iTrem

Robert Delano, Brian Parise and Leanne West

Georgia Tech Research Institute, 925 Dalney St., Atlanta, Ga, U.S.A.

Keywords: iPhone, Essential tremor, Parkinson’s, Mobile phone, Human-computer interaction.

Abstract: iTrem is an iPhone application that uses the phone’s built-in accelerometers to collect frequency and

amplitude data of hand tremor in people with Essential Tremor, Parkinson’s Disease, and other neurological

conditions. iTrem makes it possible to take frequent tremor data samples from a subject and communicate it

in real time to a doctor or researcher independent of the subject’s location. Aggregating the tremor data

collected by iTrem provides a way for patients, doctors and researchers to collaborate on a grand scale,

possibly shedding light on new avenues of treatment and therapy.

1 INTRODUCTION

Essential Tremor and Parkinson’s disease are the

most common neurological conditions defined by

tremor and collectively affect 5 to 10 million people

in the U.S. (Boelen, 2009). Parkinson’s kills one

person every 26 minutes in the U.S. alone (Xu,

Kochanek, Murphy, and Tejada-Vera, 2007)! There

is no definitive diagnosis for either condition; both

are a clinical diagnoses based on symptoms and

response to medication (Anwar and Sweeny, 2010).

Very little data about tremor is available to track the

effectiveness of medication and therapy over time.

Essential Tremor affects nearly 1 out of 5 people

over 65 and is usually progressive (Beers and Jones,

2005). Commonly beginning in young adulthood,

Essential Tremor becomes more obvious with age,

usually resulting in some degree of disability.

Originally called “The Shaking Palsy”,

Parkinson’s is a progressive disorder of the central

nervous system caused by a loss of Dopamine in the

brain. The patient experiences four primary

symptoms: tremor, rigidity, slow movements, and

loss of balance. One of the most common

neurological disorders 5 to 10 percent of people with

Parkinson’s present will symptoms before the age of

forty (Grimes, 2004).

Dopamine replacement is the “gold standard” for

treating Parkinson’s disease. Patients experience

what doctors refer to as “on/off” times with the

medication refer to. On is when there is some

symptom relief and off is when the symptoms return.

To sustain a therapeutic effect it is crucial to

anticipate the off times and take medication in

advance to reduce the occurrence of off times. Some

patients take medication every two hours and others

just a few times daily. Being able to track the

severity of symptoms throughout the day will help

patients and doctors define an optimum daily

medication schedule.

Tremor is the most common movement disorder

and is defined as an involuntary, rhythmic,

oscillating movement of nearly constant amplitude

(Anwar and Sweeny, 2010). The amplitude of a

tremor determines the severity of the disability

where the frequency can be characteristic of a

specific disorder (see Table 1).

There are three categories of tremor, resting,

postural, and kinetic or intention tremor. Resting

tremor occurs when the body is at rest and supported

against gravity. Postural tremor occurs when holding

any intentional posture, such as the arms out

stretched. Kinetic or intention tremor may appear

during any voluntary movement. Many patients

experience a combination of all three types of

tremor. (DeMeyer, 2004).

There are three categories of tremor, resting,

postural, and kinetic or intention tremor. Resting

tremor occurs when the body is at rest and supported

against gravity. Postural tremor occurs when holding

any intentional posture, such as the arms out

stretched. Kinetic or intention tremor may appear

during any voluntary movement. Many patients

experience a combination of all three types of tremor

(DeMeyer, 2004).

115

Delano R., Parise B. and West L..

iTrem.

DOI: 10.5220/0003177701150118

In Proceedings of the International Conference on Health Informatics (HEALTHINF-2011), pages 115-118

ISBN: 978-989-8425-34-8

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

Table 1: Tremor Frequencies (Rohkamm, 2004).

Frequency Disorder

2.5-5 Hz Cerebellar tremor, Holmes tremor

3-6 Hz Parkinsonian tremor

7-9 Hz Essential tremor, postural tremor in parkinsonism

7-12 Hz Physiological tremor, exaggerated physiological

tremor

12-18 Hz Orthostatic tremor

The severity of tremor is most often determined

by a subjective clinical observation made by a

Movement Disorder Specialist (MDS). The doctor or

patient bases medication and therapy choices on an

observed response to medication.

Reproducible objective tremor data, as is

collected by the iTrem application, will provide

doctors with more information when choosing daily

medication schedules and types of therapy. With a

larger quantity of reliable data, researchers will have

another avenue to spot trends in therapy. Actively

involving the patient may also provide some level of

comfort for the patient, increased communication

with the doctor, and more satisfaction with the care

provided (Sheldon and Kaplan, 1985).

2 TREMOR MEASUREMENT

The cost and challenge of gathering frequent tremor

data discourages the collection and aggregation of

data. In many cases, therapy choices are based solely

on the subjective clinical observation of a doctor.

These observations could work in conjunction with

objective data collection, thereby creating a

multidimensional approach towards the treatment of

tremor (Tarsy, Vitek and Lozano, 2003).

2.1 Clinical Rating Scales

Most tremor rating scales are based on a simplistic

rating score of 0 to 4, with 0 being no detectable

tremor. Although still used, many scales have not

been validated or found to be reliable. The

Washington Heights-Inwood Tremor Rating Scale is

the only validated clinical tremor rating scale (Chen

and Swope, 2010).

The Washington Heights-Inwood Tremor Rating

Scale first has the patient hold his/her arms in front

with wrists pronated and then in a “wing” position.

The patient is also asked to pour water from one cup

into another, use a spoon to drink water, perform

finger to nose movements, and draw Archimedes

spirals. Tremor is rated on a scale 0 to 3, with 0

being normal. Although, this technique has been

validated, a scale of 0 to3 does not allow for subtle

differences in frequency and amplitude. The

Washington Heights-Inwood Tremor Rating Scale is

also not designed for the testing of resting tremor

(Tarsy, Vitek and Lozano, 2003).

2.2 Professional Measurement Systems

The expense and time required for doctors to use

electronic-physiologic methods like the EEG and

EMG usually discourages the use of these devices.

By attaching 16-20 electrodes to a patients scalp, an

Electroencephalography (EEG) can record the

electrical activity along the scalp caused by the

firing of neurons in the brain (Niedermeyer and da

Silva, 1999).

An electromyography EMG uses a very thin

needle electrode that is placed through the skin into

the muscle. The electrode then picks up the electrical

activity produced by the muscles. An EMG may

cause mild discomfort where the electrodes are

inserted and may leave the tested muscle tender

afterwards.

Both EEG and EMG are expensive and time

consuming. An average EMG takes around 15-20

minutes to complete and may cause discomfort. An

EEG can last about 1-½ hours. EEG and EMG

procedures also require the patient to be on location.

iTrem is affordable, mobile and painless.

2.3 Off-the-shelf Measurement Systems

Although off-the-shelf products are available for

tremor measurement, there are no devices that

collect and store tremor data for review and

research. They also require that the patient and

doctor be in the same location.

Two examples of off-the-shelf tremor

measurement devices are the Tremorometor

(www.managingtremor.com) and the Stressometer

(www.tnr.fr). Designed to test for tremor in an office

or clinic, the Tremorometer® provides acceleration

measurements by taping a sensor to the patient’s

index finger for 15-second tests. The Stressometer

has the ability to measure tremor caused by

neurological disorders, but instead focuses on the

microscopic tremors of the nervous system as they

relate to stress.

HEALTHINF 2011 - International Conference on Health Informatics

116

Table 2: Feature comparison.

Feature iTrem

Tremorometer Stressometer

Measure tremor Yes

Yes Yes

Real-time

patient doctor

communication Yes No No

Aggregate patient

data Yes No No

Expandable Yes No No

Hardware Cost >=$199 $595 ~$650

3 iTrem

Both a phone and a breakthrough Internet device, the

iPhone garnered over 25 percent of the mobile

market by the end of 2009 (Tabini, 2010). The

iPhone’s sensitive accelerometers, Internet access,

and popularity made it an ideal hardware choice for

creating an application to track tremor and

communicate that data instantaneously to doctors.

iTrem is a newly developed mobile iPhone

application that provides an affordable and easy

method to collect reproducible, objective tremor data

(See Figure 1). Resting, postural, and intention

tremors can be measured with the iTrem system.

iTrem can connect the doctor, patient, and

researchers in real time without requiring them to be

in the same geographic location. Tremor data can be

transmitted to a web repository for doctor review or

aggregated for research. Through the patient’s web

profiles, doctors can remotely schedule tests, test

reminders, and medication reminders.

To collect data with iTrem, the patient holds the

iPhone and performs simple arm or hand movements

to test resting, postural, and action tremors (See

Figure 2) that are based on the movements that a

patient would perform in the Doctor’s office. To

collect data with iTrem, the patient holds the iPhone

and performs simple arm or hand movements to test

resting, postural, and action tremors (See Figure 2)

that are based on the movements that a patient would

perform in the Doctor’s office. The tests last 10 to

15 seconds each. Holding the iPhone during these

tests, the iTrem application measures tremor data,

including amplitude and frequency, using the

iPhone’s accelerometers.

To collect data with iTrem, the patient holds the

iPhone and performs simple arm or hand movements

to test resting, postural, and action tremors (See

Figure 2) that are based on the movements that a

patient would perform in the Doctor’s office. The

tests last 10 to 15 seconds each. Holding the iPhone

during these tests, the iTrem application measures

tremor data, including amplitude and frequency,

using the iPhone’s accelerometers.

Figure 1: Actual tremor index meter.

Doctors and researchers can program the phone

via the web server. Test schedules and

configurations are stored on the server and

synchronized to the phone periodically. The iPhone

will remind the patient when to take each test with

an alert or reminder. The phone will then briefly

instruct the patient as to what actions to perform for

each test including, resting, postural, and action.

The results of every test are stored in a Web

Repository.

After collection, patient data is uploaded to a

private profile on a secure web server. Doctors can

then review the objective data to assess medication

response, progression, and tremor severity. This

information will also help the patient and doctor

anticipate “off” times and adjust the daily

medication schedule accordingly.

Figure 2: Resting, postural and action tremor tests.

Access to the data is restricted to the patient and

their doctor. Researchers could have access to an

anonymized version of the entire patient databases

(See Figure 3). The tremor data will assist doctors in

diagnosis, as well as helping patients refine their

daily medication schedules to minimize “off” times.

iTrem

117

Figure 3: Flow of information.

3.1 Data Collection

Accelerometers inherently produce a small amount

of noise, and the iPhone accelerometer is no

different. iTrem compensates for with custom filters

and data smoothing algorithms. To date, the

collected data has been consistent. Validation has

been performed by comparing results from iTrem

with results gathered by the Mayo clinic using their

measurement instrumentation. More rigorous

validation will occur in the next phase of

development with the use of a motion capture

system. Using the high fidelity tracking capabilities

of the modern IR-based motion capture systems

iTrem will be able to measure in 3 dimensions hand

tremor displacement and compare that with the data

collected by the iTrem application.

4 FUTURE DEVELOPMENT

While continuing to develop the tremor tests the

iTrem team is now collaborating with a Movement

Disorder Specialist (MDS) and clinical researcher

whose specialty is the study and treatment of

Parkinson’s disease. The MDS has agreed to

determine the effectiveness of iTrem as a tool in the

exam room as well as the validity of the

measurements.

Simple games are also being developed to record

tremor data. The game is designed so that the motion

of the game play can be filtered out to reveal the

tremor data. Games also offer the capability to

gather baseline data from individuals without tremor

who are willing to play the game and contribute their

movements to the tremor database as well as allow

for the collection of data from patients who are

focusing on using tremor affect hands and arms to

perform tasks such as keep a character in the game

still.

ACKNOWLEDGEMENTS

Funding for this effort to date has been provided by

of the Georgia Tech Research Institute Electro-

Optical Systems Laboratory’s Landmarc Research

Center through Internal Research and Development

funding.

REFERENCES

Ahmed, Anwar, Sweeny, Patrick Tremor. Retrieved

March 2, 2010 from Cleveland Clinic:

http://www.clevelandclinicmeded.com/medicalpubs/di

sease management/neurology/tremors/.

Beers, Mark H. and Jones, Thomas V. The Merck Manual

of Health & Aging. Ballantine Books, New York,

2005. 357, 360.

Boelen, Miriam Health Professionals’ Guide to Physical

Management of Parkinson’s Disease. Human Kinetics,

Champaign, IL, 2009. 69.

Chen, Jack J. Swope, David M. Retrieved March 3, 2010

from Medscape: Essential Tremor: Disability

Associated with Essential Tremor: Clinical,

Functional, and Psychosocial. http://www.medscape.

com/viewarticle/461397_9.

Greenfield, Sheldon, M. D., Kaplan, Sherrie, Ph.D., M. P.

H; and Ware, John E., Jr., PhD Expanding Patient

Involvement in Care: Effects on Patient Outcomes.

Annals of Internal Medicine , April 1, 1985.

Grimes, David A. Parkinson’s: everything you need to

know. Firefly Books, Buffalo, New York, 2004.

DeMeyer, William Technique of the Neurologic

Examination. McGraw-Hill Companies, Inc., 2004.

347-348.

Rohkamm, Reinhard Color Atlas of Neurology. Thieme,

Stuttgart, Germany, 2004.

Niedermeyer, Ernst, da Silva, F. H. Lopes

Electroencephalography, 4th Ed., Lippincott Williams

& Wilkins, Philadelphia, PA 1999.

Stressometer: www.tnr.fr

Tabini, Marco comScore: Apple gained US smartphone

market share in December. Retrieved March 4, 2010

from MacWorld: http://www.macworld.co.uk/

Tarsy, Daniel, Vitek, Jerrold Lee, and Lozano, Andres M.

Surgical treatment of Parkinson’s disease and other

movement disorders. Humana Press Inc., Totowa,

New Jersey, 2003.

Tremorometer: www.managingtremor.com/.

Xu, Jiaquan, M. D., Kochanek, Kenneth, M. A., Murphy,

B. S. and Tejada-Vera, B. S. National Vital Reports.

U.S. Department of Health and Human Services,

Hyattsville, MD, 2007.

HEALTHINF 2011 - International Conference on Health Informatics

118