SYSTEM FOR IDENTIFYING THERMAL VARIATIONS IN BREAST TISSUE
FIELD OF THE INVENTION
The present invention pertains to the field of microwave and infrared radiation detection and in particular to systems and procedures for identifying thermal variations in breast tissue.
BACKGROUND
Devices for examining potentially diseased areas of the body have been used for many years for diagnostic purposes. Magnetic resonance imaging (M 1) machines, ultrasound devices, CAT scanners and X-ray imagers are but a few of the commonly utilized tools that medical radiologists have at their disposal to characterize and understand many of the maladies that afflict patients.
Another technique for imaging tissues is ther ography. Thermography relies on the fact that diseased tissues, such as tumors, tend to have a different temperature than normal tissue due to a different metabolic activity and vascularity of tumors. For example, the tumors tend to appear as hot spots in a thermogram,
A common type of thermography is infrared thermography. Diagnostic techniques using electromagnetic emission in the infrared region of the spectrum have been available for many years and have proved useful in measuring surface temperature distributions in the body. However, body tissue rapidly absorbs electromagnetic energy at the infrared frequencies. Since the heat associated with a subcutaneous tumor is transferred by radiation as well as convection and conduction, the thermal pattern seen at the skin surface due to such a tumor can be altered significantly. In fact, in some cases, a relatively deep tumor may not appear at all in an infrared thermogram of the affected area. Thus, infrared thermography is essentially limited to surface measurements which can vary greatly in response to external factors such as physical activity, menstrual cycle in women, substance intake, etc.
As previously identified it has been noticed that skin temperatures are typically elevated at a location of a malignant tumor projection. This thermal signature at these locations can provide mammary gland cancer diagnostic through the use of TR-thermography. The infrared thermography method allowing skin temperature measurement and visualization had wide spread use in the 1960-70's. However, IR electromagnetic waves are typically unable to penetrate deep into a body and as a result IR thermographs are typically unable to measure temperature in the vicinity of a malignant tumor under the skin. Wherein IR~ thermographs can measure skin temperature that is more or less related to internal tissue temperature. In 1975, A. Barrett suggested the use of microwave radio-thermomctry for providing a mammary gland cancer diagnostic procedure and is defined in A. H. Barrett & Ph. C. Myers, "Subcutaneous Temperature: A Method of Noninvasivc Sensing", Science, Nov.14, 1975, vol.190.
Since biological tissues are relatively transparent to electromagnetic waves within the radio-band, these electromagnetic waves enable the measurement of temperatures in regions located a few centimeters under the surface of the skin. Barrett measured tissue electromagnetic self-radiation at the wavelengths of 23 and 9 cm as described in A. H. Barrett, Ph. C. Myers, N. L. Sadovsky " Microwave Thermography in the Detection of Breast Cancer" AJR: 134, February 1 80.
In order to deleft thermal Electromagnetic radiation he us d a Dike's super-heterodyne receiver which is commonly used in radio astronomy. Noise diodes served as a source for the calibration of the signal. A segment of a rectangular wave-guide filled with a dielectric was used as an antenna for this device.
In order to collect data relating to the internal temperature of mammary glands at several locations, a patient would lie supine and an antenna would be placed at one location on the right mammary gland and remain there for 15 seconds enabling the receiver to perform noise signal integration. This collected signal would be subsequently transformed into a temperature by a microprocessor, The antenna would subsequently be moved to the corresponding symmetric location on the opposite mammary gland and remain there for the next 15 seconds for data collection. This procedure would be repeated until readings had been collected for 9 spots located on each mammary gland.
According to Barrett, upon the collection of this information and the determination of the respective temperature values the microprocessor associated with the system calculated three values: a) thermo-asymmetry which is a temperature difference between symmetric locations on the right and left mammary glands, b) difference of the mean temperatures of the right and left mammary glands and c) difference between the temperature of hottest spot on a mammary gland and the mean temperature thereof. Barrett evaluated a number of mathematical combinations of the measured values in order to identify a preferred procedure for segregating patients with and without mammary gland cancer, The selected identifier which appeared to provide the most efficient determination of mammary gland cancer was the maximal thermo-asymmetry. For example, in a situation wherein the maximal thermo-asymmetry exceeded a temperature threshold, the particular patient received a thermo-positive diagnosis and vice versa,
While this method can be perceived as quite simple and visual, it however may not provide high accuracy of detection. Barret did not use measurement results visualization in terms of temperature fields, he only used a single indicator, namely maximal thermo- asymmetry, in order to determine a thermo-positive or thermo-negative diagnosis. In addition to Dike's radiometer, which Barrett was using for internal temperature measurements, is known in the art to possess significant inaccuracies as a result of variations of impedance of a patient's tissue input, in addition the impedance of the tissue of a patient's mammary glands may vary within a wide range pnd the efore potentially cause significant measurement inaccuracies. It was suggested that using a noise-adding method in order to increase the accuracy of the measurement of a biological object's internal temperature may be beneficial. This noise addition was provided by antenna mismatching.
Henceforth various noise-adding methods were implemented in radiometers. It was proposed that a balanced null radiometer with a sliding compensating circuit for antenna input reflection, with a low-inertia heated resistor used as a noise generator may be beneficial. A disadvantage of using a heated resistor is the fact that the device is unable to measure temperatures that are below the ambient temperature level, Waisblat A.V,
"Meditsinskii Radio-Thermometer". Biomeditsinskϋ Technologii Radioelectronica 2001, No 8.
During the performance of screening examinations, the time required for the completion of an examination can be important. This parameter is directly determined by the time required to measure a temperature at one location. Barrett identified the measurement time as being about 15 seconds. As an example, it should be noted that according to Dike's formula the accuracy of measurements can fluctuate based on an integrator time- constant the radiometer's intrinsic noise, losses in its input circuits and front-end bandwidth. Generally, radio-thermometers have an integrator with a fixed time-constant wherein an RC-circuit can be used as the integrator. Measurement time for radio- thermometers of this form can be determined by transition time, being typically equal to 3τ, where τ is the RC-circuit time-constant. For example Barrett in particular used a RC-circuit with a time-constant of 5 seconds and as such the time required to take one measurement was approximately 15 seconds. In addition, according to Dike's formula, the fluctuation of the accuracy of a reading is determined by the RC-circuit time- constant, therefore by providing a predetermined RC-circuit time-constant, accuracy may not be improved by increasing the measurement time at a particular location, as such an expert or technician may not be ab to improve the accuracy of the result by increasing measurement time, and vice versa, It should be noted that with radiometers having a fixed lime-constant, the measurement and noise signal accumulation processes are continuous ones.
Several other sources of inaccuracy also exist in the prior art. For example, the internal temperature of the components of a radiometer may vary depending on the ambient temperature, thereby resulting in an inaccuracy of mammary gland internal temperature measurement. For example, in order to avoid this problem, the heating of the radio thermometer itself and its components thereby maintaining a constant temperature thereof, irrespective of the ambient temperature can reduce this form of inaccuracy. For example, the circuit boards of the radio-frequency unit can be heated up to a predetermined temperature level and maintained at this level regardless of variations in ambient temperature. It should be noted that increasing internal temperature measurement accuracy using this method can result in a significant increase in the power consumption of the system, and therefore increase the cost of operation.
In addition, a method for detecting cancerous breast tumors using microwave radiometry is described in US Patent No. 5,983,124 (Carr), This method screens for tumors by detecting radiometrically the increased energy emitted in the microwave band by the relatively hot cancerous tumors. The device however, does not account for energy reflection at the interface between the antenna and the skin surface therefore potentially resulting in greater levels of inaccuracy of this device. In addition, Carr defines the step of digitization of information as occurring once a signal has been collected such that this signal is representative of the temperature for the particular location on the first breast being scanned. Upon the digitization of a second signal representative of a temperature at the symmetric point on the other breast and the subsequent comparison of these two digitized temperature signals is performed in order to determine if further testing in the particular region in warranted. This method of temperature detection however, does not provide a means for adjustment of the radiometer during the collection of the signals. Furthermore, only two points are required for a comparison to be performed, wherein this type of comparison does not take into account surrounding tissue of the breasts during the comparison analysis.
Therefore there is a need for a new system for identifying thermal variations in breast tissue that accounts for interface reflection of energy and provides reduced scanning time for each location, while providing a means for self-calibration during scanning. In addiϊion, there is a need for new procedures for the identification of thermal variations in b east tissue tha accounts for surrounding tissue effects.
This background information is provided for the purpose of making known information believed by the applicant to be of possible relevance to the present invention. No admission is necessarily intended, nor should be construed, that any of the preceding information constitutes prior art against the present invention,
SUMMARY OF THE INVENTION
An object of the present invention is to provide a system and procedures for identifying thermal variations in breast tissue. In accordance with an aspect of the present invention, there is provided a system for detecting temperature in breast tissue of a
patient, said system comprising: an antenna positioned at a predetermined point of the breast tissue of a patient, the antenna tuned to predetermined frequency band enabling collection of microwave energy signals from the breast tissue at the predetermined point; electrical circuitry interconnected to the antenna, said electrical circuitry for receiving the microwave energy signals from the antenna and amplifying said microwave energy signals thereby producing amplified analog signals; an analog to digital converter interconnected lo the electrical circuitry for receiving the amplified analog signals and converting the amplified analog signals into digitised raw data; and a digital filler interconnected with the analog to digital filter, said digital filter filtering the digitised raw data thereby enabling the determination of a temperature reading.
BRIEF DESCRIPTION OF THE FIGURES
Figure I is a schematic diagram of a device including a channel for internal temperature measurement and a channel tor skin temperature measurement.
Figure 2 is a cross-section elevation view of the microwave antenna according to one embodiment.
Figure 3 is a cross-section plan view of the microwave antenna according to one embodiment.
Figure 4 is a schematic of a portion of the device defining the radiometer && illustrated in Figure 1.
Figure 5 is a schematic of a portion of a device according to one embodiment of the present invention.
Figure 6A and 6B are graphs representing the transient response for different integrating circuits.
Figure 7 is a graph of an averaged transient characteristic for different integrating circuits.
Figure 8 is a graph of the relationship between internal temperature measurement inaccuracy and measurement time,
Figure 9 is a schematic of a portion of a device according to another embodiment of the present invention.
Figure 10 is a schematic of a portion of a device according to another embodiment of the present invention.
Figure 11 is a diagram of mammary glands identifying temperature evaluation locations according to one embodiment of the present invention.
Figure 1 is a diagram of mammary glands identifying temperature evaluation locations according to another embodiment of the present invention.
Figure 13 is a diagram illustrating the internal temperature fields of a patient's mammwy glands detected using a device according to the present invention.
Figure 14 is a diagram of an internal temperature thermogram.
Figure 15 is a series of graphs illustrating risk areas according to 6 signs of mammary gland cancer according to one embodiment of the precent invention.
DETAILED DESCRIPTION OF THE INVENTION
Definitions The term "digitised raw data" is used to define raw data collected from the subject in an analog formal and subsequently digitised prior to integration thereof.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The present invention provides a system and procedures for identifying thermal variations in breast tissue. The system comprises at least two temperature measuring channels and a device enabling the evaluation and visualization of the results of a scan. One channel is used to measure skin temperature using an IR-sensor and a further channel is used to measure internal temperature using a microwave radiometer. The present invention further provides procedures for determining temperature variations of breast tissue based upon the thermal measurements collected using the system of the present invention.
Figure 1 illustrates a system for detecting thermal variations within tissue wherein the collected data is maintained in an analog format during filtering and sensing and is only converted into a digital signal prior to transmission to a temperature indicator and/or a computing device. This system comprises at least two channels for temperature measurement. One channel enables skin temperature measurement by means of an IR- sensor 19 and a further channel represented by components 1-15 forms a radiometer or radio-thermometer that provides a means for the measurement of internal temperatures of a subject in the region of the placement of the antenna 1. The radiometer comprises an antenna 1, switch 2, circulator 3, radio-frequency amplifier (RFA) 4, band-pass filter (BPF) S, amplitude detector 6, tow-frequency amplifier (LFA) 7, multiplier 11, integrator 12, amplifier 13 and a resistor heated by a Peltier element 10. The reference voltage generator (RVG) 8 controls the switch 2 and multiplier 11. This schematic represents a balanced null-radio eier wherein during operation the temperature of register 10 is representative of tins internal temperature of the patient. The temperature of the resistor tO is being measured by the temperature sensor 9 and this signal is being amplified by the amplifier 14 and accumulated in the integrator 15, Through the use of a mode select switch 22, switching between the measurement modes for internal and skin temperatures, microwave and infrared detection respectively, is provided. This signal subsequently arrives at an analog-to-digital converter (ADC) 16, wherein upon conversion, this information passes to the temperature indicator 17 and the personal computer 18, where the evaluation of the results together with temperature field visualization occurs.
In one embodiment of the invention, the antenna is a printed radiating slot antenna with a dielectric support, wherein this antenna is placed inside a metal shield fabricated as a
cylinder which can improve interference protection. Figure 2 illustrates an elevation cross-section of this form of antenna, where the coaxial input 33 is attached to a metal plate 30 at points 34 and 35. The metal plate 30 is mounted on a dielectric disk 31, with a metal shield 32 enclosing the antenna. Figure 3 is a plan cross-section view of the antenna, showing the shape of the metal plate 30 and the slot 36, and the positions of the attachment points 34 and 35. Other designs of a microwave antenna are possible for integration into the detection device of the present invention and would be readily understood by a worker skilled in the art.
In one embodiment of the invention, a high accuracy IR sensor with a measuring inaccuracy of less than 0.5% is incorporated into the system. For example, an infrared probe of this type comprises a parabolic reflector, a rotating modulator or shutter and optical sensors for detecting the infrared radiation, wherein these components are installed within a housing. During operation, the entry of radiation into the probe is modulated by the shutter which eclipses an aperture at a predetermined frequency. During the open phase, energy enters into the infrared probe and is directed towards the parabolic mirror where this energy is oc se and redirected to ths optical sensor. This optical sensor converts the detected radiation into a signal that is sent to an analog to digital converter prior to transfer to a temperature indicator and a computing device, for example, Other forms of infrared sensors could be integrated into the present invention as would be readily known to a worker skilled in the art.
According to the present invention, the determination of the internal temperature reading performed by the system takes into account the reflection of energy at the antenna/patient boundary. Specifically, thermal radiation from the patient which is proportional to the internal temperature Tx arrives at the antenna/patient boundary and a fraction of this energy proportional to Txf2 reflects from the boundary and decays within the patient, wherein the value of r is the reflection factor for the boundary condition, As such the energy which enters the antenna is proportional to Tx l-r2), With reference to Figure 4, when the switch 2, is in an closed state and the antenna is consequently connected to the circulator 9, the resistor heated by the Peltier element 10 is also transmitting energy at a level of Tπ towards the antenna 1 wherein this energy is delivered via the circulator 3. Upon reaching the antenna/patient boundary a portion of this energy from the resistor 10 is reflected back to the circulator 9 with this being
proportional to T T2 and the remainder of this energy, Tn(l-r2) travels into the patient As such the energy that is being transmitted by the circulator 9 to the radio-frequency amplifier (RFA) 4, when the switch is in a closed configuration, is equal to Tx(l-r2) + T„r2. Therefore in the case where Tx is equal to TB, compensation of the reflection of energy which occurs at the antenna/patient boundary will be realised, As such the determination of T„ enables the evaluation of Tx, since these values are equal under the desired condition. The temperature of the resistor heated by the Peltier clement 9 is directly proportional to the energy being produced thereby the temperature of the resistor is equivalent to the internal temperature of the patient at the location of the antenna, when the condition of Tx = T„ is achieved.
With reference to Figure 4, during the scanning process the switch 2, which can be for example p-i-n diodes, modulates between open and closed states at a predetermined number of cycles per second. The reference voltage generator 8 provides the control of both the switch and the multiplier 11. In a closed state the switch interconnects the antenna with the circulator at its first branch and as such energy detected by the antenna is transmitted to the circulator. The energy emanating from the patient, Tκ, is partially reflected at the antenna/patient boundary, such that T5!(l-r2) is transmitted from the antenna to the first branch of the circulator. At the same time the resistor heated by a Peltier element is transmitting energy at a level of Tm wherein this energy is a result of the resistor heating by signals transmitted thereto by the amplifier 13. This energy T„, is transmitted to ihe third branch of the circulator and subsequently transferred owards the antenna via the circulator. A portion of this energy Tnϊ'a is reflected from the antenna/patient boundary back towards the circulator at its first branch. Therefore, when the switch is in the closed state, the energy being transferred from the first branch of the circulator to the second branch thereof, is equal to Tκ(l-r2) + TnF2. This energy is subsequently passed through radio frequency amplifier 4, band pass filter 5, amplitude detector 6, a low frequency amplifier 7 tuned to a modulation frequency and finally to the multiplier 11. When the switch is in the closed state, only energy from the heated resistor 10, which is proportional to T„, is directed to the circulator wherein this energy is transferred towards the switch by the circulator and all of this energy is subsequently reflected back towards the circulator, due to the open condition of the switch. This energy subsequently passes through the same circuitry towards the multiplier 11. The processed signal arriving at the multiplier during the closed state of the switch is
assigned the opposite sign to that of the processed signal arriving during the open state of the switch, wherein this sign is provided by the multiplier which is controlled by the reference voltage generator 8. As such upon the integration of these signals by the integrator 12, the difference therebetween is determined and this difference is subsequently amplified and transmitted to the heated resistor as an input which results in a change in the temperature and energy output thereof. If the difference between the energy created by the heated resistor, Tn, is equal to the energy passed through the circulator during the closed state of the switch, namely Tx(l-r2) + Tπr2 , the condition of Tx being equivalent to Tπ is satisfied and the compensation of the reflection of energy at the antenna patient boundary is achieved. Therefore, by evaluating the temperature of the heated resistor 10 by the temperature sensor 9, provides a means for determining the internal temperature of the patient at the location of the antenna. The signal from the temperature sensor is subsequently amplified and integrated and passed to the analog to digital converter 16 prior to transmission to the temperature indicator 17 and the computing device 18 for analysis.
Having particular regard to the system as illustrated in Figure 4, according to Dike's formula the fluctuation of the accuracy of the temperature measurements from this system relates to the integrator time-constant, and this fluctuation can be defined as follows:
T + T- S - 2 —j=A, (1)
where β - fluctuation accuracy standard deviation; T - noise signal temperature, K; Tjπp - the radio-thermometer input noise temperature including losses in its input circuits;
Tim - integrator time-constant, K; Δf- the radio-thermometer front end pass bandwidth,
According to Dike's formula the integrator time-constant is defined by the fluctuation accuracy level, radiometers intrinsic noise, input circuits losses and front-end bandwidth. The radiometer according to Figure 4 has an integrator with a fixed time-
constant which is related to the particular RC-circuit being used. Therefore, the fluctuation of accuracy can be defined by the RC-circuit time-constant and does not depend on measurement time.
The system according to the present invention comprises an ADC 22 together with a digital filter having an increasing time constant 23, as illustrated in Figure 5. The use of an analog-to-digital converter and arithmetic device that operates as a digital filter with an increasing time-constant can provide a means for noise signal accumulation. With reference to Figure 5 and Figure 4, this feature can be provided by the replacement of the RC-circuit integrator 15 and the ADC 16 as shown in Figure 4, with an ADC 22 together with a digital filter having an increasing time constant 23. This digital filter can be provided its functionality by a microprocessor for example or other integrated computing device as would be readily understood. For example, a practical realization of the increasing integration constant is represented by a gradual increase in the value of ' N according to Equation 2 below, wherein this accumulation procedure should begin with small values of a time-constant.
By using a digital filter as an integrator, the digital filter arithmetic unit output temperature can be described by the following equation: r - £ ι ' (2)
It is important to note thftt for a digital filter with an increasing timc-constanl the ccurac of measurement will depend on the measurement time. Therefore in order to get a higher accuracy a physician or technician can increase the measurement time for a particular location, while at the same time, a measurement with a reduced accuracy may be achieved relatively fast. With reference to Figures 6A and 6B, experimental transient responses for a system having a RC-circuit integrator and the system according to the present invention having a digital filter with an increasing time-constant are plotted for comparison. These figures indicate that the integrator with a gradually increasing time- constant possesses higher operating speed and that the results become more stable and the accuracy increases as the measurement time increases,
In one embodiment of the present invention the transition response is represented by a random sample. Figure 7 presents the experimental transient response characteristic
averaged over six samples plotted against measurement time, This plot demonstrates the advantages of a digital filter with a gradually increasing integration or time constant, as opposed to a RC circuit ith a fixed time constant. It can be seen from this plot that the measurement time is determined by the time required for noise signal accumulation and by the transient response time.
In one embodiment, one is able to determine the time required to reach a predetermined accuracy fluctuation level as defined by Dike's formula and illustrated graphically in Figure 8. Based on this factor, the measurement time for a digital filter is determined to be equal to 2τ. In addition, Figure 8 presents a transient response characteristic for the RC-circuit integrator at the temperature difference ΔT - 1 degree.
Δ = ΔT -(l -e(-l t>) (3)
Upon the comparison of these two plots in can be determined that for an accuracy level better than 0.1 degrees Celsius, the measurement time is determined by the transient response time. In particular, in order to reach a fluctuation accuracy of 0.15 degrees one should accumulate noise for a period of 1.5 sec and in order lo obtain the same accuracy level as defined above, the required transient response time should not be less than 5.7 sec, as illustrated in Figure 8. Having regard to a regular RC-circuit integrator, the transient response time is 3τ and therefore as the integration time-constant increases the measurement time also increases. According to the present invention with a system incorporating a digital filter having an increasing time-constant, the measurement procedure commences with small t-values, which results in the transient response being brief. The additional time during which the measurement of a desired position on the patient is being collected can provide a required level of fluctuation accuracy as determined by Dike' s formula.
According to alternate embodiments of the present invention, the following provides a theoretical basis for additional features that are incorporated into the radiometer according to Figure 5, wherein these additional features are schematically presented in Figures 9 and 10.
Ti - patient internal temperature, βC;
Tr- heated resistor temperature, °C;
T0- temperature inside the radio-frequency compartment of the radiometer, °C;
Tj„d- indicator temperature, wherein the indicator temperature has a linear relationship with that of heated resistor;
Tiή<i-A Tr+B, ( ) wherein A and B are a scaling factor and a constant component that can be controlled during the system calibration procedure,
Taking into account a feedback loop for the schematic under consideration and illustrated in Figure 5, the measurement inaccuracy, I, may be described as follows:
wherein;
C = (αs- r2αtt- I K)/αs; (5a) αs- the switch transmission gain; αB - the circulator transmission gain; r2 - reflection factor for the switch in the open mode; K « an open feedback loop transmission gain.
This equation is determined based on the fact that the temperature inside the radiometer radio-frequency compartment is equal to T0, wherein this temperature varies along with that of the ambient air.
As the radiometer radio-frequency housing is typically made of metal, one can assume as 3 first, approximation that all the parts of the radio-frequency compartment are at an equivalent temperature, wherein this assumed condition can simplify formulas that describe the radiometer.
According to Equation 5 the temperature being measured is proportional to the temperature on the inside of the compartment and as such depends on the ambient temperature. Thus if the measurement inaccuracy is equal to zero, then B - T0C as I ~ 0
For the system according to the present invention, Equation 5 is valid at a fixed ambient temperature,
Fn the general case of an arbitrary ambient temperature, T
0, the brightness temperature measurement inaccuracy is equal to:
ι=r
lna -τ^- (r. -r^Jc , (7)
Therefore the measurement accuracy has a linear relationship with respect to the ambient temperature T0 and the proportionality factor, C as defined by Equation 5a, wherein C depends on the characteristics of the switch and circulator in addition to the transmission gain of the open feedback loop.
Based on the above and according to another embodiment of the present invention, the accuracy of the internal temperature measurement of a subject may additionally be increased by detecting the temperature within the radio frequency compartment of the radiometer, thereby enabling the compensation of this internal temperature. This form of compensation can be enabled by the use of Equation 7 during data reduction of the collected data. For example and with reference to Figure 9, a temperature sensor 20 can be used to evaluate the temperature within the radio-frequency compartment and this sensed temperature in addition to the output of the heated resistor temperature sensor 9 can be transmitted to a two-input analog-digital converter 22. In one embodiment an arithmetic device 21 can sum these temperatures from the two temperature sensors 9 and 20 according to Equation 4, wherein the value of B is defined by the inside temperature of the radio-frequency compartment and can be determined as B - T0C, wherein 1 = 0 and Tina = Ti. In this manner the changing temperature within the radio-frequency compartment can be accounted for during the evaluation of the internal temperature of a subject.
In another embodiment of the invention and with reference to Figure 10., the incorporation of an adjustable attenuator 19 may be able to control the feedback transmission path gain K. By the incorporation of an adjustable attenuator, the value of the proportionality factor C as defined for Equation 7, may be maintained at a zero value, thereby increasing the accuracy of the internal temperature measurements made using the system of the present invention. In this manner there may no longer be a dependence between the measured internal temperature and the ambient temperature.
In one embodiment of the present invention, the circuitry for the radiometer can be integrated onto a circuit board that can subsequently be integrated into a computer for the further data reduction in order to identify thermal variations in breast tissue.
Optionally, the system can be separate from the computing device, personal computer or laptop or other device as would be readily understood.
In the embodiment of the present invention, the subroutines and algorithms contained within the microprocessor associated with the radiometer can be altered thereby providing a means for adjusting these subroutines and algorithms as new versions are developed thereby providing a means for keeping the detection devices up to date independent of the date of manufacture, This can be performed if the microprocessor has flashable memory or other procedures, for example as would be readily understood by a worker skilled in the art.
Digitizing Data Prior to Integration
The design of the radiometer shown in Figure 4 employs a process whereby the signal from the output of the amplifier 14 is integrated by the integrator 15, and then digitized by the ADC 16. This integration process provided by this analog method loses the details and results of the output of the temperature sensor, providing only a cumulative total,
A significant improvement can be achieved by digitizing the information prior lo integration, thereby resulting in digitized raw data, wherein a large amount of detailed information generated by the radiometer is then made available for use in a range of useful applications. Figures 5, 9 and 10 show radiometer configurations according to the present invention wherein the output of the amplifier 14 is digitised by the ADC 22 into digitized raw data and then is digitally integrated by a digital filler 23, such that the output from the digital filter 23 is comparable to the output of the ADC 16 illustrated in Figure 4. In addition the digitised raw data can be stored and manipulated using a PC, an onboard microprocessor or other similar computing device, wherein an advantage of this radiometer design is the retention of significantly more information than in the system where the analog signal is integrated and then digitized.
This detailed information provided by the digitization of the raw data resulting in the creation of digitized raw data may be usefully employed in range of applications,
In one embodiment the digitised raw data provides a measure of the signal with time as the radiometer senses the information relating to the temperature of a sample over a time period. It is probable that this measured temperature variation with time will not be a straight line, but will have both randomness and a pattern. This variation may be caused by the sample under test varying, for example the sample heating up due to the presence of the probe, or may be due to a variation in time in the operation of the radiometer. By analyzing the discrepancy between measured value and the expected linear variation, assessments may be made for the reasons of occurrence and corrections rectifying the situation can be implemented. Such an analysis may use a look-up table of data, and a set of signatures indicative of particular radiometer mechanisms, For example certain components may change their performance in a start-up mode, requiring a brief turn on prior to testing. The use of the digitised raw data allows a detailed insight into the variation of the temperature measurement.
In another embodiment the availability of digitised raw data can provide a much quicker estimate of whether the system is working properly, By analyzing the digitized raw data and performing an assessment relative to the expected input, a computing device is quickly able to determine whether the system is operating within the expected parameters. An alarm and/or a reset capability would be used to alert any operator to the problem. Such a feature is useful in allowing any appropriate warm-up time to elapse, and ensuring that the equipment is fully operational.
In a further embodiment, digitised raw data allows ths calibration of the equipment to ensure that the test system is working to its declared specification. The radiometer probe can be attached to a reference load, and the resulting digital output compared with a required standard. The standard should be a controlled simulation of a reproducible sample, equivalent to the type of tests the radiometer would normally test. Such a standard might be a thick piece of plastic with a small heating element inside. The standard methodology would require a measure over time of the temperature of the sample. The u$e of the digitised raw data would allow a detailed examination of the radiometer's accuracy. The standard would have a maximum allowable deviation.
In another embodiment, digitised raw data is able to provide information concerning the possible variation of the measured temperature by the radiometer for a specific sample.
This sample variation may provide important information concerning the subject sample under test. For example when a patient is given medication to increase the temperature of certain organs, a computing device can analyse the digitised raw data, producing a time-phased variation of the temperature of the sample. Such a variation can provide valuable evidence of a disease.
In a further embodiment, a computing device can analyse the digitised raw data to monitor the temperature of a sample, wherein the proximity of the test probe may alter the temperature of the sample, thus causing an erroneous reading by the radiometer, For example a cold probe may reduce the temperature of the sample. By monitoring the temperature given by the digitised raw data, variations in the measured temperature with time may be identified. Analysis by a computing device can generate a model of heat change of a sample to provide an indication of the impact of the measuring device on the results.
In a further embodiment, the digitised raw data can provide input to a computing device for the real-time correction to the radiometer system. For example use of self- calibration techniques can be implemented by the adjustment of attenuators or other components to correct for any inconsistency, or change in other components. There are multiplicity of opportunities within the radiometer system for such probe points and feedback correction, Such correction includes self-calibration and internal adjustment providing for s. much more repeatablβ and accurate testing for diseased tissue, resulting in a higher confidence of diagnosis.
In a further embodiment, the digitised raw data provides a more accurate comparison between different radiometers. The use of the digitised raw data can be analysed by a computing device to ensure the repeatability of the accuracy of the different radiometers, ensuring a patient measured by one unit will get a very similar result on another machine. A database of digitised raw data performances by the radiometers, taken against a measurement standard will provides assurance of consistency of performance.
In a further embodiment, the use of digitised raw data provides a better interface to a computing device such as a Digital Signal Processor, using less electronic components,
and allowing more options in the processing of information, shorter processing time, predetermined accuracies providing an improved control of the radiometer system.
In a further embodiment the use of digitised raw data is able to handle predetermined changes, such as a fast clocking reference signal or other changes in the radiometer system whereby a processor could evaluate the difference in the signal levels, and calculate associated temperature levels. For example the radiometer may be programmed to change the input bandwidth. In such a case the use of digitised raw data can easily separate the data into the different sets appropriate to the different input bands.
Other embodiments as would be envisaged by a worker skilled in the art using the digitised raw data for the improvement of the performance of the radiometer.
Use of the System for Detection of Thermal Variations in Breast Tissue
The system according to the present invention can provide __» non-invasive method for the evaluation of the potential that a patient may have breast cancer based on thermal variations of breast tissue. Upon the determination of the thermal variations a determination can be made if further analysis or testing of the breast tissue is required using alternate methods. During a breast examination using the present invention, a patient lies supine with her hands above the head. The evaluation of the temperature of the mammary glands can be performed following a common procedure, for exam le as described in A.H. Barrett, Ph. C, Myers, N.L. Sadovsk (t Microwave Thermography in the Detection of of Breast Cancer" AJR; 134, February 1980, pp.365-368 and Rahlin V.L., Alova G.E. "Radio-Thermometry in Diagnostics of Mammaiy Glands, Genitals, Prostate Gland and Spinal Column Pathology". Preprint #253, Gorki, NIRFI, 1988, p.52.
In one embodiment, the internal temperature of one position of the right mammary gland is determined and subsequently the corresponding spot of left mammary gland is evaluated. These measurements commence at the right nipple, followed by evaluation of the internal temperature of the left nipple. The evaluation of the temperature of the upper quadrants boundary of the right mammary gland is performed and then that of left mammary gland, and so on in a clockwise fashion until all 9 positions of each mammary gland illustrated in Figure 11 are evaluated. While this is one examination technique
regarding the locations of data collection in order to determine the internal temperatures of the left and right mammary glands, other examination techniques may be used that collect more data points around the mammary glands or fewer data points. The concept for data collection is to enable the determination of a temperature profile within the mammary glands and an increase in the number of data collection points may improve the accuracy of the temperature profile for the subject, there is also an increase in the examination time as would be known to a worker skilled in the art. In addition, the sequence of the collection of temperature measurements from the mammary glands may not be in the sequential fashion as above described, but the complete evaluation of one mammary gland may be performed prior to the evaluation of the second or any other sequence for example two location on the first mammary gland and the symmetric two locations on the second mammary gland.
In one embodiment, the evaluation and measurement of the internal temperature of the mammary glands is followed by auxiliary area temperature measurements. In order to perform a measurement an ex ert or technician places the antenna-applicator at a test position, waits a few seconds for the temperature reading to stabilize and this evaluation of the stability of the reading can be electronically indicated by for example a green light at the computer screen. Upon the determination of the stability of the measurement the measurements can be logged into the computer, wherein this logging action of the measurement can be activated by the pushing of a button associated with the system by the ex e t or technician. The antenna-applicator can subsequently be moved to the symmetric position of the other mammary gland for internal temperature evaluation thereof. Upon the completion of the internal temperature measurements, the evaluation of skin temperature at the same locations can be determined.
In one embodiment, after the completion of the temperature measurements, evaluation of the results is performed and graphical representations can be generated. For this purpose internal and skin temperature fields are created and plotted. In order to generate temperature field images as illustrated in Figure 13, a thermal contour map can be approximated using the measured temperatures of the predetermined positions of each mammary gland. Subsequently, the temperature contour map can be colour coded for example assigning a color to each temperature value or range of temperature values. For example, higher temperatures can be assigned the colour red, low temperatures can be
assigned the colour blue and temperature values therebetween can be assigned the colour green. In addition, thermograms as illustrated in Figure 14 can be plotted in a manner as defined in by Rahlin and Alova. In Figure 14, corresponding locations on the right and left mammary glands are plotted side by side and these measurements can be connected by a solid as illustrated. For example the measured temperature of a position on the right mammary gland is assigned an "O" and a measured temperature on the left mammary gland is assigned an "X", In Figure 14, the first position plotted is the null- position of the right mammary gland, then the null-position of the left mammary gland, subsequently the first position of the right gland and the first position of the right one, etc.
In one embodiment, the evaluation of the results of a scan using the present invention is performed using a number of indicators, wherein these indicators can provide a means for one to segregate patients with a possibility of mammary gland cancer from healthy ones, such that psueudo-positive patients may require further consideration. Some of the indicators used are as follows;
1, Difference between nipple temperature and mammary gland mean temperature: S^t„ -7^, where t -- Ϋ -V-
i - temperatures in 8 spots of the mammary gland (1...8 as illustrated in Figure 11), t„ - nipple temperature,
2. Nipple temperature difference between right and left mammary glands;
3. Maximal temperature difference in the corresponding positions of the right and left mammary glands; S3 = maχ|/r ( -t _j, i =1,8;
temperatures at spots I - 8 are being compared.
4. Mean-square value of temperature differences in the corresponding positions of the right and left mammary glands:
temperatures at spots 1...8 arc being compared in pairs, see Figure 11.
5. Moan-square value of temperature difference in the corresponding positions of the right and left mammary glands:
_ 2 , where At, = /_,, -/,.,, t
6. Mean-square value of temperature scattering in one of the mammary
15 Being calculated for each mammary gland separately using spots 0...8, the bigger of two values being taken for further consideration.
7. Difference in mammary gland mean temperature for the patient under examination and that within given age group: ηn v _ nt- ^'Im ■•""——
*υ ^o 2 ■"'«-»«« "
These indicators that are determined for a patient being evaluated are compared with the respective indicators of at least one individual having verified mammary gland cancer. This information relating to the subject under examination and the verified cancer 25 patient(s) can be plotted on 6" two-dimensional charts with X0 along X-axis and Sj along Y-axis as illustrated in Figure 15, wherein each patient corresponds to a single plotted point. For example, if the determined indicators for a patient under examination fall
inside those of a group of patients that had verified mammary gland cancer, this patient can be assigned a thermo-positive result. Alternately, if the determined indicators for a patient under examination fall outside those of a group of patients that had verified mammary gland cancer, this patient will be assigned a therm o-πegative result. In addition, for example the location can be disclosed on the temperature field chart and on the thermograms thereby potentially identifying a possible location of a tumor, for example.
It should be noted that the evaluation procedures of measurement data as described above do not account for mammary gland skin temperature measurements. Therefore, in one embodiment, an improved diagnostic accuracy may be realized if measurements of internal and skin temperatures at two additional reference spots as illustrated in Figure 12, namely RSI and RS2. The evaluation of these additional reference points is made together with the evaluation of the temperatures of the mammary glands also as illustrated in Figure 12. Subsequently, the integral thermo-asymmetry index can be calculated and compared to a predetermined threshold. The integral ther o asymmetry index depends on local internal and surface thermo-asymmetry, skin and surface temperatures ratio and on the internal temperature ratio of the predetermined locations of the mammary glands,
In one embodiment of the invention, an important task with regard to evaluating mammary glands temperature variations is the reduction of uncertainty of pseudo- positive and pseudo-negative results. Virtually all the procedures based on an internal temperature measurement deal with the mammary gland internal temperature measurements only. Thus during the course of examination, it can be essential to determine if the mammary gland temperature is the only one that varies with respect to a particular patient, or if the temperature variation occurs within the whole body of the patient. Therefore the present invention incorporates the measurement of internal and skin temperatures at two reference positions. For example, the first reference position can be located at the center of the chest, 5 cm above the epigastric spot and the second reference measurement position can be located at the epigastric position projection. These reference positions are graphically defined in Figure 12. Other reference positions can be used, however these may be appropriate due to the proximity to each of the mammary glands and therefore potentially limiting the number of reference
positions. For example the reference positions may be located at a position on the chest cavity proximate to each arm, however in this example additional reference positions may be necessary since one set of reference positions may only be appropriate for evaluation of the mammary gland proximate thereto.
In one embodiment of the invention, the procedure for the collection of data by the system may commence with the evaluation and collection of a measurement of the internal temperature at the first reference position, followed by the internal temperature measurement at the second reference position in the epigastric area. Subsequent to evaluation of the internal temperature of the reference positions, the internal temperature of the mammary glands is evaluated and measured. In one embodiment the internal temperature measurements can start with the measurement of the right mammary gland nipple; followed by the left nipple, proceeding to the first position at the left mammary gland, etc., up to the 8lh position as identified in Figure 12. Subsequent to the internal temperature evaluation of these locations, the skin temperature can be determined for the same positions. This se uence of measurement may be modified in a plurality of ways including the complete collection of measurements associated with a first mammary gland prior to the collection of the information relating to the second mammary gland or for example the collection of two locations with regard to the first mammary gland and the subsequent measurement of the symmetric locations on the second mammary gland. In addition, the collection of the internal temperature value for one location may be followed or preceded by the measurement of the skin temperature at the same location.
In one embodiment of the invention, upon the completion of the collection of the temperature measurements for a particular patient, the subsequent step is the analysis of the collected information. For this purpose internal and skin temperature fields are created and plotted. In order to generate temperature field images as illustrated in Figure 13, a thermal contour map can be approximated using the measured temperatures of the predetermined positions of each mammary gland. Subsequently, the temperature contour map can be colour coded for example assigning a color to each temperature value or range of temperature values, For example, higher temperatures can be assigned the colour red, low temperatures can be assigned the colour blue and temperature values therebetween can be assigned the colour green. In addition, thermogra s as illustrated in Figure 14 can be plotted. In Figure 14, corresponding locations on the right and left
mammary glands are plotted side by side and these measurements can be connected by a solid as illustrated. For example the measured temperature of a position on the right mammary gland is assigned an "0" and a measured temperature on the left mammary gland is assigned an "X". Tn Figure 14, the first position plotted is the null-position of the right mammary gland, then the null-position of the left mammary gland, subsequently the first position of the right gland and the first position of the right one, etc.
As was previously mentioned, in the prior art typically a single sign was used in order lo segregate patients into two groups, namely ones having a high probability of having breast cancer and ones having a low probability of having cancer, wherein this is determined by local thermo-asymmetry, for example, the maximal temperature difference between corresponding locations of the right and the left mammary gland.
According to the present invention, as presented above, a procedure is provided for evaluating the internal temperature measurement results by correlating several factors.
An improvement of this analysis may be to incorporate the internal and skin temperature measurements obtained from the reference positions in addition tq the skin temperature ratio of the left and right mammary glands (local thermo-asymmetry of the skin temperature). By taking these additional factors into account the accuracy of the results from system according to the present invention may be improved.
In addition, the analysis of the measurement results of a patient having large mammiry gland diameters should be performed separately from those patients having small mammary gland diameters. For example, a patient having small mammary gland diameters, the surface or skin temperature may be of great importance and for a patient having large mammary gland diameters the internal temperature thereof may be of more significance.
Barrett had identified that the local thermo-asymmetry between the internal temperature of the mammary glands is an extremely important sign for the identification of mammary gland cancer. However, additional factors should be taken into consideration as well.
In one embodiment, the extent to which the mean internal temperature of the mammary glands of a patient exceeds the mean temperature of the mammary glands for the given age group may also be an important indicator. For example, mammary gland cancer can be accompanied by the elevation of the mean temperature of the mammary glands.
In another embodiment, the difference between the skin temperature of the mammary glands and the internal temperatures thereof may be another important factor. For example, it has been determined that the internal mammary gland temperature of healthy patients is typically approximately Ϊ.5 to 2 °C above the skin temperature of the mammary glands. For patients with mammary gland cancer, the skin temperature of the mammary glands tends to be closer to the internal temperature thereof.
In another embodiment, the difference between the mean internal temperature of the mammary glands and the internal temperature of the first reference position is another important factor. For example, it has been determined that for healthy patients, the difference between the mean internal temperature of the mammary glands and the internal temperature of the first reference position is typically 1.5 - 2 °C. For patients wilh mammary gland cancer (his difference tends to be smaller.
In one embodiment of the present invention, the evaluation of a patient in terms of determining whether the patient is either pseudo-positive or pseudo-negative for mammary gland ca ce can be determined using the following technique, The following defines the symbols which are used during the analysis øf th@ collected data in order to evaluate the diagnostic condition of a patient. This analysis can be modified depending on the number of locations that have been measured as would be known to a worker skilled in the art.
T T
Represents the internal temperature at positions i - 0 - 8 of the right and left mammary glands, respectively, wherein the location of positions 0 - 8 on the left and right mammary glands are illustrated in Figure 12.
'Sic i, T & $k. i
Represents the skin or surface temperatures of at positions i - 0 - 8 of the right and left mammary glands, respectively, wherein the location of positions 0 - 8 on the left and right mammary glands are illustrated in Figure 12.
T T
TΛRSJ ' 1(1RS_2
Represents the internal temperature at the first and the second reference positions, respectively, wherein the locations of the reference positions are illustrated in Figure 12.
T T skRS_l ' s :RS_2 Represents the surface temperature of the first and the second reference positions, respectively, wherein the locations of the reference positions are illustrated in Figure 12.
The local internal thermo-asymmetry or internal-temperature difference between the corresponding positions of the right and left mammary glands arc defined as follows: ΔTlnj =Tlni_R -Tlnj_L > 1 = 0 — 8
The local surface thermo-asymmetry or surface-temperature difference between the corresponding positions of the right and left mammary glands arc defined as follows: ΔTsk. = TSk. R - TSk. L , i = 0 - 8
The ma? mal internal thermo-asymm ry or maκimal value of the internal temperature differences between the corresponding positions of the right and left mammary glands are defined as follows: ΔTm„M =MAX|ABS(&Tini J , I - 0 - 8
The maximal surface thermo-asymmetry or maximal value of the surface temperature differences between the corresponding positions of the right and left mammary glands are defined as follows:
ΔTsk_M - MAX|ABS(ΔTsfc, J, 1 - 0 - 8
The maximal value of the surface and internal thermo-asymmetry sums' is defined as follows;
TM = MAX|ABS(ΔTlni +ΔTSkl ), i = 0 - 8 .
The mean internal temperature of the mammary glands is defined as follows
The mean surface temperature of the mammary glands is defined as follows: ■ β Tsk, _ + τskι R.
T& - ∑-
The difference between the internal temperature at the first reference position and the mean internal temperature of the mammary glands is defined as follows; ΔTltiRS-M
-Tsin
The difference between the surface temperature at the first reference position and the mean surface temperature of the mammary glands is defined as follows:
The difference between the surface tem eratures of the first and ths second reference positions is defined as follows:
ΔTsk S-12 =TskRSJ -"TskRSj ■
The difference between the internal temperatures of the first and the second reference positions is defined as follows: ΔTRRS_I 2 =TRRS_, -TRRS_2
The difference between the internal and surface temperatures of the first reference position is defined as follows;
ΔTjπ_SkRSJ =TrnRSJ ""TskRSJ '
The difference between the mean temperature of the mammary glands of the patient under examination to the mean temperature of the mammary glands for patients of the corresponding age group is defined as follow:
Through experimentation, the mean temperature values of the mammary glands for various age groups are presented in Table 1.
Upon the completion of the above evaluations of the information collected during the examination of a patient, an integral thermo-asymmetry value, ΔT, can be calculated and subsequently compared with a predetermined threshold, G. The determination of the integral thermo-asymmetry is defined as follows;
ΔT=Tln M+TSL M+ΔTln-
As an example, for a mammary gland having a diameter of approximately 22 cm, the value of ΔT]„ can be determined by the following: ΔTln =ΔTι +ΔTj +ΔT3+ΔT4, wherein;
if ΔΪIΠRS-M <0
As a further example, for a mammary gland having a diameter of 21 cm, the value of ΔTin can be determ ined by the fol lowing: ΔTin =ΔTι +ΔT2 +ΔT3 +ΔT4 +ΔT5 wherein;
if ΔTskRs-M 1
if ΔTskRs- ≥l
ΔT-, = — if Δ>0
2.5
0 if Δ≤O
Δ=T -T r" S-M skRS-M
For example, in the case where the integral thermo-asymmetry, ΔT exceeds the threshold, G, the patient is assigned to the high-risk group for mammary gland cancer. Alternately, if ΔT is less than threshold, G, the risk of mammary gland cancer is small. It has been established through experimentation that the integral thermo-asymmetry threshold, G, is equal to 2.2 βC.
Upon the collection of the temperature measurements as defined above, wherein these temperature measurements can be related to the temperature of the skin and/or internal temperatures of predetermined test positions or locations on a patient being examined, these measurements are analysed. This analysis entails the correlation between the measured skin temperatures and internal temperatures with respect to patients having confirmed cases of breast cancer. For example, if there is a close correlation, within a predetermined threshold, between the measurements for the patient being examined and the measurements of the collection of patients having confirmed breast cancer, the patient under ejtamination will be given a pseudo thermo positive result. If the correlation was outside the predetermined threshold, the patient would be given a pseudo thermo negative result.
The embodiments of the invention being thus described, it will be obvious that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the invention, and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.