1
Portable Patient Training Device for Lung Cancer Treatment
University of Wisconsin-Madison
College of Engineering
Biomedical Engineering 301
Midterm Report
March 12th, 2004
Team Members
Thomas Chia
Brent Geiger
Jason Ethington
Kawai Chan
Client
Dr. Bhudatt Paliwal, PhD
Department of Human Oncology
Advisor
Willis Tompkins, PhD
Department of Biomedical Engineering
Problem Statement
4-D tomotherapy treatment outcomes for lung cancer depend on the stability and repeatability of a patient’s breathing pattern. A portable patient training system is desired to allow more breathing pattern practice for patients. The portable device should measure a patient’s breathing pattern, process the breathing pattern waveform signal and display the signal against an established reference waveform on a portable pocket PC. The display should provide feedback to the patient to allow him/her practice a stable and repeatable breathing pattern.
Background
Tomotherapy
Tomotherapy is a relatively new technique used to deliver radiation for the treatment of cancer. This technique uses high-energy radiation in order to damage and eventually kill target tissues, such as a tumors or cancerous tissue. One of the underlying principles of tomotherapy is to damage the target tissue more than the surrounding healthy tissue. By concentrating the radiation on the target tissue and avoiding the healthy surrounding tissue the cancer can be treated with minimal damage to the body. Tomotherapy has several unique features that allow it to be effective.
Before treatment a CT image of the target tissue is made. This allows the physicians to know exactly what the tissues and tumors look like, and where they are before beginning treatment. This is a very important step in avoiding the healthy tissue damage and limiting misplaced radiation.
Tomotherapy also uses a helical radiation delivery pattern. This means that the radiation source rotates around the patient as it delivers the radiation. This allows the radiation to delivered from many different angles and to effectively concentrate on the target tissue without damaging the surrounding healthy tissue (Figure 1).
Figure 1: Helical delivery pattern used in tomotherapy.
Figure 1 represents a transverse plane of a patient receiving tomotherapy radiation. The red center represents the target tissue and the blue lines represent the radiation beams. Notice how the radiation concentrates itself at the target tissue because of the many angles of attack. One can see how more angles of attack will concentrate more radiation at the target and relatively less to the surrounding tissue.
Tomotherapy utilizes technology called Intensity Modulation Radiotherapy, or IMRT. IMRT is the changing of the intensity and shape of the radiation that is delivered to the target tissue. It does that by using a multileaf collimator, or MLC. A MLC uses leafs, or plates, that quickly move back and forth obstructing the radiation. This changes the intensity and shape of the radiation. This allows further control for the delivery of the radiation, making it more effective in killing the target tissue.
Design Motivation
In the case tomotherapy treatments for lung cancer, the tumor will move as the patient is breathing during the treatment. This adds another level of difficulty. Not only does the tomotherapy device have to track the tissue in three dimensional space it also has to track the tumor as the patient breaths, adding time as a fourth dimension. The tumor can be tracked as the patient breathes but it is important that the patient breathes in a steady repeatable pattern. By taking several CT images throughout the patient’s breathing cycle the exact location of the tumor can be known with respect to where the patient is in his or her breathing pattern. The physicians currently use several techniques to track the patient’s breathing patterns during the procedure. These consist of a sensor or reflector that is placed on the chest of the patient and either infrared light or laser beams. This enables the physician to know the position of the patient’s chest and the breathing pattern of the patient. By giving the patient a guiding cycle the patient can follow it and the position of the target tissue will be known.
However, an ideal breathing pattern is hard to achieve in many patients. Even normal breathing has deeper or shallower breaths and often change in frequency. The situation a person is in will also change the pattern. This is obviously seen in breathing patterns when exercising compared to at breathing patterns at rest. This can also come into effect when a person is nervous, for example before a radiotherapy session. Our client believes that through at home practice, the patient will be able to breathe a repeatable pattern when it comes time for the tomotherapy procedure.
Client Requirements
Our client would like a portable breathing trainer that would allow a patient practice breathing in the comfort of his/her home. Currently, this is not feasible with the equipment used in the hospital for a number of reasons. First of all, the equipment used is extremely expensive. It is not possible to loan this equipment to the patient. Also, the current devices are large and bulky and would require special transport. The devices are also not very user friendly and it would take some training for the patient to use it at home.
In order to construct a portable breathing trainer, the client has specified a number of requirements for the device. The first requirement is that the device should be portable and be able to be used at the patient’s home. The display on the device should also show the patient’s breathing pattern in real-time on top of guiding cycles. Since these are patients suffering from lung cancer, the device should be easy to use and comfortable to the patient. Finally, the client informed us of what he envisioned as the three major components to the device:
1. Breathing measurement device
2. Signal Processor
3. Personal Digital Assistant (PDA) + Software
Design Constraints
There are several aspects that the design must encompass in order for it to be an effective training device fora repeatable breathing pattern. These aspects all revolve around one main issue; the patient needs to practice their breathing pattern outside of the hospital on their own.
First of all proposed designs must give a guiding cycle that the patient can follow. This guiding cycle would be an example of an ideal breathing pattern. It should be specialized for each patient, since everyone’s breathing pattern is different. This guide will be constructed from a typical pattern of the patient found through a number of calibration breaths. The device will then display the real-time breathing pattern of the patient. This is important so that the patient can accurately follow the guiding cycle and train for a repeatable pattern.
Since the device is to be used at home it is important that the device is small and portable, and user friendly. It is also important that the device is user friendly. The patient will need to use the device on their own so it must be simple and straightforward. If the device is too complicated the patient may become frustrated and not use the device at all, which in turn decreases the effectiveness of their treatment.
Alternative Solutions
The proposed design will consist of three discrete component stages (Figure 2). The first component of the overall design is the breathing pattern measurement hardware. This hardware must consistently measure the patient’s breathing pattern and output a signal that can be analyzed, processed, and graphically displayed. In the following section three proposed breathing measurement design alternatives will be outlined and evaluated.
The second component of the design is the analog todigital signal conversion and software used to analyze/graph the output signal from the breathing measurement hardware. Two programming languages, Microsoft Visual Basic® and National Instruments Labview 7.0®, will be considered for use in the final design.
Finally, the last component of the design will be the user interface and final display of the patient’s breathing waveform. Display of the breathing pattern waveform can be accomplished either on a laptop computer or a personal digital assistant (PDA).
Figure 2: Three basic component stages of overall proposed design.
Thermistor Design
The first proposed design alternative to measure patient breathing patterns involves the use of a nasal thermistor or thermocouple sensor. This device detects the inherent changes in nasal airflow that result from normal breathing using a thermistor or thermocouple. Then the corresponding change in resistance would be processed using a low power microcontroller with an A/D converter for data acquisition and signal processing. Potentially, a wireless RF link to a personal computer or PDA could be implemented for long-term data acquisition and storage (Javanov et al.)
An advantage of this device is that it can generate a relatively accurate breathing pattern waveform (Graph 1). It is also capable of consistently measuring a patient’s breathing waveform for an extended period of time.
Graph 1: Breathing pattern waveform generated with a nasal thermistor breathing pattern measurement device. BAL corresponds to breath amplitude while BI corresponds to breath interval (Javanov et al.).
The main disadvantage of this type of design is invasiveness. Many patients would feel uncomfortable having a thermistor placed inside their nostril for and extended period of time. Also, another major disadvantage of this type of device is that it only measures breathing from one nostril. This means that if the patient breathes through his/her mouth the pattern will not be detected by the thermistor device. Yet another disadvantage of this device is that it requires rather complex digital microcontroller circuit design and construction.
Figure 3: Thermistor measuring airflow through nostril.
Inductance Pnuemography
The second proposed design alternative involves the use of a technique called impedance or inductance pnuemography. In each case instrumentation is used to measure either a change in inductance or electrical impedance corresponding to a patient’s breathing pattern. Impedance pneumography involves passing a very small electrical current across ECG chest electrodes and measuring the change in impedance as the chest volume changes. The impedance change is caused by air (which is a poor electrical conductor), moving into the lungs and thereby changing the volume of the lungs (WelchAllyn). Inductance pneumography involves strapping a coaxial cable around the patient’s torso (Figure 4). Since the inductance of the cable depends on cable length as the patient breathes the measured change in impendence corresponds to magnitude of respiratory volume.
An advantage of this technique is that it is accurate and minimally invasive. However, this design is prone to motion artifact. If the patient moves his/her chest in a fashion other than regular breathing it will cause breathing measurement errors. Furthermore, it requires proper electrode or wire placement on the patient’s chest based upon their individual breathing patterns. For example, some people breathe mainly through chest expansion, while others breathe mainly through stomach expansion. The electrode leads or wire strap must be positioned according to the patient’s breathing habits in order to obtain the clearest signal. Finally, this design would require the construction of some simple circuitry to amplify, filter, and process the breathing pattern signal.
Figure 4: An example of an inductance pnumography device.
Linear Position Transducer
A third proposed design alternative involves using a technique called resistance pnueomograghpy, which is similar to the technique described above. However, instead of measuring the inductance change of a cable, a miniature cable extension linear position transducer is used to obtain a signal corresponding to the breathing pattern waveform.
The position transducer is attached to an elastic strap tightened around the patient’s chest (Figure 5).
Figure 5: Example of a position transducer attached to a chest strap.
As the patient breathes and the circumference of the chest expands the cable will extend a given distance based upon the magnitude to patient inhalation (Figure 6). The cable extension corresponds to a change in resistance of the potentiometer. Using a simple voltage divider the output is an analog signal ranging from zero to supply voltage with magnitude proportional to the amount of cable extension. The output signal can then be processed using an A/D converter and interfaced with a computer for data storage and breathing waveform display (Webster 2004).
Figure 6: Image and schematic of the miniature linear displacement transducer. The diameter is approximately 20 cm.
The advantages of this design are that it is accurate, minimally invasive, and does not require any additional circuitry for amplification or filtering of the analog output signal. The main disadvantage is price. Also it is necessary to place the elastic strap and potentiometer on the patient’s chest based upon their breathing habits.
Design Matrix
The design matrix contains four categories: accuracy, price, simplicity and invasiveness. Accuracy is an important factor because reliable measurements are vital to the establishing a repeatable breathing pattern. The inductor method is the most accurate among the three measuring device because it is the least affected by the different kinds of breathing types. By placing one sensing strap around the chest and the other across the stomach both breathing types can be measured. The potentiometer method acheivesmoderate accuracy, however, it cannot measure both chest and stomach breathing simultaneously. This problem can be solved by advising the patient to attach the device to the same area of the torso toget a reproducible signal. The thermistor method is the least accurate because it needs to be placed in the same location (i.e. same depth inside the nostril). This can prove difficult and invasive for the patient to accomplish.
The price for the displacement transducer includes buying a linear potentiometer, spring, and nylon strap. The estimated cost for this is approximately $225.00. The other two methods are more expensive because we would likely have to buycomplete package in the case of the inductance pnuemography or deal with complicated electronics to build a thermistor measurement device.
Simplicity is an important attribute for the measurement device. This is both for the sake of limited time to design and build the device as well to keep things simple for the patient. The potentiometer is the right choice because the basic working theory is a voltage divider. By connecting the potentiometer with a power source and attaching it to a strap, the device is already outputting a meaningful signal. The other two required a rather complex theory and a more careful design process.
Invasiveness is the other main concern in this design. The higher the score in this category, the less invasive the device is. The potentiometer and inductor methods are both ranked to the highest because they are placed around patients’ chest. As their components are lightweight, both of these devices will not obstruct normal breathing. On the other hand the thermistor design may require putting a thermistor inside the patient’s nostril which may cause annoyance to them.
Each category is ranked on a one-to-five scale where a five means the best. A weight factor is applied to each category based on the relative importance. Since the simplicity and invasiveness is more important than the accuracy and price, the score in these categories are multiplied by two. The total score shows that the potentiometer method is ranked the best and this is the method to be developed in this semester (Chart 1).
Chart 1: Design matrix for breathing measurement hardware.
Data Processing and Data Displaying Units
After the breathing data is acquired from the measurement devices, it is important to interpret the data into a meaningful output. Therefore, we need a data processing unit that will convert the analog voltage signal into a digital signal that can be used to display the breathing pattern; this can be achieved through the use of an analog-to-digital (A/D) converter. By plugging in the voltage output to an A/D converter and connecting the converter to a PC, the analog signal can then benefit from digital signal processing.
The search for an A/D converter has lead to two possible options, the Measurement Computing PMD-1208LS and the National Instruments DAQ-6024E PCMCIA card. The PMD has a 10-bit analog output, while the 6024E has a 12-bit analog output. Either one should have more then enough resolution then we need for our application. The PMD has a USB output which will make it easy to interface with a PC and costs $109.00. The 6024E will has many more features and will interface nicely with Labview software. Unfortunately, the cost for this card is $600.00. Which card we choose to use in the device has yet to be decided and depends largely on whether we use a PC or PDA.
After the D/A converter, the PC still needs instruction on how to organize the digital data to a meaningful way (i.e. by writing a program to tell the PC how to manage the input data). The two programming language being considered are Visual Basic and Labview.
Visual Basic versus Labview
Visual Basic is one of the most popular programming tools developed by Microsoft. The language is rather simple compared with other high-level languages. The main advantage is that it is relatively easy to learn can be used by various Windows applications (Predko, 2003). A disadvantage to this language is that it is slow (Tompkins, 2004).