Detection and monitoring of pain centralisation in a patient

Abstract

The invention relates to a device (1) for aiding with the detection and monitoring of pain centralisation in a patient. A processor (5) is arranged to execute a graphic function which allows a patient, during a use session of the device (1), to delimit painful areas on a cartographic representation (CR) of their body, displayed by a screen (3), and to assign a coefficient which is representative of the intensity of the corresponding pain to each area. A graphic data set, which includes a time marker, position data of the painful areas and the coefficients assigned respectively, is stored in a memory means (7). A computer (9) is arranged to calculate a pain distribution index with respect to a source point and to calculate a centralisation index as a function of the difference between the respective pain distribution indices of two consecutive sessions.

Description

[0001] Description

[0002] Title: Detection and monitoring of centralization of a patient's pain

[0003] The field of the invention relates to the detection of a phenomenon of centralization of a patient's pain and to its monitoring within the framework of therapeutic care.

[0004] For example, low back pain, which refers to pain in the lumbar vertebrae, is a common condition and is therefore one of the leading causes of work incapacity. In particular, chronic low back pain - characterized by pain present for more than three months - affects nearly 10% of patients and has the effect of significantly limiting the practice of certain daily activities.

[0005] Among the known methods for treating low back pain, the McKenzie method - also referred to by the acronym MDT for "Mechanical Diagnosis and Therapy" (registered trademark) in the English-language literature - appears to be a promising approach. The McKenzie method is based on two principles: centralization and directional preference.

[0006] Centralization was first described by RA McKenzie (1981) in the publication “The lumbar spine: Mechanical diagnosis and therapy” (Wellington, New Zealand, Spinal Publications) and corresponds to a rapid and sustained abolition of pain in a disto-proximal pattern in response to mechanical stress of the musculoskeletal system. As such, centralization constitutes a phenomenon of spatio-temporal convergence. The spatial component lies both in the direction of pain displacement and in the reduction in the extent of the painful area(s). The temporal component lies in the rapid and sustained nature of the abolition of pain. [Fig. 1] illustrates, as an example, a phenomenon of centralization in which pain in a lower limb moves to the calf, then the thigh, then the lumbar region to finally reduce to a point Po: the source point.

[0007] Directional preference, discussed by R. Donelson et al. (1991) in the publication “Pain response to sagittal end-range spinal motion A prospective, randomized, multi-centered trial” (Spine 16; S206-S212), refers to the direction of movement or posture that provides pain relief or improved range of motion.

[0008] The McKenzie method involves achieving centralization through directional preference. In other words, the healthcare professional—e.g., a physiotherapist—seeks to abolish distal pain, and then proximal pain, through repeated movements or specific postures performed with directional preference, such as flexion, extension, lateral sliding, or rotation. Further details regarding the synergy between centralization and directional preference in the treatment of low back pain can be found in the publication by MW Werneke et al. (2011) “Association between directional preference and centralization in patients with low back pain” (Journal of Orthopaedic & Sports Physical Therapy, vol. 41, no. 1, pp. 22-31).

[0009] Beyond low back pain, centralization and directional preference apply, among other things, to mechanical back pain, non-severe neuralgia and more generally to musculoskeletal disorders.

[0010] Although widely used, the McKenzie method continues to be applied empirically. Thus, there is currently no tool that can reliably determine whether a centralization phenomenon occurs in response to exercises that mechanically stress the musculoskeletal system, and therefore whether a patient responds favorably to the McKenzie method.

[0011] Furthermore, over the last few decades, it has been noted that the centralization phenomenon, far from being limited to the McKenzie method, was a relevant indicator that could be adapted to different therapeutic approaches aimed at analgesia to assess the temporal evolution of a patient's pain for a wide spectrum of musculoskeletal disorders.

[0012] In particular, the spatial component of centralization does not necessarily follow the disto-proximal pattern as initially described by R.A. McKenzie (1981), even though the general principle remains the same: the spatial evolution of pain results in centripetal convergence. In other words, the pain moves to a point whose position can be predicted from the initial distribution of pain, and this movement is accompanied by a decrease in the extent of the painful area(s). For example, the progression of ankle algodystrophy may follow an inverse pattern, namely a proximo-distal pattern. Other bodily pain conditions may be mentioned as examples of pathologies whose evolution can be described using the principle of centralization, for example, facial pain after trauma caused by a blunt object or perineal pain following childbirth.

[0013] Thus, centralization - in the general sense of the term - is of interest in the context of therapeutic treatments of different kinds: pharmacological approaches for example, with the application of topical remedies, that is to say drugs to be applied to the painful area, based on capsaicin or lidocaine, but also neurochemical or neuroelectric approaches - notably neuromodulation -, and, of course, physical approaches such as the McKenzie method mentioned in detail previously.

[0014] It is understood from the above that, beyond the McKenzie method and a classic centralization obeying a disto-proximal pattern, there is a real need to be able to detect the presence of a centralization phenomenon, therefore a progressive displacement of the pain towards its origin in response to a treatment relating to the therapeutic approach adopted, and to ensure its monitoring throughout the therapeutic care of the patient.

[0015] The present invention improves the situation. As such, the invention relates to a device for assisting in the detection and monitoring of a patient's pain centralization, comprising:

[0016] - a screen arranged to display a map representation of a patient's body,

[0017] - a processor arranged to execute a graphic function allowing a patient, during a session of use of the device, to delimit one or more painful areas on the cartographic representation and to attribute to each painful area a coefficient representative of the intensity of the corresponding pain, and

[0018] - a memory arranged to store, for each patient, graphic data sets, each graphic data set including a time marker corresponding to a session of use of the device, position data of the painful area(s) delimited by the patient during the session of use of the device as well as the coefficient(s) respectively assigned.

[0019] The processor is further arranged to select, based on the respective time markers of the patient's graphical data sets stored in the memory, a reference patient's graphical data set and to derive therefrom a source point of the patient on the map representation.

[0020] The device further comprises a calculator arranged to calculate, for a patient's graphical data set, a pain distribution index defined as follows: Yes k is the pain distribution index of the k-th patient's graphical dataset,

[0021] - N k is the number of points, on the map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0022] - P o is the source point of the patient,

[0023] - P k iis the i-th point, on the map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0024] - d P^ Po) is the distance between point P k i and the source point P o , and - s(p k i ') is the coefficient attributed to the painful area within which point P is located k i .

[0025] The calculator is further arranged to calculate a centralization index of a patient based on the difference between the respective pain distribution indices of two sets of graphical data of the patient including respective time markers corresponding to consecutive sessions of use of the device.

[0026] The screen is typically a touch screen.

[0027] In one or more embodiments, the graphing function allows a patient to assign to each painful area a coefficient to be selected from a finite set of unaffected natural numbers, a coefficient being greater the greater the intensity of the pain.

[0028] For example, the finite set of natural numbers is {1, 2, 3, 4}.

[0029] In one or more embodiments, the processor is arranged to select, as the patient's reference graphics dataset, the patient's graphics dataset stored in the memory whose time stamp corresponds to the oldest session of use of the device.

[0030] In one or more embodiments, the calculator is arranged to calculate a centralization index of a patient as follows:

[0031] > -i — 4

[0032] Jk ~ j lk-l where: J kis the patient's centralization index calculated from the consecutive pain distribution indices I k-1 and I k .

[0033] In one or more embodiments, the processor is further arranged to compare the centralization index of a patient with a predetermined discrimination threshold and to return information on the presence of centralization of the patient's pain if the centralization index is greater than the predetermined discrimination threshold or information on the absence of centralization of the patient's pain if the centralization index is less than or equal to the predetermined discrimination threshold. Advantageously, the predetermined discrimination threshold is between 0.53 and 0.65, and preferably equal to 0.59.

[0034] The invention also relates to a method for assisting in the detection and monitoring of a patient's pain centralization. The method is implemented by the device described above during a session of use thereof and comprises:

[0035] - display a map representation of a patient's body on the screen,

[0036] - perform a graphic function allowing the patient to delimit one or more painful areas on the cartographic representation and to assign to each painful area a coefficient representative of the intensity of the corresponding pain,

[0037] - store, in the memory, a set of graphic data including a time marker corresponding to the session of use of the device, position data of the painful area(s) delimited by the patient during the session of use of the device as well as the coefficient(s) respectively assigned.

[0038] The method further comprises:

[0039] - select, based on the respective time markers of the patient's graphical data sets stored in the memory, a reference graphical data set of the patient and deduce therefrom a source point of the patient on the cartographic representation,

[0040] - calculate, for the patient's graphical data set, a current pain distribution index as follows: Yes k is the pain distribution index of the k-th patient's graphical dataset,

[0041] • N k is the number of points, on the map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0042] • P o is the source point of the patient,

[0043] • P k iis the i-th point, on the map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0044] • d P^ Po) is the distance between point P k i and the source point P o , And

[0045] • s(P fc ,i) is the coefficient attributed to the painful zone within which point P is located k i ,

[0046] - calculate a patient centralization index based on the difference between the current pain distribution index and the pain distribution index of the graphical dataset including a time marker corresponding to the previous session of use of the device.

[0047] Finally, the invention also relates to a computer program comprising instructions whose execution, by a processor, results in the implementation of the method described above.

[0048] Other characteristics, details and advantages will appear on reading the detailed description below, and on analyzing the attached drawings in which:

[0049] [Fig. 1] illustrates a phenomenon of centralization of a patient's pain,

[0050] [Fig. 2] illustrates a device for assisting in the detection and monitoring of a centralization of a patient's pain according to the invention,

[0051] [Fig. 3A] illustrates a screenshot of the device of [Fig. 2] during a first session of use thereof,

[0052] [Fig. 3B] illustrates a screenshot of the device of [Fig. 2] during a second session of use thereof,

[0053] [Fig. 3C] illustrates a screenshot of the device of [Fig. 2] during a third session of use thereof,

[0054] [Fig. 3D] illustrates a screenshot of the device of [Fig. 2] during a fourth session of use thereof,

[0055] [Fig. 4] illustrates a method for assisting in the detection and monitoring of a centralization of a patient's pain according to the invention, and

[0056] [Fig. 5] illustrates a ROC curve used to determine an appropriate discrimination threshold for a centralization index involved in the method of [Fig. 4]. [Fig. 2] illustrates a device 1 to aid in the detection and monitoring of centralization of a patient's pain.

[0057] Device 1 is intended to be used by a healthcare professional as part of managing a patient's pain using a therapeutic approach exploiting the principle of centralization.

[0058] In the context of the present invention, centralization generally refers to the centripetal convergence of a patient's pain, i.e. the progressive displacement of pain towards a source point accompanied by a reduction in the extent of the painful area(s), in response to the treatment relating to the therapeutic approach adopted. In the particular case of the McKenzie method, the pain evolves according to a disto-proximal pattern.

[0059] For example, device 1 can be used by a physiotherapist during a musculoskeletal physiotherapy session.

[0060] It should be noted that the use of device 1 is part of a post-diagnostic context: the healthcare professional has already established a clinical picture of the patient, i.e. the list of symptoms and signs of clinically observable pathological conditions, and has noted the presence of musculoskeletal disorders, for example at the level of the spine or extremities.

[0061] Device 1 helps the healthcare professional, after identifying the cause of musculoskeletal disorders, to select the most appropriate therapeutic approach. As such, Device 1 is particularly useful for determining whether the patient is likely to respond favorably to a treatment based on directional preference such as the McKenzie method since it allows for the detection of a centralization phenomenon and, if necessary, for monitoring it until complete abolition of pain. Generally speaking, the use of Device 1 is of interest in any therapeutic approach based on the centralization phenomenon.

[0062] Still referring to [Fig. 2], the device 1 comprises a screen 3, a processor 5, a memory 7 and a calculator 9. Such a device 1 can take various and varied forms. For example, the device 1 can be a mobile device, such as a personal digital assistant, a laptop, a smartphone or even a tablet computer, which facilitates its transport. However, the device 1 can also be a desktop computer - or fixed computer - and be intended for use in a place dedicated to the reception and care of a patient, for example a hospital, a clinic or even a medical or paramedical office.

[0063] The device 1 may be specifically designed for the purpose of detecting and, where appropriate, monitoring the centralization of a patient's pain. Alternatively, the device 1 may be obtained from a pre-existing device whose operation is modified, for example by installing software or downloading an application.

[0064] The screen 3 is arranged to provide a display function to the device 1. In particular, the screen 3 makes it possible to view information relating to the care of a patient and the evolution of their symptoms. In the context of the invention, the screen 3 allows a healthcare professional to have a visual representation of a possible centralization phenomenon and, where appropriate, to observe the displacement of the pain from one session to another.

[0065] The screen 3 may be a liquid crystal display (better known by the acronym LCD for “liquid crystal display”). In such a case, the interaction of a user – the healthcare professional or the patient – ​​may be carried out by means of an input device such as a keyboard or a mouse. Alternatively, the screen 3 is a touch screen which, in addition to the display function, therefore gives the device 1 a pointing function. It is then possible to interact with the screen directly with a finger or a stylus.

[0066] The processor 5 is arranged to evaluate a distribution of a patient's pain and to detect the possible presence of a centralization phenomenon.

[0067] Such operation of the processor 5 may result from software or the execution of an application. Such an application may be downloaded via an application store - usually referred to by the English term "store" - or be present by default on the device 1.

[0068] In particular, the processor 5 is arranged to implement at least one display function and one graphics function.

[0069] As illustrated in [Fig. 2], the implementation of the display function results in the display, on screen 3, of a CR map representation of a patient's body.

[0070] To do this, a CR map representation of each patient's body can be stored in memory 7.

[0071] It is known to those skilled in the art that there are several techniques for generating a CR map representation of the body of an individual, and of a patient in particular.

[0072] Such a CR map representation can be constructed from images or photographs of the patient, for example using a 3D scanner. In the medical context, it is possible to use images acquired using a medical imaging technique such as computed tomography (CT) or magnetic resonance imaging (MRI).

[0073] The device 1 may also be equipped with a digital camera or camera which may be used to acquire images or photographs of the patient for the purpose of generating a CR map representation of the patient's body.

[0074] Another possibility is to use software capable of generating a CR map representation from patient morphological data such as body mass index (BMI).

[0075] As examples, the problem of obtaining a CR map representation as faithful as possible to the patient's body is addressed by European patents EP 3 043 706 B1 and EP 3 043 707 B1 which deal generally with mapping a patient's pain and converting it into square centimetres (cm 2 ) pixels of a painful area drawn by the patient on the CR map representation.

[0076] The implementation of the graphic function results in the provision to a user - typically the patient - of graphic tools making it possible to delimit, on the CR cartographic representation displayed by the screen 3, one or more painful zones and to attribute to each of these zones a coefficient representative of the intensity of the pain.

[0077] An example of the features and options associated with the graphing function are shown in [Fig. 3A], [Fig. 3B], [Fig. 3C] and [Fig. 3D] which illustrate the CR mapping representation of the same patient during multiple sessions of use of the device 1.

[0078] Such features and options allow a patient to draw a painful area in the form of a closed curve. It is possible for the patient to either delineate a painful area and then select the coefficient to be assigned to the delineated area, or directly select the coefficient and then delineate the painful area which is then automatically assigned the selected coefficient.

[0079] The graphics function can allow free drawing or offer predetermined geometric shapes, including a circular shape, the dimensions of which can be adjusted by the patient. In addition, the graphics function can offer the patient classic features and options of a drawing application such as an eraser - to erase all or part of the drawing produced -, a zoom, a palette for selecting the thickness of the drawing tools, undoing the last action performed or even the possibility of moving, via rotation or translation, to access all areas of the CR map representation.

[0080] Advantageously, the graphic function includes a self-calibration mechanism to compensate for any approximations by the patient when indicating the painful areas on the screen 3.

[0081] Here again, the European patents EP 3 043 706 B1 and EP 3 043 707 B1 mentioned above detail a possible solution for programming such self-calibration. To do this, the healthcare professional can ask the patient to indicate, on the CR map representation, target points designated by contact on the patient's physical body. The target points can be chosen from a predefined list of morphological landmarks. Any deviations between the target points and the points designated by the patient are calculated and then used to adapt the configuration of the graphic function in order to subsequently correct the inaccuracies in the patient's perception of the positions of the painful points on their body.

[0082] Furthermore, the processor 5 is arranged to generate, at each session of use of the device 1 , a graphical data set. Such a graphical data set makes it possible to restore the patient's interactions with the device 1 , namely the delimitation of the painful areas and the coefficient attributed to each. Given that the phenomenon of centralization corresponds to a temporal evolution of the distribution of the pain, each graphical data set includes a time marker - also designated by the English term "timestamp" - which corresponds to a session of use of the device 1 . Such a time marker makes it possible, within the device 1 , to keep a history of the sessions of use of the device 1 and therefore to record the displacement of the pain.

[0083] In addition to the time marker, a graphical data set includes position data of the painful area(s) demarcated by the patient during the session of use of the device 1 as well as the coefficient(s) respectively assigned.

[0084] As detailed in the remainder of the description, the processor 5 is further arranged to select a reference graphic data set - which acts as a starting point for establishing the presence of a centralization phenomenon - and to deduce a source point therefrom.

[0085] [Fig. 3A], [Fig. 3B], [Fig. 3C] and [Fig. 3D] each show, on the CR map representation, the source point Po.

[0086] The processor 5 can be implemented in any known manner, for example in the form of a microprocessor, a programmable logic device (PLD) or a dedicated chip of the FPGA (Field Programmable Gate Array) or SoC (System on Chip) type, a grid of computing resources, a microcontroller or any other specific form having the computing power necessary for detecting and monitoring a centralization of pain. One or more of these elements can also be implemented in the form of specialized electronic circuits of the ASIC (Application-Specific Integrated Circuit) type. A combination of processors and electronic circuits can also be envisaged.

[0087] The memory 7 is arranged to store instructions whose execution, by the processor 5, results in the operation of the device 1.

[0088] In particular, the memory 7 is arranged to store data relating to the display function and the graphics function detailed above. The memory 7 is structured such that such stored data are respectively associated with the patients to whom they relate.

[0089] Thus, the memory 7 is arranged to store, for each patient, a CR cartographic representation of the patient's body.

[0090] The memory 7 is further arranged to store, for each patient, graphic data sets resulting from the patient's interaction with the device 1. As explained previously, each graphic data set corresponds to a session of use of the device 1 and includes, as such, a time marker, position data of the painful area(s) delimited by the patient during the session of use of the device 1 as well as the coefficient(s) representative of the intensity of the pain respectively attributed.

[0091] Memory 7 can refer to any data storage medium designed to receive and store digital data, for example a hard disk, a solid-state drive (SSD) or more generally any computer hardware allowing the storage of data on flash memory. Memory 7 can also be a random access memory or a magneto-optical disk. A combination of several storage media can also be envisaged.

[0092] Finally, the calculator 9 is arranged to carry out calculations from graphic data sets stored in the memory 7.

[0093] More particularly, the calculator 9 is arranged to calculate a pain distribution index at the end of each session of use of the device 1. Such a pain distribution index makes it possible to evaluate the relative position of a patient's pain in relation to the source point.

[0094] The calculator 9 is further arranged to calculate a centralization index. Such a centralization index makes it possible, at the very beginning of the patient's care, to determine whether a centralization phenomenon is manifesting in the patient and then, if so, to monitor the centralization of the pain from one session to the next.

[0095] In [Fig. 2], the processor 5 and the computer 9 are represented as separate entities. However, those skilled in the art will understand that this distinction is purely functional and that the processor 5 and the computer 9 may therefore form a single entity.

[0096] Advantageously, the device 1 may comprise a communication module (not shown in [Fig. 2]) for transmitting data to a server. Such a server may thus receive, from several devices such as the device 1, data relating to patients and store them. Such data typically includes, for each patient, one or more pain distribution indices and one or more centralization indices. The data received and stored by the server may also include, for each patient, one or more graphical data sets.

[0097] Reference is now made to [Fig. 4] which illustrates a method for assisting in the detection and monitoring of a patient's pain centralization.

[0098] It should be noted that this method is implemented by the device 1 described above during a session of use thereof. Such a session of use of the device 1 is part of the care of a patient suffering from musculoskeletal disorders such as lower back pain by a health professional, for example a physiotherapist.

[0099] In particular, we can distinguish, on the one hand, the very first session of use of device 1 which occurs at the start of the patient's care and makes it possible to determine the initial distribution of the pain using the pain distribution index and, on the other hand, the subsequent sessions of use of device 1 which each make it possible to evaluate the displacement of the pain compared to the previous session using the centralization index.

[0100] Between two consecutive sessions of use of device 1, the healthcare professional provides patients with the treatment related to the therapeutic approach adopted. In the case of the McKenzie method, this treatment takes the form of exercises that mechanically constrain the musculoskeletal system, namely repeated movements or specific postures performed according to a directional preference. Device 1 allows, after such treatment, to take stock of centralization.

[0101] The second session of use of device 1 allows the calculation of a first centralization index which can be used by the healthcare professional to determine whether a centralization phenomenon occurs and therefore whether the therapeutic approach tested is appropriate.

[0102] Each patient can be associated with a profile that allows them to be uniquely identified on the device 1. The healthcare professional, or even the patient, can access this profile with a username and password. Connection to the patient's profile allows the processor 5 to retrieve the patient's data from the memory 7, in particular the CR map representation of the patient's body and the graphic data sets already stored.

[0103] During an operation 400, the processor 5 performs the display function. A map representation CR of the patient's body is then displayed on the screen 3 of the device 1.

[0104] During an operation 410, the processor 5 executes the graphic function. The patient then has the graphic tools to delineate, on the CR map representation, the painful areas of his body and to assign, to each painful area, a coefficient representative of the intensity of the pain.

[0105] Typically, the patient can select such a coefficient from a finite set of unharmed natural numbers. Such a coefficient is larger the higher the pain intensity. For example, the finite set of natural numbers is {1, 2, 3, 4}. Coefficient 1 corresponds to mild pain, coefficient 2 corresponds to moderate pain, coefficient 3 corresponds to severe pain, and coefficient 4 corresponds to very severe pain.

[0106] Of course, it is possible to configure the graphing function to reduce or expand the size of the finite set of natural numbers from which the patient can select the coefficient representing the intensity of the pain.

[0107] As such, [Fig. 3A], [Fig. 3B], [Fig. 3C] and [Fig. 3D] represent the painful areas delimited by the patient on screen 3 during separate sessions of use of device 1 as well as the coefficient attributed to each of the painful areas.

[0108] In [Fig. 3A], the patient drew a first painful zone Z1 whose intensity corresponds to coefficient 4 and a second painful zone Z2 whose intensity corresponds to coefficient 3.

[0109] [Fig. 3A] corresponds, for example, to the very first session of use of device 1 which takes place at the beginning of the patient's care. This session of use of device 1 provides an overview of the initial distribution of the patient's pain.

[0110] In [Fig. 3B], the patient drew a third painful zone Z3 whose intensity corresponds to coefficient 4 and a fourth painful zone Z4 whose intensity corresponds to coefficient 3.

[0111] [Fig. 3B] corresponds, for example, to the second session of use of device 1 and therefore makes it possible to know the evolution of the distribution of the patient's pain. In the context of the McKenzie method, this second session of use of device 1 is separated from the first by exercises of mechanical constraint of the musculoskeletal system. The health professional can then determine, using device 1, if a phenomenon of centralization occurs.

[0112] In [Fig. 3C], the patient drew a fifth painful zone Z5 whose intensity corresponds to coefficient 3 and a sixth painful zone Z6 whose intensity corresponds to coefficient 2.

[0113] Finally, in [Fig. 3D], the patient drew a seventh painful zone Z7 whose intensity corresponds to coefficient 2 and an eighth painful zone Z8 whose intensity corresponds to coefficient 1.

[0114] Analysis of [Fig. 3A], [Fig. 3B], [Fig. 3C] and [Fig. 3D] reveals a centralization phenomenon. Indeed, from one figure to the next, we observe that the pain converges towards the source point Po. Furthermore, this displacement is concomitant with a reduction in the body surface area affected by pain. In other words, the total extent of the painful areas gradually decreases. Thus, the total surface area of ​​Z3 and Z4 is less than the total surface area of ​​Z1 and Z2; the total surface area of ​​Z5 and Z6 is less than the total surface area of ​​Z3 and Z4; and the total surface area of ​​Z7 and Z8 is less than the total surface area of ​​Z5 and Z6.

[0115] In [Fig. 4], operations 400 and 410 are shown as separate operations implemented sequentially. However, those skilled in the art will understand that operations 400 and 410 may be implemented simultaneously.

[0116] During an operation 420, that is to say after the patient has drawn all the painful areas and indicated for each the appropriate coefficient, the processor 5 generates a set of graphic data corresponding to the current session of use of the device 1.

[0117] The graphical data set includes a time marker corresponding to the session of use of the device 1, position data of the painful area(s) delimited by the patient during the session of use of the device 1 as well as the coefficient(s) respectively assigned. The graphical data set is stored in the memory 7 and is associated there with the corresponding patient, or more precisely with an identifier of the corresponding patient. A history of the sessions of use of the device 1 is thus stored in the memory 7 for each patient.

[0118] During an operation 430, the processor 5 selects, based on the respective time markers of the patient's graphic data sets stored in the memory 7, a reference graphic data set.

[0119] The reference graph dataset helps identify the source point, i.e. the point to which the pain should move.

[0120] In the particular case of the McKenzie method, the source point is the proximal origin of the pain. In contrast to distal pain, that is, pain close to the extremity of a limb, the source point is proximal and is therefore located close to the root of the painful limb. Therefore, the source point depends on the initial distribution of the pain, which may be cervical, thoracic, lumbar, or, in the example in [Fig. 1], lumbar-radicular.

[0121] As explained previously, the McKenzie method is not the only therapeutic approach to exploit centralization and the latter can result in a spatial evolution other than the disto-proximal pattern - provided that this evolution corresponds to a displacement of the pain towards a point, that is to say the source point, whose position is predictable. Thus, for pain at the extremity of a limb, for example algodystrophy - or complex regional pain syndrome (CRPS) - of the ankle, the source point is at the ankle. Centralization then manifests itself by a centripetal convergence of the painful areas.

[0122] As an example, facial pain caused by a traumatic injury to the supraorbital nerve can be mentioned. The source point is then at the facial emergence of this nerve. Here again, centralization corresponds to a centripetal convergence of the painful areas.

[0123] Thus, still during operation 430, the processor 5 identifies the position - for example in the form of coordinates - of the source point from the position data of the painful area(s) delimited by the patient. As mentioned previously, [Fig. 3A], [Fig. 3B], [Fig. 3C] and [Fig. 3D] each show, on the map representation CR, the source point Po.

[0124] Typically, the reference graphics dataset corresponds to the graphics dataset whose time stamp corresponds to the oldest usage session of device 1 stored in memory 7. Normally, the oldest usage session of device 1 is also the first.

[0125] Advantageously, the position of the source point of each patient can be stored in the memory 7. Thus, it is not necessary, at each session of use of the device 1, to retrieve again the reference graphic data set and to proceed again with the identification of the source point.

[0126] The device 1 may also be configured to offer the healthcare professional the possibility of selecting, for the determination of the source point, the session of use of the device 1 of his choice. The processor 5 then retrieves, from the memory 7, the graphic data set whose time marker corresponds to the selected session of use of the device 1 and deduces the source point therefrom. Such an option may be useful in the event of atypical displacement of the pain noted by the healthcare professional. The position of the source point can then be updated in the memory 7.

[0127] During an operation 440, the computer 9 calculates, for the patient's graphic data set, a pain distribution index which depends on the extent of the painful areas and the coefficient assigned to each. For each point - or pixel - of the map representation CR, the position data allows the computer 9 to determine whether or not it is present in a painful area and, if so, the coefficient assigned to the painful area within which it is present.

[0128] In particular, Calculator 9 calculates the pain distribution index as follows: Yes k is the pain distribution index of the k-th patient's graphical dataset,

[0129] - N k is the number of points, on the CR map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0130] - P ois the source point of the patient,

[0131] - P k i is the i-th point, on the CR map representation, located in a painful area of ​​the k-th graphical data set of the patient,

[0132] - d P^ Po) is the distance between point P k i and the source point P o , And

[0133] - s(P fc ,i) is the coefficient attributed to the painful zone within which point P is located k i .

[0134] The In(-) function - or natural logarithm - makes it possible to give the coefficient s less importance than the distance d P k i , P0') in the calculation of the pain distribution index l k .

[0135] A standardized value of the pain distribution index can also be calculated:

[0136] 1 I k K = ~ JV lk K Yes kis the normalized value of the pain distribution index l k ,

[0137] - N is the number of points in the CR cartographic representation.

[0138] In an operation 450, the calculator 9 calculates a patient centralization index based on the difference between the respective pain distribution indices of two sets of patient graphical data including respective consecutive time markers.

[0139] Here, the term "consecutive" refers to the chronology of device 1 usage sessions. Time markers allow graphical datasets to be timestamped. Therefore, graphical datasets including consecutive time markers correspond to consecutive device 1 usage sessions. The sequence of pain distribution indices {I k} is chronologically ordered. For example, calculator 9 calculates the patient's centralization index as follows:

[0140] > -i — 4 Jk ~ j lk-l where: J k is the patient's centralization index calculated from the consecutive pain distribution indices I k-1 and I k .

[0141] In the following description, the centralization index is considered to be the one calculated with the previous formula. It can also be noted that, in this formula, the pain distribution index can be replaced by the normalized pain distribution index: the value of the centralization index is unchanged.

[0142] The centralization index can be displayed on screen 3 - possibly as a percentage (100 x J k ) - to allow the healthcare professional to read it.

[0143] In the case of a session using device 1 for monitoring the patient's centralization, the method may be interrupted at the end of operation 450. The centralization index allows the healthcare professional to verify that the pain continues to move as expected. Thus, when the therapeutic approach adopted is the McKenzie method, the centralization index allows the healthcare professional to ensure that the pain moves in a disto-proximal pattern.

[0144] However, in the case of a session of use of the device 1 aimed at detecting a possible centralization phenomenon and therefore determining whether the chosen therapeutic approach is appropriate, the process continues. This case, which typically corresponds to the second session of use of the device 1, is that shown in [Fig. 4],

[0145] Thus, during an operation 460, the processor 5 compares the calculated centralization index with a predetermined discrimination threshold.

[0146] Such a discrimination threshold can be determined from a receiver operating characteristic (ROC) curve. The ROC curve is a measure of the performance of a binary classifier. In this case, the binary classifier is the centralization index, which allows each patient to be classified into one of two categories: patients for whom centralization occurs and those for whom no centralization occurs.

[0147] If the centralization index is greater than the discrimination threshold, the patient is classified in the category of patients for whom a centralization phenomenon occurs. Conversely, if the centralization index is less than or equal to the discrimination threshold, the patient is classified in the category of patients for whom no centralization phenomenon occurs.

[0148] The purpose of the ROC curve is to determine the discrimination threshold that allows the most reliable results possible to be obtained.

[0149] The ROC curve represents sensitivity as a function of antispecificity, i.e. the 1's complement of specificity. Sensitivity corresponds to the rate of true positives classified as positive, while antispecificity corresponds to the rate of true negatives classified as positive. Each point on the ROC curve corresponds to a discrimination threshold value. In the case of [Fig. 5], the points corresponding to the following values ​​are indicated: +00; 0.9; 0.8; 0.6; 0.2 and 0.

[0150] The ROC curve shown in [Fig. 5] has an area under curve (AUC) of approximately 0.9907, indicating that the centralization index is able to generate a classifier that is close to the ideal classifier and significantly different from the random classifier.

[0151] The analysis of the ROC curve of [Fig. 5] allowed the Applicant to conclude that a discrimination threshold of between 0.53 and 0.65 makes it possible to obtain a sufficiently reliable classifier with a sensitivity close to 1.00 (100%) and an antispecificity close to 0.052 (5.2%). Preferably, the discrimination threshold is substantially equal to 0.59 to be as close as possible to the ideal point, i.e. the point with coordinates (0,1). Thus, if the centralization index is less than or equal to the discrimination threshold, the processor 5 returns, during an operation 470, information on the absence of centralization.

[0152] Conversely, if the centralization index is greater than the discrimination threshold, the processor 5 returns, during an operation 480, information on the presence of centralization.

[0153] The information on the presence or absence of centralization is for example displayed on screen 3. This information can be stored in memory 7.

[0154] Device 1 and the method for assisting in the detection and monitoring of a patient's pain centralization were subjected to several tests to verify the relevance of the pain distribution index and, above all, the reliability of the centralization index.

[0155] A sample of these tests is presented below.

[0156] This sample corresponds to twelve patients, all presenting with common chronic low back pain, with pain in the back and legs. For each of these patients, the Applicant proceeded, on the one hand, to calculate a centralization index during an initial physiotherapy session and, on the other hand, to collect the physiotherapist's opinion as to the presence or absence of centralization.

[0157] The results obtained for these patients, each bearing an identification number from 1 to 12, are shown in the table below. The indication "CP+" indicates that the physiotherapist concluded that centralization was present, while the indication "CP-" indicates that the physiotherapist concluded that centralization was absent. The patients are classified, from top to bottom, by decreasing centralization indices.

[0158] This sample illustrates a correlation between the centralization index and the physiotherapist's conclusion. Taking a discrimination threshold C=0.59, it appears that only patient 4 is in a situation in which device 1 and the physiotherapist disagree about the presence of centralization. On the other hand, the same result is given by device 1 and the physiotherapist for all other patients, i.e. approximately 91.67% of the patients in the sample.

Claims

Claims

1. Device (1) for assisting in the detection and monitoring of a centralization of a patient's pain, comprising: - a screen (3) arranged to display a map representation (CR) of a patient's body, - a processor (5) arranged to execute a graphic function allowing a patient, during a session of use of the device (1), to delimit one or more painful zones (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) on the cartographic representation (CR) and to attribute to each painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) a coefficient representative of the intensity of the corresponding pain, and - a memory (7) arranged to store, for each patient, graphic data sets, each graphic data set including a time marker corresponding to a session of use of the device (1), position data of the painful zone(s) (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) delimited by the patient during the session of use of the device (1) as well as the coefficient(s) respectively assigned, said device (1) being characterized in that the processor (5) is further arranged to select, as a function of the respective time markers of the graphic data sets of a patient stored in the memory (7), a reference graphic data set of the patient and to deduce therefrom a source point of the patient on the cartographic representation (CR), in that the device (1) further comprises a calculator (9) arranged to calculate, for a graphic data set of a patient,a pain distribution index defined as follows:, Yes k is the pain distribution index of the k-th patient's graphical dataset, - N k is the number of points, on the cartographic representation (CR), located in a painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) of the k-th set of patient graphic data, - P o is the source point of the patient, - P k i is the i-th point, on the cartographic representation (CR), located in a painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) of the k-th graphical data set of the patient, - d P^ Po) is the distance between point P k i and the source point P o , And - s(P fc ,i) is the coefficient attributed to the painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) within which point P is located k i, and in that the calculator (9) is further arranged to calculate a centralization index of a patient as a function of the difference between the respective pain distribution indices of two sets of graphical data of the patient including respective time markers corresponding to consecutive sessions of use of the device (1).

2. Device (1) according to claim 1, characterized in that the screen (3) is a touch screen.

3. Device (1) according to claim 1 or 2, characterized in that the graphic function allows a patient to assign to each painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) a coefficient to be selected from a finite set of unharmed natural integers, a coefficient being all the greater as the intensity of the pain is high.

4. Device (1) according to claim 3, characterized in that the finite set of natural integers is {1, 2, 3, 4}.

5. Device (1) according to one of the preceding claims, characterized in that the processor (5) is arranged to select, as the patient's reference graphic data set, the patient's graphic data set stored in the memory (7) whose time marker corresponds to the oldest session of use of the device (1).

6. Device (1) according to one of the preceding claims, characterized in that the calculator (9) is arranged to calculate a centralization index of a patient as follows: > -i — 4 Jk ~ j lk-l where: J k is the patient's centralization index calculated from the consecutive pain distribution indices I k-1 and I k .

7. Device (1) according to one of the preceding claims, characterized in that the processor (5) is further arranged to compare the centralization index of a patient with a predetermined discrimination threshold and to return information on the presence of centralization of the patient's pain if the centralization index is greater than the predetermined discrimination threshold or information on the absence of centralization of the patient's pain if the centralization index is less than or equal to the predetermined discrimination threshold.

8. Device (1) according to claim 7 taken in combination with claim 6, characterized in that the predetermined discrimination threshold is between 0.53 and 0.65, and preferably equal to 0.

59.

9. Method for assisting in the detection and monitoring of a centralization of a patient's pain, said method being implemented by the device (1) according to one of the preceding claims during a session of use thereof and comprising: - display (400) a map representation (CR) of a patient's body on the screen (3), - execute (410) a graphic function allowing the patient to delimit one or more painful zones (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) on the cartographic representation (CR) and to attribute to each painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) a coefficient representative of the intensity of the corresponding pain, - storing (420), in the memory (7), a set of graphic data including a time marker corresponding to the session of use of the device (1), position data of the painful zone(s) (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) delimited by the patient during the session of use of the device (1) as well as the coefficient(s) respectively assigned, said method being characterized in that it further comprises: - select (430), based on the respective time markers of the patient's graphic data sets stored in the memory (7), a data set patient reference charts and deduce a patient source point on the map representation (CR), - calculate (440), for the patient's graphical data set, a current pain distribution index as follows: Or : • / kis the pain distribution index of the k-th patient's graphical dataset, • N k is the number of points, on the cartographic representation (CR), located in a painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) of the k-th graphic data set of the patient, • P o is the source point of the patient, • P k i is the i-th point, on the cartographic representation (CR), located in a painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) of the k-th graphical data set of the patient, • d P^ Po) is the distance between point P k i and the source point P o , And • s(P fc ,i) is the coefficient attributed to the painful zone (Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8) within which point P is located k i , - calculate (450) a patient centralization index based on the difference between the current pain distribution index and the pain distribution index of the graphical data set including a time marker corresponding to the previous session of use of the device (1).

10. Computer program comprising instructions whose execution, by a processor (5), results in the implementation of the method according to claim 9.

Images