) and inter-ventricular (V-V) intervals. Although invasive

measurements of first maximal derivative of left ventricular pressure (dP/dt

max

) are used to optimize lead

placement and A-V and V-V intervals in CRT, it would be preferable to have a non-invasive assessment of

dP/dt

max

. Echocardiographic dyssynchrony and left ventricular function are current parameters for non-

invasively evaluating responders to CRT, but they are not recommended due to their poor reproducibility.

We applied recent advances in technology to develop a device called the digital ballistocardiograph (dBG

which assesses the mechanical function of the heart using triaxial accelerometry. We show that dBG

cardiac events are valid in comparison to 2D transthoracic echocardiography and reliable in comparison to

cardiac magnetic resonance imaging. We present preliminary data to support our position that the dBG

could be used as a non-invasive assessment of dP/dt

max

in heart failure patients to identify responders and

optimize CRT.

1 INTRODUCTION

The science of ballistocardiography (BCG) was con-

ceived over a century ago (Starr et al., 1955, Starr et

al., 1953) as the study of body motion resulting from

myocardial contraction and blood flow from the

heart to the periphery (Noordergraff, 1961). During

the 1930s to 1960s, there was a surge of studies

showing the importance of BCG measurements in

clinical cardiology (Starr et al., 1950, Starr and

Hildreth, 1952, Starr, 1964), especially in relation to

identifying patients with coronary heart disease

(Baker, 1950, Scarborough, 1952, Baker, 1968) and

development of circulatory abnormalities such as co-

arctation of the aorta (Brown et al., 1949, Starr,

1964). Although conceptually attractive, BCG was

limited in practice, as the devices were cumbersome

requiring fixed installation. In addition, the ability

to analyze the signals electronically was not yet

available. As such, BCG was impractical to use on a

large scale and was abandoned in the 1970s.

Bayevski and colleagues developed seismo-

cardiography (SCG) in 1964. The technique con-

sisted of an accelerometer attached over the sternum

area of the chest, which recorded compression

waves transmitted through the chest wall from heart

contractions during each cardiac cycle. Over the

years, SCG has been refined as a technique for

cardiac stress monitoring (Jerosch-Herold et al.,

1999), left ventricular monitoring during ischemia

(Salerno and Zanetti, 1991, Korzeniowska-Kubacka

et al., 2005), estimation of left ventricular function

(Korzeniowska-Kubacka and Piotrowicz, 2002), and

detection of coronary artery disease (Salerno et al.,

1991). However, SCG devices were not imple-

mented on a large scale due to the emergence and

sudden interest in echocardiography.

251

Burrows M., Jahns G., Houlton G., van Gelder B. and Marcus F..

SEISMOCARDIOGRAPHY: A NOVEL APPLICATION FOR THE NON-INVASIVE ASSESSMENT OF THE FIRST MAXIMAL DERIVATIVE OF LEFT

VENTRICULAR PRESSURE.

DOI: 10.5220/0003845202510256

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2012), pages 251-256

ISBN: 978-989-8425-91-1

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

At present, modern accelerometer-based tech-

nology is revitalizing the science of BCG and SCG,

allowing the motion of the heart to be recorded and

analyzed quickly and efficiently for the assessment

of cardiac function (Alametsa et al., 2009). We have

applied recent advances in hardware and software

technologies to develop a new medical device called

the digital ballistocardiograph (dBG

), which allows

rapid, non-invasive assessment of cardiac events and

the force of the heart’s contraction, lending itself to

patient monitoring and assessment.

Clinical trials have demonstrated that cardiac

resynchronization therapy (CRT) results in improved

clinical status and lower mortality in selected

patients (Abraham et al., 2002, Bristow et al., 2004).

However, approximately one third of CRT patients

fail to respond due to the inability to accurately

1) identify non-responders prior to treatment,

2) optimize coronary sinus lead placement during

the procedure, and 3) optimize the atrio-ventricular

(A-V) and inter-ventricular (V-V) intervals

(Abraham et al., 2002, van Gelder et al., 2004,

Cleland et al., 2005, Jansen et al., 2006). Although

invasive measurements of the first maximal

derivative of left ventricular pressure (dP/dt

max

) can

be used to increase the number of CRT responders

via optimization of lead placement, A-V and V-V

intervals, (Kurzidim et al., 2005, van Gelder et al.,

2008, van Gelder et al., 2009), it would be preferable

to have a non-invasive assessment of dP/dt

max

(Houthuizen et al., 2011). Echocardiographic dys-

synchrony and left ventricular function are current

parameters for non-invasively evaluating responders

to CRT (Altman et al., 2011, Bai et al., 2011).

However, even though echocardiographic variables

have been proposed as surrogates for left ventricular

dP/dt

max

, they are not highly recommended due to

their poor reproducibility (Thomas et al., 2009).

Cardiac timings recorded non-invasively by BCG

and SCG have been shown to provide valuable

insight into the heart’s function (Starr, 1964, Crow,

1994, Lyseggen et al., 2005). As such, if the dBG

could non-invasively predict dP/dt

max

across a range

of heart rates, it would have potential to be a

valuable tool for the non-invasive assessment of

dP/dt

max

for identification of CRT responders and

optimization of CRT.

In this position paper, we present preliminary

data from animal studies to support our position that

the dBG

could be used as a non-invasive

assessment of dP/dt

max

in heart failure patients to

identify responders and optimize CRT.

2 METHODS

2.1 The Digital Ballistocardiograph

The dBG

(Heart Force Medical Inc., Vancouver,

Canada) consists of three main components: the

sensor containing the triaxial accelerometer and two

exposed pads for electrocardiograph (ECG) elec-

trodes, the digitizing transceiver unit which

conditions and samples the signals and transmits the

data to a PC via Bluetooth™, and the software

application used for device control and manual data

analysis (Figure 1A). The sensor is placed on the

sternum in the midline, with its lower edge ap-

proximately 3 cm above the xiphoid process (Figure

1B). The triaxial accelerometer (Figure 1C) detects

the SCG vibrations generated by the heart’s motion.

A single, non-diagnostic ECG similar to a lead 1 is

also recorded. From specific peaks on the SCG, the

following cardiac events can be determined: mitral

valve closure (MVC), aortic valve opening (AVO),

aortic valve closure (AVC) and mitral valve opening

(MVO). Isovolumetric contraction time (IVCT; t

MVC

– t

AVO

) and isovolumetric relaxation time (IVRT;

AVC

– t

MVO

) are derived variables (Figure 2).

Figure 1: A) The digital ballistocardiograph

sensor and transceiver; B) Digital ballistocardiograph

sensor placement, C)

Digital ballistocardiograph

sensor axes from the perspective of the observer; x – from right to left, y – from head to toe, z

– from back to chest. Abbreviations: dBG - digital ballistocardiograph; ECG – electrocardiograph.

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

252

We investigated the accuracy of cardiac events

using the dBG

against the clinical reference

standard of 2D transthoracic Doppler echocardio-

graphy, and found that the cardiac events measured

by dBG

were equivalent to cardiac events measured

by 2D echocardiography (95% of dBG

cardiac

events fell within ±2SD of echocardiography cardiac

events). We have also determined within-device,

intra- and inter-operator reliability for the dBG

measurements. Precision was calculated as the root

mean square error coefficient of variation

(RMSECV, %) and was less than 4% (within), 8%

(intra) and 10% (inter) for all cardiac events, which

is comparable to the gold standard technique of

cardiac magnetic resonance imaging whose

reliability is known to be between 2.9–9% (Chuang

et al., 2000). As such, dBG

cardiac events are both

valid and reliable for clinical use.

2.2 The Digital Ballistocardiograph

in Cardiac Resynchronization

Therapy

We used a swine model to investigate if the dBG

variables were predictive of dP/dt

max

across various

heart rate conditions as well as following induced

changes in blood volume. We studied 10 hybrid

farm pigs (10–16 weeks old), utilizing the dBG

device to measure cardiac timings (MVC, AVO,

AVC, MVO), and catheterization to measure left

ventricular pressure for computation of dP/dt

max

A guiding catheter (Medtronic, MN, USA) and

sensor-tipped PressureWire

(St. Jude Medical Inc.,

MN, USA) were inserted into the left femoral artery

and placed in the apex of the left ventricle under

fluoroscopic guidance (HICOR/ACOM-TOP, Sie-

mens, Erlangen, Germany). The PressureWire

was

connected to the RadiAnalyser

Xpress (St. Jude

Medical Inc., MN, USA) for the measurement of left

ventricular pressure. The RadiAnalyser

Xpress

Figure 2: Example digital ballistocardiograph

waveform annotated for four valve timings. Abbreviations: Mitral valve

closure – MVC; Aortic valve opening – AVO; Aortic valve closure – AVC; Mitral valve opening – MVO; ECG –

Electrocardiogram; X – dBG

sensor axis from right to left, Y – dBG

sensor axis from head to toe, Z – dBG

sensor

axis from back to chest.

SEISMOCARDIOGRAPHY: A NOVEL APPLICATION FOR THE NON-INVASIVE ASSESSMENT OF THE FIRST

MAXIMAL DERIVATIVE OF LEFT VENTRICULAR PRESSURE

253

system utilized the PhysioMon™ software (Version

2.02, St. Jude Medical Inc.) for computation of

dP/dt

max

. The output from the RadiAnalyser

Xpress

was routed to a signal processing unit (CMS,

Module M1006A, Phillips Medical Systems, MA,

USA) for simultaneous monitoring of left ventricular

pressure. A catheter was inserted into the right

femoral artery and placed into the aortic arch under

fluoroscopic guidance (HICOR/ACOM-TOP,

Siemens). The catheter output was routed to the

signal-processing unit (CMS, Module M1006A,

Phillips Medical Systems) for simultaneous

monitoring of aortic blood pressure. All data output

from the CMS were routed to a Biopac MP150

(Biopac Systems Inc., CA, USA). A dBG

pro-

prietary sensor (Heart Force Medical Inc.,

Vancouver, Canada) was placed on the midline of

the sternum with the lower edge of the sensor placed

approximately 3 cm above the xiphoid process. A

pacing wire (Medtronic) was inserted into the left

femoral vein, advanced into the right atrium, and

connected to an external, single chamber pacemaker

(Medtronic) for pacing of the heart at specific heart

rate (HR) conditions. We paced animals via the right

atrium for 10 counterbalanced heart rate conditions:

90, 100, 110, 120, 130, 140, 150, 160, 170 and 180

bpm. At each HR, we observed a period of 5

minutes for normalization of hemodynamics, after

which we collected all pressure and dBG

data

simultaneously for 1 minute. After the pacing

protocol was completed, each animal had blood

withdrawn equivalent to 10% of body weight, which

was subsequently reperfused. After each blood

withdrawal/reperfusion, all pressure and dBG

data

were collected simultaneously at 1 minute

immediately after withdrawal/reperfusion, and 3

minutes after withdrawal/reperfusion.

The left ventricular pressure, dP/dt

max

and dBG

data collected during the counterbalanced HR

conditions were used to assess the relation between

dP/dt

max

and dBG

variables and devise a regression

equation to predict dP/dt

max

non-invasively. The left

ventricular pressure, dP/dt

max

and dBG

data

collected during the blood volume conditions were

used only to assess if changing blood volumes

affected the relation between dP/dt

max

and dBG

variables. The aortic blood pressure signals were

captured only to facilitate time alignment of the

dBG

waveforms to the dP/dt waveforms.

We created a proprietary software tool (Heart

Force Medical Inc.) to manually annotate the

merged data files. We also used the tool to offset the

data streams to achieve heart beat synchronization

for all data. We manually annotated MVC, AVO,

AVC, MVO and a dBG

amplitude measure (dBGv)

calculated from the measured cardiac timings on the

dBG

waveform, and dP/dt

max

on the dP/dt

waveform. We used a repeated measure regression

model to assess if dBGv could predict catheter-based

measurement of dP/dt

max

. Heart rate, HR order and

pig weight were used as covariates, and the within

pig (between HR and within HR) and between pig

variation were assessed. Alpha was set at P <0.05.

All statistical analyses were performed by an

independent statistician using R (www.r-

project.org).

3 RESULTS AND DISCUSSION

dBGv was a significant predictor of dP/dt

max

<0.0001), with a one unit change in dBGv equiv-

alent to 4.01 unit changes in dP/dt

max

(95% CI 3.00–

5.05 units). The relation between dBGv and dP/dt

max

was consistent across HRs (P >0.8), but varied

across pigs (P <0.0001). dBGv remained a sig-

nificant predictor of dP/dt

max

(P <0.0001) even when

blood volumes were decreasing and increasing.

These preliminary data show a strong predictive

capability of dBG

v for dP/dt

max

across heart rates,

suggesting that dBGv is a non-invasive surrogate of

dP/dt

max

and is able to detect the force-frequency

relation between HR and dP/dt

max

. The potential of a

cardiac amplitude variable to predict dP/dt

max

has

been suggested previously, with peak endocardial

acceleration (PEA) amplitude (measured using a

micro-accelerometer located on the tip of a trans-

venous pacing lead), shown to be related to global

ventricular contractility independent of recording

site and atrial rhythm (Bongiorni et al., 1996). Acute

variations in PEA, measured using endocardial leads

and implantable device, have also been shown to

closely parallel changes in dP/dt

max

(Rickards et al.,

1996).

Despite the remarkable clinical results associated

with CRT (Houthuizen et al., 2011), the percentage

of non-responders is still 25–35% (Bristow et al.,

2004, Cleland et al., 2005). Acute measurements of

left ventricular dP/dt

max

have shown to be necessary

to identify non-responders prior to treatment, to

optimize coronary sinus lead placement, and to

optimize A-V and V-V intervals during CRT to

obtain maximal hemodynamic benefit in patients

(van Gelder et al., 2004, Jansen et al., 2006, van

Gelder et al., 2008, van Gelder et al., 2009).

Although catheterization is the gold standard for the

assessment of ventricular function, it is invasive,

costly and time consuming, and is therefore limited

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

254

in its clinical utility. The dBGv measured in this

study is a non-invasive surrogate for dP/dt

max

as it

describes the rate of change of contractile force,

while being measured from an accelerometer placed

on the sternum. During CRT, heart rate is a known

variable that, in conjunction with a measured dBG

could be inserted into an equation to derive a pre-

dicted dP/dt

max

without left ventricular catheter-

ization. An Electrophysiologist could use the pre-

dictive equation during CRT to 1) identify patients

who experience an increase in dP/dt

max

at time of

coronary sinus lead insertion and therefore likely to

be CRT responders, and 2) optimize coronary sinus

lead placement and A-V and V-V intervals. The fact

that the relation between dP/dt

max

and dBGv remains

strong across changes in blood volume suggests that

the predictive relation between dBG

v and dP/dt

max

robust. The use of the dBG

during CRT could help

prevent implantation of CRT devices in the one third

of patients who do not currently benefit from this

therapy (van Gelder et al., 2004).

4 CONCLUSIONS

In this position paper, we presented a brief history of

BCG and SCG and noted that modern technology

has revitalized the science of BCG and SCG,

allowing the motion of the heart to be recorded,

captured and used in the assessment of cardiac

function. A new medical device called the dBG

was

introduced. The use of the dBG

in CRT

optimization was discussed and preliminary data

presented showing the strong predictive capability of

the dBG

to track dP/dt

max

across a wide variety of

heart rates in swine. The dBG

is small and

relatively inexpensive. It appears to have potential to

assist in identifying CRT responders and optimizing

lead position and A-V and V-V intervals. Clinical

studies in heart failure patients are required to

document the ability to assist in selecting patients

who will benefit from CRT, as well as its use in

optimizing A-V and V-V intervals.

ACKNOWLEDGEMENTS

The authors would like to thank Dr Rick White

(University of British Columbia, Canada) for

conducting the independent statistical analysis

herein; Brandon Ngai (Heart Force Medical Inc.) for

his help in developing the position paper; and

Heather Braybrook (Heart Force Medical Inc.) for

her help in editing and formatting this manuscript.

The authors would also like to acknowledge the

work of Heather Braybrook, Hayley Corbett, Dorel

Dumencu and Gonzalo Portacio (Heart Force

Medical Inc.) in data collection for the dBG

studies

mentioned herein.

REFERENCES

Abraham, W. T., Fisher, W. G., Smith, A. L., Delurgio, D.

B., Leon, A. R., Loh, E., Kocovic, D. Z., Packer, M.,

Clavell, A. L., Hayes, D. L., Ellestad, M., Trupp, R. J.,

Underwood, J., Pickering, F., Truex, C., McAtee, P. &

Messenger, J. 2002. Cardiac Resynchronization in

Chronic Heart Failure. New England Journal of

Medicine, 346, 1845-1853.

Alametsa, J., Varri, A., Viik, J., Hyttinen, J. & Palomaki,

A. 2009. Ballistocardiogaphic studies with accel-

eration and electromechanical film sensors. Med Eng

Phys, 31, 1154-65.

Altman, R. K., McCarty, D., Chen-Tournoux, A. A.,

Tournoux, F. B., Riedl, L., Orencole, M., Park, M. Y.,

Picard, M. H. & Singh, J. P. 2011. Usefulness of low-

dose dobutamine echocardiography to predict response

and outcome in patients undergoing cardiac resynch-

ronization therapy. Am J Cardiol, 108, 252-7.

Bai, R., Biase, L. D., Mohanty, P., Hesselson, A.B., Ruvo,

E. D., Gallagher, P. L., Elayi, C. S., Mohanty, S.,

Sanchez, J. E., David Burkhardt, J., Horton, R., Joseph

Gallinghouse, G., Bailey, S. M., Zagrodzky, J. D.,

Canby, R., Minati, M., Price, L. D., Lynn Hutchins,

C., Muir, M. A., Calo, L., Natale, A. & Tomassoni, G.

F. 2011. Positioning of Left Ventricular Pacing Lead

Guided by Intracardiac Echocardiography With Vector

Velocity Imaging During Cardiac Resynchronization

Therapy Procedure. J Cardiovasc Electrophysiol.

Baker, B., Scarborough WMR, Mason RE, Singewald ML

1950. Coronary heart disease studied by ballisto-

cardiography. Transaction of the American Clinical

and Climatological Association, 62, 191-201.

Baker, B. M. 1968. Ballistocardiography: predictor of

coronary heart disease. Circulation, 37, 1-3.

Bongiorni, M. G., Soldati, E., Arena, G., Quirino, G.,

Vernazza, F., Bernasconi, A. & Garberoglio, B. 1996.

Is Local Myocardial Contractility Related to Endo-

cardial Acceleration Signals Detected by a Trans-

venous Pacing Lead? Pacing and Clinical Electro-

physiology, 19, 1682-1688.

Bristow, M. R., Saxon, L. A., Boehmer, J., Krueger, S.,

Kass, D. A., De Marco, T., Carson, P., DiCarlo, L.,

DeMets, D., White, B. G., DeVries, D. W. & Feldman,

A. M. 2004. Cardiac-resynchronization therapy with

or without an implantable defibrillator in advanced

chronic heart failure. N Engl J Med, 350, 2140-50.

Brown, H. R., Jr., Hoffman, M. J. & De, L. V., Jr. 1949.

Ballistocardiograms in coarctation of the aorta; obser-

vations before and after operation. N Engl J Med, 240,

715-8.

SEISMOCARDIOGRAPHY: A NOVEL APPLICATION FOR THE NON-INVASIVE ASSESSMENT OF THE FIRST

MAXIMAL DERIVATIVE OF LEFT VENTRICULAR PRESSURE

255

Chuang, M. L., Hibberd, M. G., Salton, C. J., Beaudin, R.

A., Riley, M. F., Parker, R. A., Douglas, P. S. &

Manning, W. J. 2000. Importance of imaging method

over imaging modality in noninvasive determination

of left ventricular volumes and ejection fraction:

Assessment by two- and three-dimensional echo-

cardiography and magnetic resonance imaging. J Am

Coll Cardiol, 35, 477-484.

Cleland, J. G. F., Daubert, J.-C., Erdmann, E., Freemantle,

N., Gras, D., Kappenberger, L. & Tavazzi, L. 2005.

The Effect of Cardiac Resynchronization on Morbidity

and Mortality in Heart Failure. New England Journal

of Medicine, 352, 1539-1549.

Crow, R. S., Hannan P, Jacobs D, Hedquist L, Salerno

DM 1994. Relationship between seismocardiogram

and echocardiogram for events in the cardiac cycle.

American Journal of Noninvasive Cardiology, 39-46.

Houthuizen, P., Bracke, F. & van Gelder, B. 2011.

Atrioventricular and interventricular delay optimi-

zation in cardiac resynchronization therapy: physio-

logical principles and overview of available methods.

Heart Failure Reviews, 16, 263-276.

Jansen, A.H., Bracke, F.A., van Dantzig, J.M., Meijer, A.,

van der Voort, P.H., Aarnoudse, W., van Gelder, B.M.

and Peels, K.H. 2006. Correlation of echo-Doppler

optimization of atrioventricular delay in cardiac re-

synchronization therapy with invasive hemodynamics

in patients with heart failure secondary to ischemic or

idiopathic dilated cardiomyopathy. Am J Cardiol, 97,

552-7.

Jerosch-Herold, M., Zanetti, J., Merkle, H., Poliac, L.,

Huang, H., Mansoor, A., Zhao, F. & Wilke, N. 1999.

The seismocardiogram as magnetic-field-compatible

alternative to the electrocardiogram for cardiac stress

monitoring. Int J Card Imaging, 15, 523-31.

Korzeniowska-Kubacka, I., Bilinska, M. & Piotrowicz, R.

2005. Usefulness of seismocardiography for the diag-

nosis of ischemia in patients with coronary artery

disease. Ann Noninvasive Electrocardiol, 10, 281-7.

Korzeniowska-Kubacka, I. & Piotrowicz, R. 2002. Seis-

mocardiography--a non-invasive technique for estim-

ating left ventricular function: preliminary results.

Acta Cardiol, 57, 51-2.

Kurzidim, K., Reinke, H., Sperzel, J., Schneider, H. J.,

Danilovic, D., Siemon, G., Neumann, T., Hamm, C.

W. and Pitschner, H.-F. 2005. Invasive Optimization

of Cardiac Resynchronization Therapy: Role of

Sequential Biventricular and Left Ventricular Pacing.

Pacing and Clinical Electrophysiology, 28, 754-761.

Lyseggen, E., Rabben, S. I., Skulstad, H., Urheim, S.,

Risoe, C. & Smiseth, O. A. 2005. Myocardial

acceleration during isovolumic contraction: relation-

ship to contractility. Circulation, 111, 1362-9.

Noordergraff, A. 1961. Further studies on a theory of the

ballistocardiogram. Circulation, 23, 413-425.

Rickards, A. F., Bombardini, T., Corbucci, G., Plicchi, G.

& on behalf of, T. M. P. S. G. 1996. An Implantable

Intracardiac Accelerometer for Monitoring Myocardial

Contractility. Pacing and Clinical Electrophysiology,

19, 2066-2071.

Salerno, D. M. & Zanetti, J. 1991. Seismocardiography for

monitoring changes in left ventricular function during

ischemia. Chest, 100, 991-3.

Salerno, D. M., Zanetti, J. M., Green, L. A., Mooney, M.

R., Madison, J. D. & Van Tassel, R. A. 1991. Seismo-

cardiographic changes associated with obstruction of

coronary blood flow during balloon angioplasty. Am J

Cardiol, 68, 201-7.

Scarborough, W., Mason RE, Davis FW, Singewald ML,

Baker BM, Lore SA 1952. A ballistocardiographic and

electrocardiographic study of 328 patient with

coronary artery disease: Comparison with results from

a similar study of apparently normal persons.

American Heart Journal, 44, 645-655.

Starr, I. 1964. Prognostic Value of Ballistocardiograms.

Studies on Evaluation of the Doctor's Experience.

JAMA, 187, 511-7.

Starr, I. & Hildreth, E. A. 1952. The effect of aging and of

the development of disease on the ballistocardiogram;

a study of eighty subjects, originally healthy, followed

from ten to fourteen years. Circulation, 5, 481-95.

Starr, I., Horwitz, O., Mayock, R. L. & Krumbhaar, E. B.

1950. Standardization of the ballistocardiogram by

simulation of the heart's function at necropsy; with a

clinical method for the estimation of cardiac strength

and normal standards for it. Circulation, 1, 1073-96.

Starr, I., Pedersen, E. & Corbascio, A. N. 1955. The effect

of nitroglycerine on the ballistocardiogram of persons

with and without clinical evidence of coronary heart

disease. Circulation, 12, 588-603.

Starr, I., Schnabel, T. G., Jr. & Mayock, R. L. 1953.

Studies made by simulating systole at necropsy. II.

Experiments on the relation of cardiac and peripheral

factors to the genesis of the pulse wave and the

ballistocardiogram. Circulation, 8, 44-61.

Thomas, D. E., Yousef, Z. R. & Fraser, A. G. 2009. A

critical comparison of echocardiographic measure-

ments used for optimizing cardiac resynchronization

therapy: stroke distance is best. European Journal of

Heart Failure, 11, 779-788.

van Gelder, B., FA, B., Meijer A, LJM, L. & NHJ, P.

2004. Effect of optimizing the VV interval on left

ventricular contractility in cardiac resynchronization

therapy. The American journal of cardiology, 93,

1500-1503.

van Gelder, B. M., Meijer, A. & Bracke, F. A. 2008. The

optimized V-V interval determined by interventricular

conduction times versus invasive measurement by LV

dP/dtMAX. J Cardiovasc Electrophysiol, 19, 939-44.

van Gelder, B. M., Meijer, A. & Bracke, F. A. 2009.

Timing of the left ventricular electrogram and acute

hemodynamic changes during implant of cardiac

resynchronization therapy devices. Pacing Clin

Electrophysiol, 32 Suppl 1, S94-7.

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

256