This patent application is related to the following utility patent applications, each of which is incorporated herein, in its entirety, by reference:
Both of the above listed patents are incorporated herein, in their entireties, by reference.
The invention generally relates to dermatological ulcers on living beings and, more particularly, the invention relates to evaluating portions of living beings for dermatological ulcers.
Open sores on an external surface of the body often form septic breeding grounds for infection, which can lead to serious health complications. For example, foot ulcers on the bottom of a diabetic's foot can lead to gangrene, leg amputation, or, in extreme cases, death. The healthcare establishment therefore recommends monitoring the foot of a diabetic on a regular basis to avoid these and other dangerous consequences. Unfortunately, known techniques for monitoring foot ulcers, among other types of ulcers, often are inconvenient to use, unreliable, or inaccurate, thus reducing compliance by the very patient populations that need it the most.
In accordance with one embodiment of the invention, a method and apparatus for evaluating foot inflammation each uses at least one temperature detection modality to form a first thermogram and a second thermogram of the sole of at least one foot. Each thermogram forms a substantially continuous set of two-dimensional temperature values across the sole of the (at least one) foot. The thermograms have features; namely, the first thermogram has first features and the second thermogram has second features. The method and apparatus thus control a device to apply at least one transformation (e.g., an affine transformation, non-affine transformation, or a combination) to one or both of the first and second thermograms to align the first features with corresponding second features, and determine, at any thermogram location, if at least one of the thermograms presents one of a plurality of patterns indicative of inflammation. Finally, the method and apparatus each produce output information indicating the result of the determination of whether the thermograms present one of the plurality of patterns.
The at least one affine transformation may include, among other things, at least one of reflection, rotation, scaling and translation. The affine transformation preferably aligns the first features and the second features to a common coordinate system. The at least one affine transformation also may be applied to a set of points corresponding to a) foot temperature, b) a grid corresponding to foot temperature, or 3) a set of equations corresponding to foot temperature. In addition to applying the affine transformation(s), some implementations apply at least one non-affine transformation to the first and second thermograms.
The two thermograms may apply to a single foot, or both feet. Thus, the first thermogram may represent the sole a left foot of a given person, and the second thermogram represents the sole of a right foot of the given person. In that case, the method and apparatus each may use the modality to obtain temperatures across the sole of the left foot at a first time, and obtain temperatures across the sole of the right foot at a second time. The first time and second time may be different times. Alternatively, the first thermogram and the second thermogram may represent the sole of the same foot of a given person. In that case, the data used to form the first and second thermograms can be obtained at different times, or at substantially the same time.
The at least one temperature detection modality may include a thermal camera. In that case, a person may hold the thermal camera in an unconstrained manner in at least three degrees of freedom in free space when the thermal camera obtains temperature data of the sole of the at least one foot. For example, when the person is holding the camera, the camera is free to move in space (relative to the sole of the at least one foot) while the person holds the thermal camera and obtains the temperature data. The at least three degrees of freedom may include at least three of translational movement in the X-axis, the Y-axis, and the Z-axis of the Cartesian Coordinate System, and rotation about the X-axis, the Y-axis, and the Z-axis of the Cartesian Coordinate System. Other temperature detection modalities may include an insole in which the foot is positioned, and an open platform having a substrate for receiving the at least one foot, and a plurality of temperature sensors that are stationary relative to the substrate. Alternative embodiments may vary the position of the temperature sensors relative to the substrate.
The method and apparatus each may control a device to orient the first thermogram and the second thermogram to a common coordinate system by changing the orientation of at least one of the first and second thermograms for roll (rotation about the X-axis), pitch (rotation about the Y-axis), yaw (rotation about the Z-axis), X-axis translation, Y-axis translation, and Z-axis translation. Moreover, the at least one temperature detection modality may obtain a plurality of discrete temperature values of the sole of the at least one foot, and calculate temperatures between a plurality of adjacent discrete temperature values to form the thermograms of the sole of each of the at least one foot.
Some embodiments control the device to orient by retrieving the first thermogram from memory, and using the orientation of the first thermogram to orient the second thermogram. To improve accuracy in some instances, the apparatus and method each may normalize the amplitude of the two-dimensional array of temperature values of the first and second thermograms against a common value.
In accordance with another embodiment, a system for evaluating foot inflammation has a thermogram generator configured to form a first two-dimensional thermogram and a second two-dimensional thermogram of the sole of the at least one foot. Each thermogram forms a substantially continuous set of two-dimensional temperature values across the sole of the at least one foot. Moreover, the first thermogram and second thermogram have respective first and second features. The apparatus also has an orientation module operatively coupled with the thermogram generator and configured to apply at least one affine transformation to the first and second thermograms to align the first features of first thermogram with corresponding second features of the second thermogram. The apparatus further has a pattern recognition system operatively coupled with the orientation module and configured to determine, at any location within the first thermogram and the second thermogram, if the thermograms present one of a plurality of patterns indicative of inflammation. Finally, the apparatus has an analyzer operatively coupled with the pattern recognition system and configured to produce output information indicating the result of the determination of whether the thermograms present one of the plurality of patterns.
Illustrative embodiments of the invention are implemented as a computer program product having a computer usable medium with computer readable program code thereon. The computer readable code may be read and utilized by a computer system in accordance with conventional processes.
Those skilled in the art should more fully appreciate advantages of various embodiments of the invention from the following “Description of Illustrative Embodiments,” discussed with reference to the drawings summarized immediately below.
In illustrative embodiments, an apparatus analyzes a patient's foot to determine the risk of an ulcer emerging on its underside (i.e., on its sole). This permits patients their healthcare providers and/or their caregivers to intervene earlier, reducing the risk of more serious complications. To that end, a modality detects foot temperatures to generate two or more thermograms. The apparatus then applies a transformation to the thermograms, normalizing/registering the thermograms so that they comply with a standard coordinate system. If the transformed thermogram presents at least one of a number of prescribed patterns, then various embodiments produce output information indicating the risk of an ulcer emerging on the patient's foot. Details of illustrative embodiments are discussed below.
It should be noted that discussion of diabetes and diabetics is but one example used simply for illustrative purposes only. Accordingly, various embodiments apply to other types of diseases (e.g., stroke, deconditioning, sepsis, friction, coma, etc.) and other types of ulcers-such embodiments may apply generally where there is a compression or friction on the living being's body over an extended period of time. For example, various embodiments also apply to ulcers formed on different parts of the body, such as on the back (e.g., bedsores), inside of prosthetic sockets, or on the buttocks (e.g., a patient in a wheel chair). Moreover, illustrative embodiments apply to other types of living beings beyond human beings, such as other mammals (e.g., horses or dogs). Accordingly, discussion of diabetic human patients having foot ulcers 12 is for simplicity only and not intended to limit all embodiments of the invention.
Many prior art ulcer detection technologies known to the inventors suffered from one significant problem-patient compliance. If a diseased or susceptible patient does not regularly check his/her feet 10, then that person may not learn of an ulcer 12 or a pre-ulcer 14 until it has emerged through the skin and/or requires significant medical treatment. Accordingly, illustrative embodiments implement an ulcer monitoring system in any of a variety of forms and modalities-preferably in an easy to use form factor that facilitates and encourages regular use.
To monitor the health of the patient's foot (discussed in greater detail below), illustrative embodiments use any of a variety of modalities to gather temperature data about a plurality of different locations on the sole of the patient's foot 10. This temperature data provides the core information ultimately used to determine the health of the foot 10. To that end,
As known by those in the art, rather than using visible light, a thermal camera (also known as a “thermographic camera,” “thermal imaging camera,” or an “infrared camera”) forms an image of an object using infrared radiation. More specifically, a thermal camera captures the heat signature of an object (e.g., the sole of a foot) in electronic form, effectively determining the temperature across the two-dimensional sole of the foot. Among other things, the thermal camera 17 can be portable/hand-held, as shown in
The person may hold the thermal camera 17 in a fully-constrained manner, a partially-constrained manner, or in an unconstrained manner (i.e., as in
When fully-constrained, the thermal camera 17 is substantially immovable relative to the patient's foot. As such, the thermal camera 17 is not movable along or about the noted axes.
When partially-constrained, the thermal camera 17 is substantially unmovable in one or more ways, but still movable in at least one other way. For example, the thermal camera 17 may be set against a flat surface and thus, be substantially stable along the Y-axis (if defined as normal to the ground). Despite this stability, the thermal camera 17 may be movable in other ways, such as translatable along the Z-axis and the X-axis. These and other freedoms of movement can lead to analysis complications when evaluating the foot. Illustrative embodiments aim to mitigate these complications.
A bathroom mat or rug are but two of a wide variety of different potential open platform form factors. Others may include a platform 16 resembling a scale, a stand, a footrest, a console, a tile built into the floor, or a more portable mechanism that receives at least one of the feet 10. The implementation shown in
The open platform 16 (and other modalities, such as the thermal camera modality) also has some indicia or display 18 on its top surface they can have any of a number of functions. For example, the indicia can turn a different color or sound an alarm after the readings are complete, show the progression of the process, or display results of the process. Of course, the indicia or display 18 can be at any location other than on the top surface of the open platform 16, such as on the side, or a separate component that communicates with the open platform 16. In fact, in addition to, or instead of, using visual or audible indicia, the platform 16 may have other types of indicia, such as tactile indicia/feedback, our thermal indicia.
Rather than using an open platform 16, alternative embodiments may be implemented as a closed platform, such as an insole, a shoe, or a sock that can be regularly worn by a patient, or worn on an as-needed basis. For example, the insole of the patient's shoe or boot may have the functionality for detecting the emergence of a pre-ulcer 14 or ulcer 12, and/or monitoring a pre-ulcer 14 or ulcer 12. The open platform 16 and thermal camera 17 modalities are discussed in greater detail in its parent patent application, U.S. application Ser. No. 13/799,828, which already was incorporated by reference.
Although it gathers temperature and other data about the patient's foot, illustrative embodiments may locate additional logic for monitoring foot health at another location. For example, such additional logic may be on a remote computing device. To that and other ends,
As noted above and discussed in greater detail below with regard to
This output message may have the output information in its relatively raw form for further processing. Alternatively, this output message may have the output information formatted in a high-level manner for easy review by automated logic or a person viewing the data. Among other things, the output message may indicate the actual emergence of an ulcer 12 or a pre-ulcer 14, the risk of the emergence of an ulcer 12 or a pre-ulcer 14, or simply that the foot 10 is healthy and has no risks of ulcer 12 or pre-ulcer 14. In addition, this output message also may have information that helps an end-user or healthcare provider 48 monitor an ulcer 12 or pre-ulcer 14.
Using a distributed processing arrangement like that shown in
Some embodiments may distribute and physically position the functional components in a different manner. For example, the platform (e.g., the thermal camera 17) may have the analysis engine 46 on its local motherboard. In fact, some embodiments provide the functionality entirely on the modality, such as on the open platform and/or within other components in the local vicinity of the platform 16. For example, all of those functional elements (e.g., the analysis engine 46 and other functional elements) may be within a housing that also contains the thermal camera 17. Accordingly, discussion of a distributed platform is but one of a number of embodiments that can be adapted for a specific application or use.
Those skilled in the art can perform the functions of the analysis engine 46 (and the other functional modules) using any of a number of different hardware, software, firmware, or other non-known technologies.
In summary, the analysis engine 46 of
The analysis engine 46 also has an orientation module 67 configured to apply at least one transformation to a thermogram—preferably to align the features of different thermograms—and an analyzer 70 configured to produce the above noted output information, which indicates any of a number of different conditions of the foot 10. For example, the output information may indicate the risk that an ulcer 12 will emerge, the emergence of a pre-ulcer 14 (i.e., the first indication of a pre-ulcer 14), the progression of a known ulcer 12, or the emergence of a new ulcer 12 (i.e., the first indication of any given ulcer 12 to the patient and associated support). Communicating through some interconnect mechanism, such as a bus 72 or network connection, these modules cooperate to determine the status of the foot 10, which may be transmitted or forwarded through an input/output port 74 that communicates with the prior noted parties across the larger data network 44.
As noted above, some or all of these modules may be implemented in hardware, software, firmware, or a combination of hardware and software. For example, some modules may be configured across several integrated circuits (e.g., microprocessors or application specific integrated circuits) on one or more printed circuit boards. Those skilled in the art may select the implementation based on the requirements of their given situation (e.g., availability of resources, additional functions, current technology, etc.).
The process begins at step 700, in which the temperature gathering modality obtains two foot temperature readings. Specifically, the modality obtains foot temperature information in two different actions. For example, the thermal camera 17 may take a first thermal image of the right foot, and a second thermal image of the left foot. As another example, the thermal camera 17 may take a first thermal image of both feet, and a second thermal image of both feet at a different time, such as the next day. As yet another example, a first thermal camera 17 may take the first thermal image, while a second thermal camera 17 may take the second thermal image. In still a fourth example, the thermal camera 17 may take the first thermal image, while the open platform 16 (i.e., a second modality) may take the second thermal image. These two thermal images may be taken at the same time, or at different times (e.g., seconds, minutes, hours, or days apart). This is in contrast to embodiments that may take thermal images of two feet in a single action (e.g., taking the thermal image of two feet at the same time in the same action with a single thermal camera 17).
This step therefore produces a matrix of discrete temperature values across the foot 10 or feet 10. For example, these discrete temperature values may be in the form of discrete pixels of a thermographic image obtained with the thermal camera 17. This temperature data also may have additional meta-data, such as the date and time of obtaining this temperature data.
Accordingly, based on the matrix of discrete temperature values, the temperature detection modality, or other functional module, forms a first thermogram and a separate second thermogram of the sole of at least one foot. For example, the first thermogram may represent the left foot, while the second thermogram may represent the right foot. Two separate actions thus were taken to obtain each of the data values used to form the thermograms. As another example, the first thermogram may represent the left foot on a given day, while the second thermogram may represent the left foot the next day. Each of these thermograms has relevant features, such as an outline, a shape, temperature information, prominent anatomical shapes, etc., that ultimately will be used to orient the thermograms (discussed below).
In simple terms, as known by those in the art, a thermogram is a data record made by a thermograph, or a visual display of that data record. A thermograph simply is an instrument that records temperatures (i.e., the platform 16). As applied to illustrative embodiments, a thermograph measures temperatures and generates a thermogram, which is data, or a visual representation of that data, of the spatially-continuous two-dimensional temperature data across some physical region, such as a foot 10. Accordingly, unlike an isothermal representation of temperature data, a thermogram provides a complete, continuous data set/map of the temperatures across an entire two-dimensional region/geography. More specifically, in various embodiments, a thermogram shows (within accepted tolerances) substantially complete and continuous two-dimensional spatial temperature variations and gradients across portions of the sole of (at least) a single foot 10, or across the entire sole of the single foot 10.
Those skilled in the art may form the thermogram in a variety of different manners. For example, the thermogram may be formed by calculating temperature values between some or all of the plurality of discrete temperature values of the foot retrieved by the modality. Among other things, these intermediate temperature values may be calculated using interpolation techniques. Reference is made to the above noted incorporated parent application for some thermogram generation examples, which involves interpolation, re-orienting, and adjusting the baseline temperature. That example may be used separately at least in part with illustrative embodiments discussed below with regard to
At this point, the process is considered to have formed the thermogram, which may be stored in memory 76. For effective use, however, it nevertheless still may require further processing. Accordingly, at step 704, the orientation module 67 applies one or more transformations to the two thermograms, thus normalizing/registering the thermograms to a standard coordinate system. Some embodiments may apply the transformations/normalize as the modality collects the data, while other embodiments, such as the one shown in
The position and orientation of the foot 10 on the platform 16 therefore is important when performing this step. For example, when using the open platform 16, to determine the position and orientation of the foot 10, the analysis engine 46 and its thermogram generator 66 simply may contrast the regions of elevated temperature on the platform 16 (i.e., due to foot contact) with those at ambient temperature. Other embodiments may use pressure sensors to form a pressure map of the foot 10.
Some embodiments may further modify the thermogram to better contrast warmer portions of the foot 10 against other portions of the foot 10.
For example, if the deviation is negative, the thermogram may have some shades of blue, with a visual scale of faint blues being smaller deviations and richer blues being larger deviations. In a similar manner, positive deviations may be represented by some shades of red, with a visual scale of faint red being smaller deviations and richer reds being larger deviations. Accordingly, and this example, bright red portions of the thermogram readily show hotspots that may require immediate attention. Of course, other embodiments may use other colors or techniques for showing hotspots. Accordingly, discussion of color coding or specific colors is not intended to limit all embodiments.
Briefly moving away from the discussion of
Before beginning this process, the orientation module 67 receives one or both of the thermograms (e.g., from memory 76 or other means). Indeed, the thermograms can be applied to the standard coordinate system, or the standard coordinate system can be applied to the thermograms. In the latter case, some embodiments may orient the first thermogram to a standard coordinate system, and then coordinate the second thermogram to the first thermogram. In either case, the two thermograms are oriented to effectively and efficiently perform the process of
The normalization process begins at step 800, which removes background information from both thermograms, leaving a respective single foot for each thermogram. For example, background radiation can be removed using a graph-partitioning method, which examines homogenous regions of the thermogram (in terms of temperature) and segments the regions to minimize the gradient across the segments in the thermogram. Alternatively, some embodiments may use simpler histogram or thresholding techniques, where the background is assumed to have a uniformly lower value than the region of interest (e.g., the feet).
Next, step 802 identifies the foot, such as by forming an outline around the perimeter of the foot in each thermogram. This outline can substantially exactly track the perimeter of the foot, or be in the form of a rectangle about the outline of the foot. Some embodiments may search for fully-enclosed regions with similar principal characteristics (e.g., length, width, or area ratio) to a foot. Alternatively, the normalization process may search the thermogram for a thermometric template of the foot, either generated for a generic subject or using previously-collected data for a specific subject. Among other ways, this search can use optimization techniques to maximize the favorability of the fit by applying affine or non-affine transformations to the template or thermogram.
The process continues by applying one or more appropriate transformations to the thermograms. In this case, as noted in step 804, the process applies one or more affine transformations to each thermogram. In general, as known by those in the art, an affine transformation generally preserves co-linearity (i.e., all points lying on a line initially still lie on a line after transformation) and ratios of distances (e.g., the midpoint of a line segment remains the midpoint after transformation). Geometric contraction, expansion, dilation, reflection, rotation, shear, scaling, similarity transformations, spiral similarities, and translation all may be considered to be affine transformations, as are their combinations. More generally, an affine transformation is a composition of rotations, translations, dilations, and shears.
Illustrative embodiments rotate and/or translate the thermograms, as needed, to a standard coordinate system defined by the principal axis of the foot. Such a technique registers features of the thermogram with the standard coordinate system. For example, one embodiment may register/orient the first thermogram to the standard coordinate system, and then simply register/orient the second thermogram to the first thermogram (effectively registering them to the same standard coordinate system). When registering, the system thus may cause the relevant thermogram to translate and rotate in one or more of pitch, roll, and yaw.
In addition to rotating and translating, step 804 also may mirror and/or align all or part of the thermograms. For example, illustrative embodiments may mirror a left foot by simply rotating its thermogram 180 degrees along its major axis so that it can be aligned with the right foot. Rather than mirroring the entire thermogram, however, some embodiments may mirror only corresponding portions of the thermograms, such as portions known to be most prone to inflammation.
Illustrative embodiments mirror the thermograms as appropriate when comparing the left foot to the right foot. Accordingly, in such cases, step 804 does not mirror a single foot over time—it is unnecessary. To that end, the process may rotate one of the transformations, and then align some or all common portions together. For example, step 804 may rotate the first thermogram, and then align the heels of both thermograms together to align other corresponding portions of the sole. It also should be noted that step 804 may align thermograms whether comparing left and right foot thermograms, or when comparing thermograms of the same foot over time.
Alternative embodiments may omit the affine transformations of step 804.
Step 806 then determines if more transformations are necessary. If so, then the process may continue to step 808, which can apply non-affine transformations to one or both of the thermograms. As such, these transformations generally do not preserve thermogram co-linearity (i.e., all points lying on a line initially still lie on a line after transformation) and ratios of distances. For example, some embodiments may dimensionally stretch, deform, represent a three-dimensional space in two-dimensions, or otherwise modify one or both thermograms in a corresponding manner. Among other ways, some embodiments may use non-affine transformations in a series to approximate a single affine transformation. In that latter case, some embodiments may skip the affine transformations.
The process then may dimensionally scale one or both thermograms (step 810). For example, both feet may not be the same size or shape, or the thermograms of the feet may not be the same size or shape. This may become an issue with the thermal camera 17, in which its distance from the sole and its rotation relative to the foot varies. This step thus may map both thermograms to a common shape, such as the shape of a foot, or even a shape that does not resemble a foot. For example, step 810 may stretch and compress the thermogram to the shape of a circle. This step preferably is executed internally to the orientation module 67 and thus, not displayed on a display device. Other steps, however, may display the thermograms as they are processed.
The process may conclude by normalizing the amplitude of the temperature signal across the entire thermogram. This may be important when using a closed platform having an elevated temperature (e.g., a shoe after exercise). Illustrative embodiments may normalize the amplitude signal in a number of manners, such as by subtracting the temperatures across two thermograms, or determining the temperature based on some prescribed temperature. Among other things, the prescribed temperature may include the mean temperature across the thermogram, the median temperature across the thermogram, or the background temperature. Continuing with the above example, when using the mean temperature, the thermogram may show that a local temperature is 1 degree C. above the mean temperature.
The amplitude also may be normalized over time to remove extraneous trends or correct for harmonic fluctuations due to time-of-day or time-of-month, or to eliminate or remove unwanted artifacts in the signal due to exogenous factors, such as the patient's activity or basal temperature.
Now that the thermogram generator 66 has generated the two normalized thermograms, the process returns to
A prescribed pattern may include a temperature differential over some geography or portion of the foot 10 or feet 10. To that end, various embodiments contemplate different patterns that compare at least a portion of the foot 10 against other foot data. Among other things, those comparisons may include the following:
As an example of the first comparison, the pattern may show a certain region of a foot 10 has a temperature that is 4 F higher than the temperature at that same region several days earlier.
As an example of the second comparison, the pattern may show that the corresponding portions of the patient's feet 10 have a temperature differential that is 4 degrees F.
As an example of the third comparison, the pattern may show localized hotspots and peaks within an otherwise normal foot 10. These peaks may be an indication of pre-ulcer 14 or ulcer 12 emergence, or increased risk of the same, which, like the other examples, alerts caregiver and patient to the need for more vigilance.
Accordingly, if no pattern indicative of relevant inflammation is detected, then the output produces a negative reading or message (step 710), indicating no or minimal risk. Conversely, if such a pattern is detected, then the process may conclude at step 712, producing an output reading indicating a risk of ulceration or pre-ulcer (or similar indication). The output reading may include the risk of an ulcer 12 emerging anywhere on the foot 10, or at a particular location on the foot 10. This risk may be identified on a scale from no risk to maximum risk. Indeed, some embodiments include evaluation of inflammation at various stages, from no inflammation, to pre-ulcer, to full ulcer. See the incorporated patent applications for some examples of such stages.
Of course, various embodiments may make similar comparisons while analyzing the thermograms for additional patterns. For example, similar to the third comparison, the pattern recognition system 68 may have a running average of the temperature of the geography of the entire foot 10 over time. For any particular spot on the foot 10, this running average may fall within a normal range between a high temperature and a low temperature for that set of thermograms over a period of time. Accordingly, data indicating that the temperature at that given spot is outside of the normal range may be predictive of a pre-ulcer 14 or an ulcer 12 at that location.
Some embodiments may use machine learning and advanced filtering techniques to ascertain risks and predictions, and to make the comparisons. More specifically, advanced statistical models may be applied to estimate the current status and health of the patient's feet 10, and to make predictions about future changes in foot health. State estimation models, such as a switching Kalman filters, can process data as they become available and update their estimate of the current status of the user's feet 10 in real-time. The statistical models can combine both expert knowledge based on clinical experience, and published research (e.g., specifying which variables and factors should be included in the models) with real data gathered and analyzed from users. This permits models to be trained and optimized based on a variety of performance measures.
Models can be continually improved as additional data is gathered, and updated to reflect state-of-the-art clinical research. The models also can be designed to take into account a variety of potentially confounding factors, such as physical activity (e.g., running), environmental conditions (e.g., a cold floor), personal baselines, past injuries, predisposition to developing problems, and problems developing in other regions (e.g., a rise in temperature recorded by a sensor 26 may be due to an ulcer 12 developing in a neighboring region measured by a different sensor). In addition to using these models for delivering real-time analysis of users, they also may be used off-line to detect significant patterns in large archives of historical data. For example, a large rise above baseline temperature during a period of inactivity may precede the development of an ulcer 12.
Alternative embodiments may configure the pattern recognition system 68 and analyzer 70 to perform other processes that identify risk and emergence, as well as assist in tracking the progressions of ulcers 12 and pre-ulcers 14. For example, if there is no ambient temperature data from a thermogram prior to the patient's use of the platform 16, then some embodiments may apply an Otsu filter (or other filter) first to the high resolution thermogram to identify regions with large temperature deviations from ambient. The characteristics of these regions (length, width, mean temperature, etc.) then may be statistically compared to known distributions of foot characteristics to identify and isolate feet 10. The right foot thermogram may be mirrored and an edge-alignment algorithm can be employed to standardize the data for hotspot identification.
Two conditions can be evaluated independently for hotspot identification. The first condition evaluates to true when a spatially-localized contralateral thermal asymmetry exceeds a pre-determined temperature threshold for a given duration. The second condition evaluates to true when a spatially-localized ipsilateral thermal deviation between temporally successive scans exceeds a pre-determined temperature threshold for a given duration. The appropriate durations and thermal thresholds can be determined from literature review or through application of machine learning techniques to data from observational studies. In the latter case, a support vector machine or another robust classifier can be applied to outcome data from the observational study to determine appropriate temperature thresholds and durations to achieve a desired balance between sensitivity and specificity.
Illustrative embodiments have a set of prescribed patterns against which the pattern recognition system 68 and analyzer 70 compare to determine foot health. Accordingly, discussion of specific techniques above are illustrative of any of a number of different techniques that may be used and thus, are not intended to limit all embodiments of the invention.
The output of this analysis can be processed to produce risk summaries and scores that can be displayed to various users to trigger alerts and suggest the need for intervention. Among other things, state estimation models can simulate potential changes in the user's foot 10 and assess the likelihood of complications in the future. Moreover, these models can be combined with predictive models, such as linear logistic regression models and support vector machines, which can integrate a large volume and variety of current and historical data, including significant patterns discovered during off-line analysis. This may be used to forecast whether the user is likely to develop problems within a given timeframe. The predictions of likelihood can be processed into risk scores, which also can be displayed by both users and other third parties. These scores and displays are discussed in greater detail below.
The output table thus may provide the caregiver or healthcare provider with information, such as the fact that Patient B has a 90 percent probability that he/she will develop a foot ulcer 12 in the next 4-5 days. To assist in making clinical treatment decisions, the clinician also may access the patient's history file to view the raw data.
Other embodiments produce output information indicating the emergence of a pre-ulcer 14 at some spot on the foot 10. As known by those skilled in the art, a pre-ulcer 14 may be considered to be formed when tissue in the foot 10 is no longer normal, but it has not ruptured the top layer of skin. Accordingly, a pre-ulcer 14 is internal to the foot 10. More specifically, tissue in a specific region of the foot 10 may not be receiving adequate blood supply and thus, may need more blood. When it does not receive an adequate supply of blood, it may become inflamed and subsequently, become necrotic (i.e., death of the tissue). This creates a weakness or tenderness in that region of the foot 10. Accordingly, a callous or some event may accelerate a breakdown of the tissue, which ultimately may rupture the pre-ulcer 14 to form an ulcer 12.
Illustrative embodiments may detect the emergence of a pre-ulcer 14 in any of a number of manners described above. For example, the system may compare temperature readings to those of prior thermograms, such as the running average of the temperature at a given location. This comparison may show an elevated temperature at that spot, thus signaling the emergence of a new pre-ulcer 14. In more extreme cases, this may indicate the actual emergence of a new ulcer 12.
The emergence or detection of a pre-ulcer 14 can trigger a number of other preventative treatments that may eliminate or significantly reduce the likelihood of the ultimate emergence of an ulcer 12. To that end, after learning about a pre-ulcer 14, some embodiments monitor the progression of the pre-ulcer 14. Preferably, the pre-ulcer 14 is monitored during treatment in an effort to heal the area, thus avoiding the emergence of an ulcer 12. For example, the caregiver may compare each day's thermogram to prior thermograms, thus analyzing the most up to date state of the pre-ulcer 14. In favorable circumstances, during a treatment regimen, this comparison/monitoring shows a continuous improvement of the pre-ulcer 14, indicating that the pre-ulcer 14 is healing. The output information therefore can have current and/or past data relating to the pre-ulcer 14, and the risk that it poses for the emergence of an ulcer 12.
Sometimes, patients may not even realize that they have an ulcer 12 until it has become seriously infected. For example, if the patient undesirably does not use the foot monitoring system for a long time, he/she may already have developed an ulcer 12. The patient therefore may undergo an analysis of his/her foot/feet to produce output information indicating the emergence of an ulcer 12. To that end, the analyzer 70 may have prior baseline thermogram (i.e., data) relating to this patient's foot 10 (showing no ulcer), and make a comparison against that baseline data to determine the emergence of an actual ulcer 12. In cases where the data is questionable about whether it is an ulcer 12 or a pre-ulcer 14, the caregiver and/or patient nevertheless may be notified of the higher risk region of the foot 10 which, upon even a cursory visual inspection, should immediately reveal the emergence of an ulcer 12.
Some embodiments manually or automatically notify the relevant people about the health of the patient's foot 10. These notifications or messages (a type of “risk message”) may be in any of a number of forms, such as a telephone call, a text message, e-mail, and data transmission, or other similar mechanism. For example, the system may forward an e-mail to a healthcare provider indicating that the right foot 10 of the patient is generally healthy, while the left foot 10 has a 20 percent risk of developing an ulcer 12, and a pre-ulcer 14 also has emerged on a specified region. Armed with this information, the healthcare provider may take appropriate action, such as by directing the patient to stay off their feet 10, use specialized footwear, soak their feet 10, or immediately check into a hospital.
Accordingly, illustrative embodiments take advantage of the continuous data provided by two thermograms to ascertain various risks to foot health. In addition, such embodiments also monitor the foot 10 using an easy to follow regimen and form factor that encourages patient compliance. Early detection can assist in avoiding foot ulcers 12, while late detection can alert patients to yet undiscovered ulcers 12, which can then be effectively treated.
Various embodiments of the invention may be implemented at least in part in any conventional computer programming language. For example, some embodiments may be implemented in a procedural programming language (e.g., “C”), or in an object oriented programming language (e.g., “C++”). Other embodiments of the invention may be implemented as preprogrammed hardware elements (e.g., application specific integrated circuits, FPGAs, and digital signal processors), or other related components.
In an alternative embodiment, the disclosed apparatus and methods (e.g., see the various flow charts described above) may be implemented as a computer program product (or in a computer process) for use with a computer system. Such implementation may include a series of computer instructions fixed either on a tangible medium, such as a computer readable medium (e.g., a diskette, CD-ROM, ROM, or fixed disk) or transmittable to a computer system, via a modem or other interface device, such as a communications adapter connected to a network over a medium.
The medium may be either a tangible medium (e.g., optical or analog communications lines) or a medium implemented with wireless techniques (e.g., WIFI, microwave, infrared or other transmission techniques). The medium also may be a non-transient medium. The series of computer instructions can embody all or part of the functionality previously described herein with respect to the system. The processes described herein are merely exemplary and it is understood that various alternatives, mathematical equivalents, or derivations thereof fall within the scope of the present invention.
Those skilled in the art should appreciate that such computer instructions can be written in a number of programming languages for use with many computer architectures or operating systems. Furthermore, such instructions may be stored in any memory device, such as semiconductor, magnetic, optical or other memory devices, and may be transmitted using any communications technology, such as optical, infrared, microwave, or other transmission technologies.
Among other ways, such a computer program product may be distributed as a removable medium with accompanying printed or electronic documentation (e.g., shrink wrapped software), preloaded with a computer system (e.g., on system ROM or fixed disk), or distributed from a server or electronic bulletin board over the larger network 44 (e.g., the Internet or World Wide Web). Of course, some embodiments of the invention may be implemented as a combination of both software (e.g., a computer program product) and hardware. Still other embodiments of the invention are implemented as entirely hardware, or entirely software.
Although the above discussion discloses various exemplary embodiments of the invention, it should be apparent that those skilled in the art can make various modifications that will achieve some of the advantages of the invention without departing from the true scope of the invention.
This patent application is a continuation of U.S. patent application Ser. No. 17/703,785, filed Mar. 24, 2022, entitled “METHOD AND APPARATUS OF MONITORING FOOT INFLAMMATION,” assigned attorney docket number 3891/11703, and naming Brian Petersen, Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler as inventors, which is a continuation of U.S. patent application Ser. No. 15/144,658, filed May 2, 2016, entitled “METHOD AND APPARATUS OF MONITORING FOOT INFLAMMATION,” assigned attorney docket number 3891/11701, and naming Brian Petersen, Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler as inventors,which is a continuation patent application of U.S. patent application Ser. No. 14/662,738, U.S. Pat. No. 9,326,723 on issue date May 3, 2016, filed Mar. 19, 2015, entitled, “METHOD AND APPARATUS OF MONITORING FOOT INFLAMMATION,” assigned attorney docket number 3891/1013, and naming Brian Petersen, Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler as inventors,which is a continuation-in-part of U.S. patent application Ser. No. 13/799,828, now U.S. Pat. No. 9,259,178, filed Mar. 13, 2013, entitled, “METHOD AND APPARATUS FOR INDICATING THE RISK OF AN EMERGING ULCER,” assigned attorney docket number 3891/1001, and naming Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler, Brian Jude Petersen, David Charles Kale, and Adam Geboff as inventors,which claims priority from provisional U.S. patent application No. 61/618,889, filed Apr. 2, 2012, entitled, “AUTONOMOUS METHOD FOR PREDICTING MEDICAL CONDITIONS IN MAMMALS,” and naming Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler, David Charles Kale and Adam Geboff as inventors. U.S. patent application Ser. No. 14/662,738, U.S. Pat. No. 9,326,723 on issue date May 3, 2016, filed Mar. 19, 2015, entitled, “METHOD AND APPARATUS OF MONITORING FOOT INFLAMMATION,” assigned attorney docket number 3891/1013, and naming Brian Petersen, Jonathan David Bloom, David Robert Linders, Jeffrey Mark Engler as inventors, also claims priority to provisional U.S. patent application No. 61/968,696, filed Mar. 21, 2014. All of the above listed patent documents (patents and applications) are incorporated herein, in their entireties, by reference.
Number | Date | Country | |
---|---|---|---|
61968696 | Mar 2014 | US | |
61618889 | Apr 2012 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 17703785 | Mar 2022 | US |
Child | 18780201 | US | |
Parent | 15144658 | May 2016 | US |
Child | 17703785 | US | |
Parent | 14662738 | Mar 2015 | US |
Child | 15144658 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13799828 | Mar 2013 | US |
Child | 14662738 | US |