US20170367644A1 - Apparatus and Method for Monitoring Rehabilitation from Joint Surgery - Google Patents

Abstract

Apparatus and methods are described for setting rehabilitation goals for a patient, measuring patient movements, storing the movement data for later transfer to a computer, displaying progress indicators and inspirational messages based on progress towards goals, reporting movement, skin temperature and swelling data to a caregiver so that they monitor compliance and be aware of potential infection. The invention consolidates data from motion, temperature and swelling sensors placed on the patient with data received directly from the patient in response to questions displayed on a tablet computer screen, to create one or more representative values indicating the patient's current status. These representative values may include the patient's degree of compliance with their prescribed exercise and icing regimen; their degree of wellness based on their self-reported pain scale and sensed temperature and limb swelling; and their current level of physical activity as detected by the motion sensors.

Description

FIELD OF THE INVENTION
• The present invention relates to apparatus and methods for monitoring a patient in various respects including the compliance of a patient to the rehabilitation regimen that is prescribed for preparation for and recovery from joint surgery such as total joint arthroplasty. More specifically, the present invention relates to the use of sensors applied to a post-surgical patient for the purpose of detecting, acquiring and measuring the patient's movement, temperature and limb circumference and for using acquired data for tracking the patient's progress during rehabilitation while reporting their progress to their caregivers.
• Joint arthroplasty is a surgical procedure for resurfacing or replacing those parts of knee, hip, elbow, shoulder and other joints that are damaged, typically from arthritis, in older adults. Knee arthroplasty is a very common procedure—more than 700,000 were done in the United States in 2015—and is rapidly increasing as a result of an aging population.
• A key factor in the success of a joint arthroplasty is the compliance of the patient with the required rehabilitation regimen. This regimen may begin prior to surgery—certain exercises and stretches are sometimes prescribed in advance of surgery to improve the chances of success—and is certainly required for some months after the surgery is complete. Rehabilitation may include such activities as flexing the affected joint through a certain range of motion, applying ice or heat to the joint, and monitoring the surgical site for signs of infection, excessive swelling or drainage.
• Rehabilitation is usually managed by a physiotherapist or other medical professional who instructs the patient in a clinical setting, then checks with the patient occasionally to monitor their progress. This means that the patient is expected to be self-motivated to follow the required regimen and accurately report to the caregiver their level of compliance. Many find this difficult to do and may not be entirely honest about their level of compliance. As a caregiver may be responsible for a large number of patients, it may be difficult for them to determine which patients are complying well and which need additional attention.
• To encourage better compliance, it would be advantageous to provide a patient with timely feedback and encouragement as to their progress, and to provide quantitative measurements as to their progress, both to the patient and their caregivers.
• Three kinds of measurements can provide information of value. The first is measurement of the number of flexions, degree of flexion and maximum and minimum amount of flexion of the affected joint; the second is measurement of the skin temperature near the surgical site; and the third is the degree of swelling of the limb. The number and the degree of flexions including the maximum and minimum flex angles is indicative of the patient's activity level and progress towards re-establishing a full range of motion. The temperature near the wound site can provide an early indication of infection as it has been known since Roman times that wound infection is indicated by the four factors of calor, dolor, rubor and tumor—heat, pain, redness and swelling. Further, icing of the wound site after surgery is indicated for improved recovery, therefore measuring the amount and duration of temperature decrease near the wound site is indicative of the patient's compliance with prescribed icing techniques. Similarly, increased swelling of the limb near the surgical site may be indicative of infection, overuse of the limb or a need for additional icing.
• Using electronic sensors to measure joint flexion has been demonstrated in the laboratory. Several published papers show the use of integrated circuit accelerometers or capacitive, resistive or inductive flex sensors to detect joint movements and range of motion. Similarly, there are many well-known ways to measure skin temperature using electronic and mechanical thermometers. Measurement of swelling is typically done manually, using a tape measure or similar device to measure limb circumference near the surgical site.
• Existing devices for measuring joint motion and temperature require separate sensors connected to a computer for collecting data for interpretation by a caregiver. These systems do not provide a convenient single unit for measuring the required parameters, nor do they provide for storage of the data for later transmission to a caregiver's computer. In addition, a single temperature sensor near the wound site may provide misleading data if the patient moves into a hot or cold environment, as there is no way, with one sensor, to tell if the temperature increase or decrease is a local effect (caused by infection or icing of the joint). Manual measurement of swelling provides only intermittent data and may be inaccurate as a result of varying measurement locations and measurement techniques.
• The prior art fails to teach the combination of data from motion and temperature sensors into a patient coaching system and caregiver management system. Such a system can be used by a caregiver to set specific goals (such as number of repetitions of joint flexion, target ranges of motion or target temperature and duration during icing) and to provide the patient with feedback and encouragement as to achievement of those goals based on measurements by the sensors.
• The prior art also fails to teach consolidating data from patient-worn sensors and patient reported data into one or more figures of merit which allow a caregiver to easily determine if the patient is doing well or not.
SUMMARY OF THE INVENTION
• The current invention describes apparatus and method for setting rehabilitation goals for a patient, measuring their movements, storing the movement data for later transfer to a computer, displaying progress indicators and inspirational messages based on progress towards goals, reporting movement, skin temperature and swelling data to a caregiver so that they monitor compliance and be aware of potential infection.
• One advantage of the current invention is the use of two temperature sensors to monitor patient skin temperature—one located on the skin near the surgical wound site and another on the skin some distance from the wound site, so that the wound site temperature can be compared to a basal skin temperature, eliminating environmental variations that might effect the temperature measurements.
• In another aspect, the sensor apparatus in accordance with the current invention provides for data storages and wireless communications between the sensor apparatus and a computer or computer network, such that readings made by the sensors can be stored within the sensor apparatus, then transmitted wirelessly to a computer or network whenever a wireless connection is available, therefore eliminating the need for the patient to remain within wireless communications range of a computer, without risking loss of measurement data.
• In yet another aspect, the current invention consolidates data from motion, temperature and swelling sensors with data received directly from the patient in response to questions displayed on a tablet computer screen, to create one or more representative values indicating the patient's current status. These representative values may include the patient's degree of compliance with their prescribed exercise and icing regimen; their degree of wellness based on their self-reported pain scale and sensed temperature and limb swelling; and their current level of physical activity as detected by the motion sensors.
BRIEF DESCRIPTION OF THE DRAWINGS
• These and other objects, features and advantages of the present invention will become apparent upon reference to the following detailed description of the exemplary embodiment presented herein and to the drawings wherein:
• FIG. 1 is a block diagram of an apparatus according to the invention.
• FIG. 2 is a schematic representation of the sensor and data logger components of the apparatus shown inFIG. 1.
• FIG. 3 is a schematic illustration of a post-operative patient's leg illustrating how the sensors, data logger and microcontroller may be applied to the patient's leg.
• FIG. 4 is a schematic illustration analogous toFIG. 3 except illustrating an alternative embodiment of an apparatus for locating the sensors on the patient's leg.
• FIG. 5 is a graphical display of how data obtained from patient-worn sensors according the present invention may be displayed on, for example, a computer monitor.
• FIG. 6 is a schematic illustration analogous toFIG. 3, except illustrating an alternative embodiment of an apparatus for measuring the flexion of the patient's leg.
• FIG. 7 is a schematic representation illustrating one preferred way that data generated according to the invention may be displayed to a caregiver who has responsibility for monitoring plural patients.
DETAILED DESCRIPTION OF THE INVENTION AND ILLUSTRATED EMBODIMENTS
• FIG. 1 illustrates the major functional components of the preferred embodiment according to the invention. Patient-worn sensors10 are connected to data logger andmicrocontroller12, both of which are described more fully hereinafter, such that the microcontroller can cause data to be read fromsensors10 and stored in memory. At pre-determined intervals,microcontroller12 tests to see if a connection tolocal computer16 can be made throughwireless data connection14, which in the preferred embodiment is a Bluetooth connection, but may be a WiFi or other data connection. If a connection is available,microcontroller12 retrieves data fromsensors10 from the memory and sends it acrosswireless data connection14 tolocal computer16.Local computer16, which in the preferred embodiment is an Android tablet computer, then transfers the data, usingInternet connection18, to database andserver20, which, in the preferred embodiment is a ‘cloud service’ such as those provided by Heroku and Amazon.
• Also connected to database andserver20 isremote computer24, via Internetconnection22, which may be any computer capable of running a web browser such as Google Chrome or the like. Thus, it can be seen that through the various devices and connections described, data fromsensors10 can be delivered to database andserver20, from where it can be retrieved byremote computer24 for viewing and interpretation by a user ofremote computer24.
• Note thatInternet connections18 and22 permit communications in the opposite direction to that described—remote computer24 can send information viaInternet connection22 to database andserver20, from whence it can be further sent tolocal computer16. In this way it is possible forremote computer24 to causecomputer16 to display messages, images, videos or other information onlocal computer16.
• FIG. 2 more fully illustrates patientworn sensors10 and data logger andmicrocontroller12. In he preferred embodiment, data logger andmicrocontroller12 is made up ofATMega32U4 processor board42, connected to SDcard memory unit50, which in the preferred embodiment is an Adafruit Feather 32u4 Adalogger (Adafruit Industries LLC, NY, N.Y.). Incorporated intoprocessor board42 isUSB connector44, serial communications connections, I2C bus connections and battery charging circuitry.Battery46 is a lithium polymer 3.7volt 12 mAh battery, which in the preferred embodiment is a PKCell LP503562, which is connected to the battery pins ofprocessor board42.
• Bluetoothradio48 is embodied with an Adafruit BlueFruit EZ Link module. This module is connected to the auxiliary power supply connections ofprocessor board42 and to the serial data transmit (TX) and receive (RX) pins ofprocessor board42.
• The remaining modules of data logger andmicrocontroller12 and patientworn sensors10 are connected toprocessor board42 using the industry standard I2C bus. This communications bus provides electrical power and digital communications to100 or more modules connected on the same set of four wires. As each device connected to the I2C bus has a unique digital address, the software running onmicrocontroller12 can request and receive data from each module as required.
• Real time clock40 is an I2C module based on the DS1307 real time clock chip. In the preferred embodiment, this is an Adafruit DS1307 Real Time Clock module, which includes a battery backup to ensure that real time clock data is preserved even ifbattery46 should become exhausted.
• Also connected to the I2C bus are two MCP9808 temperature sensors (Adafruit MCP9808) and two LSM9DSO motion sensors (Adafruit LSM9DSO).Motion sensor30 is connected to the distal end ofI2C cable39 so that it may be attached distal to the patient's affected joint as hereinafter described;temperature sensor32 is connected to thecable39 some distance proximal tomotion sensor30, such that it may be attached to the patient's skin near the surgical site; andmotion sensor34 andtemperature sensor36, are connected some distance proximal totemperature sensor32 so that they can be attached to the skin proximal to the patient's affected joint. As detailed below,temperature sensor32 is attached to the patient near enough to the surgical site that the sensor is capable of measuring increases (or decreases) in temperature at the surgical site, which could be indicative of infection (or icing).Temperature sensor36 is attached to the patient spaced away fromsensor32 by enough distance that thesensor36 measures a basal skin temperature that is not effected by an increase or decrease in temperature at the surgical site where thesensor32 is located.
• Also connected to the I2C bus is one ADS1115 analog to digital converter31 (Adafruit ADS1115). Connected to analog input A0 of analog todigital converter31 isstretch sensor33, which in the preferred embodiment is made of a length of 2 mm diameter conductive rubber cord (Adafruit 519).Stretch sensor33 is connected between 1K ohm pull upresistor35, which is connected to positive supply voltage V+ and negative supply voltage V−. This produces a voltage at the junction ofstretch sensor33 andresistor35 that varies in proportion to the extension ofstretch sensor33. This varying voltage is read by analog todigital converter31 via input connection A0.
• Cable39 connectingreal time clock40 andtemperature sensor36 includesconnector38, which allows thetemperatures sensors32 and36,motion sensors30 and34 andstretch sensor33 to be disconnected fromreal time clock40, thus making the module containingreal time clock40,processor board42,Bluetooth radio48,SD card memory50 andbattery46 separable from the sensor components.
• FIG. 3 shows how patientworn sensors10 and data logger andmicrocontroller12 might be applied to the leg ofpatient52 during recovery from knee surgery.Motion sensor30 is applied to the patient's leg below the knee and may be taped in place, attached to the surgical dressing, or tucked inside an elastic bandage applied to the leg. Analog todigital converter31 is mounted such that connectedstretch sensor33 may be wrapped around the leg immediately below the knee such that it is normally under slight tension. Similarly,temperature sensor32 is attached to the leg, but is located a near as practicable tosurgical incision54. In the preferred embodiment,temperature sensor36 andmotion sensor34 are contained within the same enclosure and are attached to the leg ofpatient52 above the knee. All of the sensors are connected withcable39, which is connected to data logger andmicrocontroller12 withconnector38.Microcontroller12 enclosesreal time clock40,processor board42,SD memory50,Bluetooth radio48,battery46 andUSB connector44.USB connector44 is accessible such thatmicrocontroller12 can be plugged into a standard USB cable to rechargebattery46 and to upload programs toprocessor board42. In use,microcontroller12 may be strapped to the leg ofpatient52 with an elastic strap, clipped on a belt, or placed in a pocket.
• FIG. 4 shows an alternative means for locating the sensors on a patient's leg in accordance with the invention, as it might be used in rehabilitation from knee surgery. In this embodiment,sensors30,3234 and36 are fastened insideelastic sleeve54, whilesensor33 and analog todigital converter31 are fastened outsideelastic sleeve54 all connected viacable39. The sensors are pre-positioned at locations inside and outside the sleeve such that when the sleeve is pulled up over the knee, the sensors are located in the desired positions. This has the advantage of simplifying the location and attachment of the sensors to the patient.
• In typical use, a caregiver usesremote computer24, to create a record for a new patient using a web application hosted by database andserver20. As part of this setup, the caregiver assignslocal computer16 topatient52, creating a link between the record for the patient andlocal computer16. The caregiver then pairs patientworn sensors10 tolocal computer16 so that data from patientworn sensors10 is transmitted tolocal computer16 usingBluetooth connection14 from where it is further transferred to database andserver20 overInternet connection18, where it is stored in a database record associated withpatient52.
• As soon as the connection is made, data logger andmicrocontroller12 begins to collect data from patientworn sensors10 and store it locally inSD card memory50. In the preferred embodiment, data is collected approximately every 1/10 second. From time to time,microcontroller12 checks to see if there is a connection tolocal computer16 available usingBluetooth connection14. If so,microcontroller12 transmits any data not previously transmitted tolocal computer16. In turn,local computer16 transmits the data to database andserver20 overInternet connection18.
• From time to time, the caregiver may choose to review the data collected by patientworn sensors10. Using a web browser onremote computer24, the caregiver can retrieve data from database andserver20. The web service running on database andserver20 retrieves the data obtained from patientworn sensors10 and performs an analysis of the data to extract features from the raw data.
• Many different techniques for extracting knee joint angles from accelerometer and gyroscope data are known in the art, many of which can be implemented with thesensors30 and34 herein described. For example, a first approximation of the knee joint angle can be determined using only the three-axis accelerometers ofsensors30 and34. In this implementation, the acceleration due to gravity is detected by each sensor to provide an X, Y and Z acceleration measurement that varies depending on the orientation of the sensor with respect to the ground. Assensor30 is attached to the shank ofpatient52 andsensor34 is connected to the thigh ofpatient52, the X, Y and Z axis readings from each sensor define a vector V that represents the orientation of the sensor on the shank or thigh, and the angle between the two resulting vectors represents the angle between the shank and thigh. The formula for determining the angle between two vectors V1 and V2 is:
• 
  θ=cos⁻¹(V1V2)/(|V1|×|V2|)
• Where  indicates the dot product of the vectors and |V| indicates the magnitude of the vector.
• Measuring only accelerations will give a reasonably accurate representation of knee flexion angle whenpatient52 is at rest, but will be less accurate when there is any motion. To improve the estimate of the actual knee angle, there are several different filtering techniques to remove signal noise and accelerations due to motions of the patient. A particularly good technique is to use the three axis gyroscopes incorporated insensors30 and34 to detect the angular rotation rate of the shank and thigh ofpatient52 when they are moving and use this data to correct the readings taken from the accelerometers. In the preferred embodiment, a Kalman filter is used to make this correction.
• The Kalman filter is an algorithm which uses a time series of measurements to estimate the next expected state of the system based on the current and previous states. It produces a statistically optimal estimate of the actual state of the system based on the measurements, even when the measurements include noise. In the case of an accelerometer and gyroscope, the accelerometer will include noise components as a result of motion, while the gyroscope will drift over time. In short, the accelerometer will give a good indication of the direction of gravity (hence the angle of the limb in question) over a long period of time, while the gyroscope will give a good indication of a change in angle over a short period of time, but will become increasingly inaccurate over longer periods of time due to drift. The Kalman filter thus uses both measurements to arrive at a good estimate of the actual orientation of the sensors.
• In the preferred embodiment, readings are taken fromsensors30 and34 every 1/10 of a second. The three acceleration measurements (X,Y and Z axes) and three gyroscope rate measurements (X,Y and Z axes) fromsensor30 are passed through the Kalman filter calculation to arrive at an estimate of the current X, Y and Z angles ofsensor30, which provides a vector representing the orientation ofsensor30 with respect to gravity. Similarly, the three acceleration measurements and three gyroscope rate measurements fromsensor34 are passed through the Kalman filter calculation to arrive at an estimate of the current X, Y and Z angles ofsensor34 with respect to gravity. As described above, the angle between the two resulting vectors is easily calculated.
• The mathematics of a Kalman filter are well known in the art. In the preferred embodiment, the Kalman filter calculation is reduced to the following:
• 
  Rate=NewRate−Bias   1)
• Where temporary value Rate is calculated as the latest gyroscope rate reading (NewRate) minus the most recently calculated Bias amount. Bias is initially set to 0 and is updated during each pass through the Kalman filter.
• 
  Angle=Angle+DeltaT×Rate   2)
• Where temporary value Angle is the previous value of Angle plus the time interval since the last reading (DeltaT) times the new Rate calculated in step 1.
• 
  P[0][0]=P[0][0]+DeltaT×(DeltaT×P[1][1]−P[0][1]−P[1][0]+Q_angle   3)
• 
  P[0][1]=P[0][1]−DeltaT×P[1][1]
• 
  P[1][0]=P[1][0]−DeltaT×P[1][1]
• 
  P[1][1]=P[1][1]+DeltaT×Q_bias
• Where P[ ][ ] is the covariance matrix, Q_angle and Q_bias are constants. This step updates the estimation error covariance.
• 
  K[0]=P[0][0]/(P[0][0]+R_Measure)   4)
• 
  K[1]=P[1][0]/(P[0][0]+R_Measure)
• Where constant R_Measure is used to update the Kalman gain matrix K.
• 
  tempAngle=newAngle−Angle   5)
• 
  Angle=Angle+K[0]×tempAngle   6)
• 
  Bias=Bias+K[1]×tempAngle   7)
• In these steps, the angle calculated during that previous pass through the Kalman filer is subtracted from the new reading of the angle from the accelerometer, newAngle to get tempAngle, the change in angle. This is adjusted by the Kalman gain K[0] calculated in the previous step to arrive at a new value of the estimated actual angle, Angle. Similarly, a new value for Bias is calculated by multiplying the Kalman gain K[1] by tempAngle.
• 
  P[0][0]=P[0][0]−K[0]×P[0][0]  8)
• 
  P[0][1]=P[0][1]−K[0]×P[0][1]
• 
  P[1][0]=P[1][0]−K[1]×P[1][0]
• 
  P[1][1]=P[1][1]−K[1]×P[1][1]
• As a final step of the Kalman filter, the values of the covariance matrix are updated based on the updated Kalman gain.
• It can been seen from the above that each of the X, Y and Z axis measurements of the inertial sensor (newAngle) can be combined with the X, Y and Z axis measurements of the gyroscope (NewRate) to arrive at a best estimate of the actual magnitude of gravitational acceleration measured by the sensors with respect to each axis. Doing this for the data read from both ofsensors30 and34 results in the two vectors from which the angle between the shank and thigh ofpatient52 can be calculated, as described above. The data comprises the number of flexions, the degree of flexion and the maximum and minimum amount of flexion of the affected joint. The maximum and minimum flex angle achieved during each flex is important for the assessment of the patient's rehabilitation because it is important to get the joint fully straight as part of the recovery process.
• This angle information read from the sensors may be presented to the caregiver in many different forms, one of which is graphically, as hereinafter described.
• Data from the two temperature sensors is also processed by database andserver20 to calculate the difference in temperature measured bytemperature sensor36 andtemperature sensor32. This difference in temperature is meaningful to the caregiver in that an elevation of the temperature measured bytemperature sensor32, which is located nearsurgical incision54, with respect to the basal temperature measured bytemperature sensor36, which is located separated from thesurgical incision54 by a great enough distance that thesensor36 will not detect an elevated temperature at the incision, may be indicative of infection ofsurgical incision54. Alternatively, a decrease in the temperature measured bytemperature sensor32 with respect to the basal temperature measured bytemperature sensor36 is a good indication that the patient is applying ice to the surgical site, which is a desirable part of the rehabilitation protocol.
• The absolute temperature measured bysensors32 and36 is also of clinical interest. A rise in basal temperature as measured bytemperature sensor36, which is removed a distance fromsurgical incision54, could indicate body heating due to exercise in the case of a small temperature rise, or a system infection causing a fever inpatient52. Similarly, a fall in the absolute temperature ofsensor32 is likely indicative of icing of the knee joint. Therefore, although there are advantages to considering the temperature differences betweensensors32 and36, either sensor can provide useful information by itself.
• The duration of temperature measured bysensor32 is of clinical value as well and is data that is collected and analyzed by the present invention. As an example, if the absolute temperature measured bysensor32 is indicative of the patient icing the joint, then determining the time that the temperature is indicative of icing allows the caregiver to know how long the patient is icing the joint.
• Data fromstretch sensor33 represents the girth of the patient's limb at the position ofstretch sensor33. In the described position, change of girth due to motion of the patient's leg (and underlying muscles) is minimized and can be averaged out over long time periods. The resulting average girth measurement is a useful indication of swelling of the patient's limb in response to infection or overuse.
• FIG. 5 shows one of many possible ways to display the data obtained from patientworn sensors10 as processed by database andserver20. In this graphical representation,vertical lines60 indicate a knee flexion. The height of the line is proportional to the degree of flexion as indicated on the vertical axis. Thus a caregiver can easily determine the degree of activity, number of times the patient has flexed their knee and by what amount.
• Line62 shows the temperature difference betweentemperature sensors32 and36. In the figure, two periods of decreased temperature would indicate to the caregiver that the patient is properly icing their knee. To the right end of the temperature curve, there is a sharp and steady rise in the temperature difference. This would indicate to the caregiver the onset of infection. In this embodiment, the temperature differential is shown, however it is clear that similarly useful information can be conveyed by showing the absolute temperature measured by either or both sensors and the duration of time either or both of thesensors32 and36 remain at a given temperature or temperature range.
• Line61 shows the time averaged girth of the patient's limb as measured bystretch sensor33. In the preferred embodiment, readings from analog todigital converter31 are taken once every second and averaged over the previous100 measurements, thus creating a time average that filters out relatively fast changes that are likely due to motion rather than swelling. As can be seen fromline61, the girth of the patient's limb (hence the amount of limb swelling) is relatively constant until it begins to rise at the right end of the graph. This would indicate to the caregiver that there is significant swelling of the limb.
• As the connection betweenremote computer24 is connected tolocal computer16 viaInternet connections22 and18 is bi-directional, it is possible for the caregiver to interact withpatient52 using email, text messaging, or video chat using any number of easily available Internet communications tools. In the preferred embodiment, this communications was facilitated using the Claris Companion Android app from Claris Healthcare Inc. (www.clariscompanion.com). The Claris Companion app was integrated with the database and server of the preferred embodiment to add additional useful information to the graphical display of data for the caregiver, as well as to provide additional useful functions. For example, the Claris Companion app is configured to allowpatient52 to voluntarily provide a “pain score” from 1-10, where 1 is no pain at all and 10 is excruciating. Pain scores64 are displayed along the time axis inFIG. 5 so that the caregiver can correlate the pain score with activity or temperatures. In addition, the Claris Companion app is configured to report whenever the patient chooses to take pain medication, as indicated bymarks66 inFIG. 5.
• In addition to the manual communication between the caregiver andpatient52 made possible by the present invention, the preferred embodiment provides automated coaching and encouragement topatient52 vialocal computer16. For example, the caregiver can set goals forpatient52 such as completing25 repetitions of a knee flex beyond 80 degrees. When database andserver20 calculates that the target repetitions are completed by analyzing the data from patientworn sensors10, it causeslocal computer16 to display a congratulatory message. Similarly, analysis of the temperature data from patientworn sensors10 can causelocal computer16 to show a confirmation message when patient52 successfully lowers the temperature ofsurgical incision54 by a desired amount, and can then start an on-screen timer to indicate how long the lowered temperature should be maintained. Further automated or manual coaching and encouragement can be provided in the form of instructional videos or photographs, encouraging messages, social interaction with similar patients, and ‘gamification’ in the form of goals, rewards and progress reporting.
• FIG. 6 illustrates an alternative sensing means for determining the degree of flexion of the knee ofpatient52. In this embodiment,distal motion sensor30 is replaced withcapacitive flex sensor70, which in the preferred embodiment is a Soft Silicon Bend Sensor (bendlabs.com) that provides a signal proportional to the angle of flexion ofsensor70.Flex sensor70 is an elongate strip attached to the leg ofpatient52 so that the strip extends above, over and below the kneejoint using anchor72 and the case that enclosessensors36 and34. Assensor70 provides a signal directly proportional to the degree of flexion of the knee ofpatient52, there is no need to perform mathematical calculations to determine the flexion angle. Although no longer used in the calculation of the flexion angle,motion sensor34 is retained in order to allow the orientation of the thigh ofpatient52 to be measured. Knowing this orientation allows a caregiver to determine the body position ofpatient52 while they are flexing their knee. For example, shouldmotion sensor34 indicate that the thigh ofpatient52 is horizontal while the knee is flexed, it would indicate thatpatient52 is performing the exercise while sitting, while ifmotion sensor34 indicates that the thigh ofpatient52 is vertical, it would indicate that the exercise is being performed while standing. Thus, data corresponding to the orientation of the limb that is proximate to the joint relative to a ground plane (i.e., a horizontal reference plane) is an effective in monitoring rehabilitation therapy.
• FIG. 7 illustrates another advantageous way to display the data obtained from patientworn sensors10 as processed by database andserver20. This embodiment anticipates that a caregiver may be responsible for several patients simultaneously, thus is it helpful to be able to determine the status of selected and predetermined patient conditions for several patients in summary form without having to examine and interpret the detailed data as presented in the embodiment shown inFIG. 5. In the alternative embodiment shown inFIG. 7, each patient is represented by a row of information icons, includingpatient photo80,patient name82,Activity icon84,Compliance icon86 andWellness icon88. Each ofActivity icon84,Compliance icon86 andWellness icon88 are labelled with a number representing, in the preferred embodiment, a score ranging from 0 to 100, calculated as hereinafter described. A further enhancement to rapid understanding of the relative score for each factor, not illustrated inFIG. 7, is that each ofActivity icon84,Compliance icon86 andWellness icon88 display as a different color depending on the score: if the score is 33 or below, the icon is colored red; if between 34 and 66, the icon is colored yellow; while if the score is above 66, the icon is colored green. Thus a caregiver can quickly determine which of the patients in his or her charge are doing well and which are doing poorly.
• The Activity score is calculated by combining data from several sources associated with the patient. One source of data is patientworn sensors10. Data fromsensors10 are used to determine how often the patient has flexed their knee during the day and comparing that to a pre-determined ideal activity level for that time of day, or to previous days for the same patient, to arrive at a value representing a percentage of normal activity. An additional score is calculated based on the number of times the patient has interacted withlocal computer device16, again compared to a pre-determined ideal activity level to arrive at a percentage of normal interaction frequency. Finally, data from patientworn sensors10 is used to determine the total period of time the patient has been moving and the total period of time the patient has been stationary, and to calculate a third activity score representing the percentage of time the patient is moving. The three scores are weighted and combined into a total activity score; in the preferred embodiment, the score representing knee flexes is weighted at 70%, the interactions withlocal computer device16 are weighted at 20% and the percentage of time moving is weighted at 10%, thereby deriving an activity score between 0 and 100. Each time the Activity score is calculated, the number displayed inActivity icon84 is updated,Activity icon84 is moved up or down in the display area, and the color of the icon is changed if merited by the score.
• The Compliance score is calculated by determining what percentage of prescribed activities the patient has completed, and how quickly they have completed them. For example, during the course of a day, the patient may be reminded to ice their knee five times at two hour intervals, be asked (via local computer device16) to report their current pain scale, reminded to perform certain exercises, or be reminded to take medications. Compliance with these prescribed activities is determined from two sources—patientworn sensors10 and the patient's interaction withlocal computer device16. For example,sensors10 will detect a decrease in temperature if and when the patient responds to a reminder to ice their knee, which would indicate compliance. Similarly,sensors10 will detect appropriate motions if and when the patient complies with a reminder to do exercises. For other reminders, such as reporting pain scale or taking medications, the patient's response vialocal computer device16 will indicate compliance, or the patient's election to skip the activity reminded. The compliance score displayed byCompliance icon86 is calculated as follows: each time an activity reminder is displayed onlocal computer device16, that activity is given a score of 100. For each pre-determined time period that elapses without the activity being completed, typically 5 minutes, one point is deducted from the score for this activity. Once the activity is completed, the resulting discounted score is recorded. At periodic intervals, the average score for all activities completed so far during the day is calculated. This becomes the revised Activity score, which is used to update the number appearing inActivity icon86 and to adjust the position and color ofActivity icon86 on the caregiver's display.
• The Wellness score is similarly calculated by combining data from patientworn sensors10 and the patient's interaction withlocal computer device16. In the case of wellness, the patient's response to a pain scale self-evaluation is given a high weighting. At regular intervals during their recovery,local computing device16 displays a message asking the patient to rate their pain on a scale of 1 to 10, where 1 is very little pain and 10 is excruciating to thereby generate a pain score. This scale is inverted to calculate wellness, wherein a pain scale of 1 becomes a pain score of 100 and a pain score of 10 becomes a pain score of 0. Added to this is a temperature score, wherein, if the temperature recorded bysensors10 is higher than a pre-determined normal temperature, the temperature score declines from 100 to 0 in proportion to the increased temperature. Finally, a swelling score is calculated in which the current girth of the limb is compared to the initial girth of the limb to determine a degree of swelling. If the measured girth is greater than the initial girth, the swelling score is discounted from 100 to 0 in proportion to the increase in girth. At periodic intervals, the total Wellness score is calculated by weighting the pain score at 80%, the temperature score at 10% and the swelling score at 10% to arrive at a total Wellness score ranging from 0 to 100. This becomes the revised Wellness score, which is used to update the number appearing inWellness icon88 and to adjust the position and color ofWellness icon88 on the caregiver's display.
• It will be evident based on the foregoing description and from the drawings that the present invention defines a method monitoring a patient and assessing the patient's condition by attaching a sensor to the patient for measuring a defined patient parameter, recording patient parameter data from the sensor, providing an input device to the patient and prompting the patient to enter patient input data into the input device, recording the patient input data; and assigning a score that is indicative of the patient parameter by combining the first patient parameter data and patient input data. As detailed, plural sensors may be used for monitoring different patient parameters such as flexion, temperature and swelling and the method is useful for generating status scores that quickly provide a caregiver with a way to assess the patient's condition in relation to the different patient parameters, including Activity, Compliance and Wellness. The methods further enable a patient's compliance with a rehabilitation regimen that has been prescribed for the patient, including prompting the patient to enter patient input data into the input device, the patient input data correlating to a desired patient condition, recording the time of the prompt, recording the patient input data entered by the patient and the time that the patient input data are entered by the patient, determining the time between when the patient is prompted and the time when the patient enters the patient input data, and generating a patient compliance score from the time between when the patient is prompted when the patient enters the patient input data, the patient compliance score providing a measure of the patient's compliance with a treatment regimen for the desired patient condition.
• Many variations on the preferred embodiment described here can be easily imagined. For example, although the invention as herein described is shown as used for a knee joint, it can easily be extended to operate in a similar fashion for any other joint on which surgery may be performed. The sensors described are one choice of many possibilities for measuring joint motion and temperature, and the choice of a data logger with local memory and periodic uploading could be eliminated in favour of real-time transfer of data fromsensors10 tolocal computer16. Furthermore, it is possible to eliminate the cable and I2C bus by having each sensor connected to a separate Bluetooth radio linked to the local computer. It is also clear that there are other mathematical techniques for filtering data from accelerometers and gyroscopes to improve their accuracy and extracting the angle betweensensors30 and34, many of which could provide equally useful measurements.
• Based on review of the foregoing disclosure and the drawings, it should also be evident to one skilled in the art that there are many equivalent ways to calculate Activity, Compliance and Wellness scores, or other scores of interest to caregivers, and to display the value of such scores graphically to allow a caregiver to quickly evaluate a number of patients. Further, the weighting given to the various data sources used to determine such scores can obviously be changed to reflect the relative importance of each data source with respect to the patient's condition.
• While the present invention has been described in terms of preferred and illustrated embodiments, it will be appreciated by those of ordinary skill that the spirit and scope of the invention is not limited to those embodiments, but extend to the various modifications and equivalents as defined in the appended claims.

Claims (21)

1. A method monitoring a patient, comprising the steps of:

a) attaching a sensor to the patient for measuring a patient parameter;

b) recording patient parameter data from the sensor;

c) providing an input device to the patient and prompting the patient to enter patient input data into the input device;

d) recording the patient input data; and

e) assigning a score that is indicative of a patient condition by combining the first patient parameter data and patient input data.

2. The method according toclaim 1 including the steps of

a) attaching plural sensors to the patient for measuring plural patient parameters;

b) recording patient parameter data from each of the sensors;

c) prompting the patient to enter patient input data;

d) recording the patient input data and

e) assigning a score to each of the patient parameters, each score determined by combining the patient parameter data and patient input data.

3. The method according toclaim 2 including the step of prompting the patient to enter patient input data correlating to a level of pain.

4. The method according toclaim 2 wherein one of the sensors is a motion sensor and the patient parameter relates to the number of times the patient has flexed a joint.

5. The method according toclaim 2 wherein one of the sensors is a temperature sensor and the patient parameter relates to the temperature of the patient at a location of interest.

6. The method according toclaim 2 wherein one of the sensors measures the girth of the patient's limb and the patient parameter relates to tissue swelling.

7. The method according toclaim 4 in which one of the scores is a patient activity score that is determined according to the following steps

a) comparing the amount of motion detected by the motion sensor with a predetermined amount of motion to thereby determine a first activity score;

b) comparing the number of times the patient has entered patient input data into the input device with a predetermined number to thereby determine a second activity score.

8. The method according toclaim 7 including the steps of:

a) with patient parameter data, determining a third activity score representing the percentage of time that the patient is moving; and

b) combining the first, second and third activity scores.

9. The method according toclaim 8 in which the first, second and third activity scores are weighted before they are combined to generate a total activity score that is between 0 and 100.

10. The method according toclaim 5 in which one of the scores is a patient wellness score that is determined according to the following steps:

a) prompting the patient to enter patient input data correlating to the patient's perceived level of pain and from the patient input data generating a pain score;

b) comparing the patient parameter data derived from the temperature sensor with a predetermined temperature to generate a temperature score; and

c) combining the pain score and temperature score to generate a wellness score.

11. The method according toclaim 2 including displaying each of the plural scores on display screen for a caregiver.

12. A method for monitoring a patient, the method comprising the steps of:

a) attaching one or more sensors to the patient, each sensor adapted for measuring a different patient condition;

b) recording patient condition data from each sensor;

c) providing an input device to the patient and prompting the patient to enter patient input data into the input device;

d) recording the patient input data; and

e) for each patient condition, calculating a status score that is indicative of patient condition.

13. The method according toclaim 12 wherein one of the sensors is a motion sensor and a first patient condition correlates to patient joint flexion at a desired joint.

14. The method according toclaim 13 wherein one of the sensors is a temperature sensor and a second patient condition correlates to the patient's temperature at a desired location proximate the joint.

15. The method according toclaim 14 wherein one of the sensors is a limb girth sensor and a third patient condition correlates to swelling of the patient's limb proximate the joint.

16. The method according toclaim 15 including the step of generating a status score for the first patient condition, the second patient condition and the third patient condition and displaying the status score for each patient condition on a caregiver's display screen.

17. A method for monitoring a patient, the method comprising the steps of:

a) providing an input device to the patient;

b) prompting the patient to enter patient input data into the input device, the patient input data correlating to a desired patient condition;

c) recording the time of the prompt;

d) recording the patient input data entered by the patient and the time that the patient input data are entered by the patient;

e) determining the time between when the patient is prompted and the time when the patient enters the patient input data; and

f) generating a patient compliance score from the time between when the patient is prompted when the patient enters the patient input data, the patient compliance score providing a measure of the patient's compliance with a treatment regimen for the desired patient condition.

18. The method according toclaim 17 wherein the step of generating a patient compliance score includes summing the time between when the patient is prompted when and when the patient enters the patient input data.

19. The method according toclaim 18 in which the patient input data correlating to a desired patient condition is assigned a predetermined weighting value and the predetermined weighting value is multiplied by the time between when the patient is prompted when and when the patient enters the patient input data before the summing step.

20. The method according toclaim 19 further comprising the step of attaching first and second motion sensors to the patient, the first sensor attached to the patient's limb proximate to a joint and the second sensor attached to the patient's limb distal to the joint, the sensors adapted for measuring a flexion of the joint.

21. The method according toclaim 20 and the method further comprises the steps of:

a) with the sensors, measuring the angle between the patient's limb proximate to a joint and distal to the joint;

b) with the sensors, detecting the number of times the patient has flexed the joint; and

c) for each flexion of the joint, detecting the maximum and minimum flex angle of the joint by determining the angular orientation between the limb distal to the joint relative to the limb proximate to the joint during flexion.

Images