Improved system for measuring red blood cell sedimentation rate and related method

Abstract

Described is a system (1) for measuring the erythrocyte sedimentation rate in a blood sample, the system comprising: a support (2) for accommodating at least one tube (P) of the blood sample to be analyzed; a rocking element (3) configured to rock the tube (P); at least one detection unit (4) configured to perform at least one optical measurement on the blood sample in the tube (P); a movement device (4m) configured to cause a relative movement between the detection unit (4) and the tube (P) during the optical measurement; and a processing unit (C) adapted to process the signal (Sgn) from the detection unit (4). Suitably, based on the signal (Sgn) from the detection unit (4), the processing unit (C) is configured to: create a reading curve (L) corresponding to the absorption of the radiation emitted by the detection unit (4) as a function of the relative movement between the detection unit (4) and the tube (P); defining, on the basis of the reading curve (L), an ideal curve (T) of the trapezoidal type adapted to approach the reading curve (L); the system (1) further comprises means (Out) for outputting a measurement result on the basis of the generated value, executing an optimization procedure of the ideal curve (T), thereby generating an optimized ideal curve (T '), and generating, on the basis of the optimization procedure, at least one value indicative of the erythrocyte sedimentation rate of the blood sample in the tube P. The system (1) further comprises an output device (Out) for outputting the measurement result on the basis of the generated value.

Description

[0001] manual Technical Field

[0002] The present invention relates to a system and related methods for analyzing biological samples, in particular for measuring erythrocyte sedimentation rate in blood samples. The following description relates to this field of application, and its sole purpose is to simplify its discussion. Background Art

[0003] The measurement of the erythrocyte sedimentation rate (ESR) is a very common laboratory test used to quickly identify inflammatory processes. Specifically, the rate at which red blood cells in a blood sample fall to the bottom of a tube is assessed.

[0004] As is well known in the art, the benchmark method for calculating ESR is the Westergreen method, which measures the precipitate formed after one hour of diluting the blood to be analyzed with sodium citrate in a graduated tube. Thus, the calculation method prescribes the use of a dedicated tube and a predetermined timing.

[0005] There are instruments that can also measure the ESR on standard tubes (e.g. normal blood cell counting tubes), whereby a measurement corresponding to the benchmark values ​​of the above-mentioned Westergreen method is obtained by optical absorption measurement of the blood sample in the tube. In this case, the measurement is carried out in a relatively short time, for example, the blood sample is allowed to stabilize for about twenty minutes before the last reading is taken.

[0006] In instruments of the type described above, it is crucial to develop optimal calculation procedures for analyzing the optical absorption of blood samples and to obtain all the required parameters for evaluating the ESR through said procedures. In particular, it is very important to be able to quickly obtain measured values ​​that correspond to reference standards, which is often not guaranteed by current calculation methods.

[0007] The technical problem of the present invention is to design a system and the related method having structural and functional characteristics capable of overcoming the limitations and disadvantages of the prior art and, in particular, ensuring a very high reliability and precision of the measurement of the erythrocyte sedimentation rate while allowing to maintain a suitable analysis rate. Summary of the invention

[0008] The solution concept forming the basis of the present invention is to develop a system for measuring the erythrocyte sedimentation rate (ESR) of a blood sample in a tube (or a blood cell counting standard tube) by optical measurements, in particular by radiation absorption measurements of the blood sample, wherein the results of the optical absorption measurements are processed by an innovative calculation process based on sedimentation analysis, wherein the experimental reading curve obtained by the above optical absorption measurements is approximated by a theoretical curve of the trapezoidal type suitable for describing the sedimentation process. Thus, by optimizing the trapezoidal type theoretical curve, parameters allowing the estimation of the ESR can be obtained in a very accurate and fast way.

[0009] Based on the solution concept, the aforementioned technical problem is solved by a system for measuring the erythrocyte sedimentation rate in a blood sample, the system comprising: a support for at least one tube, the at least one tube containing the blood sample to be analyzed; a shaking element, which is configured to shake the tube; at least one detection unit, which is configured to perform at least one optical measurement (especially optical absorption measurement) on the blood sample in the tube; a movement device, which is configured to cause a relative movement between the detection unit and the tube during the optical measurement process, so as to illuminate the tube in different parts of the system; a processing unit, which is suitable for processing signals from the detection unit, wherein, based on the signals, the processing unit is configured to: create a reading curve corresponding to the variation of the absorption of the radiation emitted by the detection unit with the relative movement between the detection unit and the tube; based on the reading curve, define an ideal curve of the trapezoidal type suitable for approximating the reading curve; perform an optimization procedure on the ideal curve, thereby generating an optimized ideal curve; and based on the optimization procedure, generate at least one value indicating the erythrocyte sedimentation rate of the blood sample contained in the tube, for example, an output value of a fitting algorithm that can subsequently be converted into an ESR value.

[0010] More particularly, the present invention includes the following additional and optional features, which may be used alone or in combination as necessary.

[0011] According to one aspect of the invention, the system can also include output means configured to output measurement results based on the generated values ​​(obviously, the generated values ​​can already correspond to the displayed measurement results, or can be appropriately processed by the central unit through known formulas or conversions and then output as described above). For example, the ESR value can be shown as an output result, but this is not strictly necessary, and other results related to the calculations performed can also be displayed, possibly together with the ESR.

[0012] According to one aspect of the invention, the processing unit can be configured to perform optimization of the ideal curve by least-squares minimization.

[0013] According to one aspect of the invention, the processing unit can be configured to perform an optimization procedure according to the Levenberg-Marquardt algorithm, wherein the quantity to be minimized is calculated according to the following expression:

[0014] in, L i Represents a reading curve consisting of multiple discrete points, T i represents an ideal curve comprising a number of discrete points (in particular, a number equal to the number of points of the reading curve), and is the weight associated with each point.

[0015] According to one aspect of the present invention, the weight Calculated according to the following expression:

[0016] Among them, p 2 is a parameter suitable for defining the weight of a single point of the reading curve, and ε is the correction factor, and where p 2 is a parameter stored in a storage unit of the processing unit at a default value, and the p 2 Can be modified to improve the optimization of the curve.

[0017] According to one aspect of the present invention, the processing unit can be configured to define an ideal curve of the trapezoidal type to be optimized by creating a parameter vector based on a Cartesian reference system, the parameter vector including: at least two ordinate values, which are easy to identify the two parallel bases of the trapezoid; and at least four abscissa values, which are easy to identify the four vertices of the trapezoid, the processing unit is also configured to fill the vector with two initial abscissa values ​​and four initial ordinate values ​​calculated by processing the reading curve obtained; and after the optimization procedure of the ideal curve, provide an optimized vector, the optimized vector including optimized abscissa values ​​and optimized ordinate values, for example, the optimized ordinate values ​​can correspond to the blood (plasma) level in the tube and the sedimentation level in the tube, respectively, especially when a certain sedimentation time has passed (therefore after a certain time relative to the first baseline reading).

[0018] According to one aspect of the present invention, the processing unit can be configured to calculate the initial ordinate values ​​as the maximum and minimum values ​​of the reading curve, respectively, possibly discarding extreme measurement values, respectively calculate the initial abscissa values ​​corresponding to the vertices on the trapezoid as the first point and the last point of the reading curve that are less than or equal to a certain threshold, and allow the difference in ordinate values ​​and abscissa values ​​indicating the rising / falling phase of the reading curve to be predefined, and based on this definition, calculate the initial abscissa value for calculating the lower vertex of the trapezoid, respectively as the first point of the curve that satisfies the following expression:

[0019] The difference in ordinate values ​​and the difference in abscissa values ​​are parameters stored in a storage unit of the processing unit with default values, and can be modified to improve the optimization of the curve.

[0020] In an example, the processing unit can also be configured to calculate and / or detect specific points of the optimized theoretical curve, and use the points to calculate the ESR, wherein the reference point can be modified by modifying appropriate parameters, such as the point being on the hypotenuse of the trapezoid; thus, the levels of plasma and sediment (vertical coordinates) and the rate at which the sedimentation process occurs (e.g., by taking into account the hypotenuse) can be calculated.

[0021] According to one aspect of the invention, the detection unit can comprise at least one emitter and a corresponding detector arranged to illuminate the tube and to collect the radiation after it has passed through said tube.

[0022] According to one aspect of the invention, the emitter can be an LED configured to emit substantially white light or infrared radiation, or generally any suitable wavelength.

[0023] According to one aspect of the invention, the detection unit can be arranged on the movement device, which is configured to move the detection unit along the longitudinal axis of the pipe, thereby allowing acquisition of a plurality of measurement points along the longitudinal axis.

[0024] According to one aspect of the invention, the support for the tubes can be included in a chain structure that is movable and defines a closed path for the tubes, the chain structure comprising a plurality of supports for a corresponding plurality of tubes that are movable integrally with the chain structure.

[0025] According to one aspect of the invention, the system can comprise four detection units arranged along the chain structure, such that each of the four detection units is configured to analyse the tube moved by the chain structure at a respective instant in time, ie at a different settling time.

[0026] According to one aspect of the invention, the system can also comprise a receiving area for racks, which are apt to receive tubes to be analyzed, and grippers configured to pick up tubes from respective racks and arrange them in supports on said chain structure.

[0027] According to one aspect of the invention, the system can include an image detector configured to acquire an image of the rack within the receiving area, wherein the processing unit is configured to process the image acquired by the image detector and detect the presence of the tube and the position of the tube in the rack based on the processing, and transmit this information to the control device of the gripper, for example, in order to ensure that the gripper moves directly to the position where the tube to be picked up is located.

[0028] According to one aspect of the present invention, the shaking element may include a guide member engaged with an engaging element (e.g., an edge portion) of the chain structure, the chain structure being constructed into a plurality of parts that are connected to one another and configured to rotate around an axis parallel to the direction of advancement of the tube, the shaking element including a moving device (e.g., a motorized device / rotor coupled to a gear or a belt / pulley system) configured to move the guide member, thereby rotating the portion of the chain structure engaged with the guide member.

[0029] The present invention also relates to a method for measuring the erythrocyte sedimentation rate in a blood sample, which comprises the following steps: shaking a tube containing a blood sample to be analyzed; performing at least one optical measurement of the blood sample in the tube by means of at least one detection unit, the optical measurement involving relative movement between the detection unit and the tube, thereby irradiating the tube at different parts; creating a reading curve corresponding to the radiation emitted by the detection unit that changes with the relative movement between the detection unit and the tube; based on the reading curve, defining an ideal curve of a trapezoidal type suitable for approaching the reading curve; executing an optimization procedure for the ideal curve, thereby generating an optimized ideal curve; based on the optimization procedure, generating at least one value indicating the erythrocyte sedimentation rate of the blood sample contained in the tube; and outputting a measurement result based on the generated value.

[0030] According to one aspect of the present invention, the ideal curve may be optimized by a least squares solution according to the Levenberg-Marquardt algorithm, wherein the minimization amount is calculated according to the following formula:

[0031] in, L i represents the reading curve comprising a plurality of discrete points, T i represents an ideal curve consisting of a number of discrete points, and is the weight associated with each point.

[0032] According to one aspect of the present invention, an ideal curve of the trapezoidal type to be optimized can be defined by creating a parameter vector based on a Cartesian reference system, the parameter vector including at least two ordinate values, which are suitable for identifying two parallel bases of the trapezoid; and at least four abscissa values, which are suitable for identifying four vertices of the trapezoid, wherein the method can include the following steps: filling the vector with two initial abscissa values ​​and four initial ordinate values ​​calculated from the processed reading curve; and providing an optimized vector after the optimization procedure of the ideal curve, the optimized vector including optimized abscissa values ​​and optimized ordinate values, wherein the optimized ordinate values ​​correspond to the blood level (plasma level) in the tube and the sedimentation level at the bottom of the tube, respectively.

[0033] The invention also relates to a computer program product for measuring the erythrocyte sedimentation rate in a blood sample, said computer program product comprising code portions facilitating the execution of the above method.

[0034] Features and advantages of the system and method according to the invention will become apparent from the following description of an embodiment thereof, which provides a non-limiting example with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In said drawings: Figure 1 is a general scheme of a system for measuring erythrocyte sedimentation rate according to the present invention; Figure 2 is a stereoscopic view of an optical detection unit according to an embodiment of the present invention; Figure 3 is a top view of a system according to an embodiment of the present invention; Figure 4 is a perspective view of a tube shaking element according to an embodiment of the present invention; Figure 5 is a perspective view of a tube holder and a moving system thereof according to an embodiment of the present invention; Figure 6 shows an example of a reading curve obtained using the system of the present invention; and Figure 7 is an example of a theoretical curve of the trapezoidal type that is easily close to the true reading curve; Figure 8 is an example diagram illustrating the parameter definition of a theoretical curve of trapezoidal type starting from a real reading curve; Fig. 9 An example of a theoretical curve of trapezoidal type superimposed onto a real reading curve obtained by optical absorption measurement is shown; Fig.10 Shows Fig. 9The error minimized by the minimization procedure is highlighted in the curve in; and Fig.11A and Fig. 11B The influence of the choice of weight values ​​in the optimization procedure of a theoretical curve of trapezoidal type is shown. DETAILED DESCRIPTION

[0036] With reference to the accompanying drawings, Figure 1 A system for measuring the erythrocyte sedimentation rate in a blood sample according to the invention is indicated overall and schematically, the system operating according to the associated method.

[0037] It should be noted that the drawings represent schematic views and that the drawings are not always drawn to scale but are drawn in order to emphasize important features of the invention. Furthermore, in the drawings, various elements are represented in a schematic manner and their shapes may vary depending on the desired application. It should also be noted that in the drawings, the same reference numerals refer to elements that are identical in shape or function. Finally, specific features described in relation to the embodiments shown in the drawings may also be used in the other embodiments shown in the drawings.

[0038] It should also be noted that unless explicitly stated otherwise, the process steps described may also be reversed where necessary.

[0039] The present invention provides a system and method for measuring the erythrocyte sedimentation rate (ESR) of a blood sample in a tube (identified by reference numeral P, not limited to a specific type). Tube P can actually be a common blood cell counting tube, but it should be noted that the inventive aspects described herein are not limited to the above-mentioned tube type.

[0040] In its most general aspect, the present invention enables to estimate the ESR in a very accurate way, by means of an innovative analysis procedure of the collected data, thereby calculating the parameters of interest, and to estimate the validity of the measurements made, by means of an innovative analysis of the process dynamics. The following description thus illustrates the basic steps for defining the system 1 and the associated method which are the object of the present invention.

[0041] In order to allow the execution of the operations described below, the system 1 comprises a processing unit (identified by the reference numeral C), which comprises a specific memory unit MEM, which is appropriately programmed and designated for its management, for automatic control and for analyzing the measurement data. For example, the processing unit C can be a computerized unit integrated inside the system or outside the system and operatively connected to the system. In addition, it should be noted that the processing unit C can be a single unit or can include a plurality of local and / or remote units, which may communicate with each other, each of which is designated for executing a specific operation. Thus, the processing unit C is capable of controlling the system 1 to obtain the desired blood sample analysis. In any case, the invention is not limited to the architecture for the control unit C, which can generally be any suitable computerized unit, which includes one or more units according to the needs and / or environment.

[0042] It should also be noted that the term "system 1" refers to a general analytical device provided with a suitable housing and comprising a plurality of components cooperating with each other to obtain the desired calculation procedure, and that the device is in no way limited to a particular type.

[0043] In any case, the invention will be explained below with reference to a specific example in which a tube P containing a sample to be analyzed is moved along the various reading stations by a chain system, even though, as mentioned above, the teaching described herein is not limited to this embodiment and is also applicable to many other types of devices with different configurations.

[0044] See also Figure 1 In the most general form, the system 1 comprises a support 2 for receiving at least one tube P containing a blood sample to be analyzed. In particular, the system 1 comprises a plurality of supports 2 for receiving a corresponding plurality of tubes 2 .

[0045] The system 1 also comprises a shaking element 3 configured to shake the tube P, thus allowing the sedimentation process to be subsequently evaluated. The shaking element 3 is not limited to a specific configuration and depends mainly on the type of support used to house and possibly move the tube. An example will be given later in which the shaking element 3 cooperates with a movable chain structure on which the tube P is arranged, but the scope of protection is not limited to said configuration; indeed it should be noted that when the tube is arranged on other types of supports, such as a circular plate, the shaking device is significantly different and adapted to the specific case.

[0046] Then, there is at least one optical detection unit (indicated with reference numeral 4 , also referred to below as reading unit) which is configured to perform optical measurements on the blood sample contained in the tube P, in particular optical absorption measurements.

[0047] In particular, the detection unit 4 comprises at least one emitter 4' and an associated detector 4", both arranged to illuminate the tube P with electromagnetic radiation and to collect the radiation after it has passed through said tube P. The emitter 4' is preferably an LED configured to emit substantially white light, so that the presence of labels on the tube P (and other external factors) do not affect the measurement. Obviously, the emitter 4' can emit radiation of any suitable wavelength, without being limited to a specific value.

[0048] The detection unit 4 thus allows obtaining, by absorption measurement of the blood sample in the tube P, a reading curve which will serve as a starting point for the calculation procedure of the invention, as will be described in detail below.

[0049] Suitable moving means 4m are also provided, which are configured to cause a relative movement between the detection unit 4 and the tube P during the light absorption measurement, so as to illuminate different parts of the tube P and obtain a plurality of n discrete measurement points (here and below denoted by ) forming a reading curve. i In particular, a single point on the reading curve is identified as i In particular, the moving device 4m is configured to cause a stepping movement of the detection unit 4. As will be explained in detail below, the obtained reading curve represents the detected light intensity as a function of the reading step (the step may be switchable at a time instant).

[0050] like Figure 2 , which shows a non-limiting example of a reading unit 4, the moving means 4m can comprise a trolley moved by a suitable motor unit 4mu (comprising its own control drive board, also indicated by reference numeral 4mu). The detection unit 4 (in particular the emitter 4' and the detector 4'') are therefore arranged on the trolley 4m, thereby allowing a movement of the detection unit 4 in a direction substantially parallel to the longitudinal axis HH of the pipe P, thus allowing the acquisition of a plurality of measurement points along said longitudinal axis HH. In this example, the motor unit 4mu moves a worm 4v which in turn causes the trolley 4m to move. Furthermore, in Figure 2 In the example of FIG. 4 , both the emitter 4 ′ and the detector 4 ″ are moved by a carriage 4 m which is suitably shaped to allow the tube (not in FIG. 4 ) to be moved. Figure 2 The housing (shown) is located substantially in the center of the trolley. All of the above components are supported by the support member 4s, so the support member 4s serves as the load-bearing structure of the detection unit 4.

[0051] However, it should be noted that the present invention is not limited to the configuration of the detection unit 4 at all, and thus any other suitable configuration, for example with respect to the movement of the detectors / emitters or their arrangement, may be adopted. Therefore, any suitable structure of the detection unit (and any emitter / detector components) falls within the scope of the present invention.

[0052] In an embodiment of the present invention, when there is more than one reading unit, the relative calibration (alignment) between the various reading units can be performed by acting on an adjustment device to adjust the position of the emitter / detector relative to a test tube (not shown in the figure) so as to adapt the reading curve, especially the detected light intensity, relative to the reference provided by the test tube.

[0053] Reference now Figure 3 In a particular embodiment of the invention, as previously described, the system 1 comprises a chain structure 10 on which the support 2 for the pipes P is formed. The chain structure 10 is movable and defines a closed path for the pipes P, which is substantially placed in a horizontal plane, for example parallel to the surface on which the system 1 is located.

[0054] The chain structure 10 comprises a plurality of parts connected to each other, each part providing support for at least one respective pipe p. In an embodiment, the chain parts 10 are connected to each other by ball joints and are rotatable relative to each other.

[0055] Thus, the support 2 for the tube P is contained in a chainable structure 10 that is mobile and defines a closed analytical path for said tube P, said chain structure 10 comprising a plurality of supports for a corresponding plurality of tubes P that therefore move integrally with the chain structure.

[0056] In a particular embodiment, the system 1 comprises at least two detection units, preferably four detection units, arranged along the chain structure 10 , such that each of the four detection units is configured to analyse the pipe P moved by the chain structure 10 at a specific moment in time, ie at a specific settling time.

[0057] In particular, the first detection unit 4a takes a first reading curve (thus a baseline reading) shortly after shaking the tube, while the second reading unit 4b, arranged at a different position along the chain 10, takes measurements after a certain settling time (e.g., after 20 minutes) of the tube P. There can also be optionally two further detection units (e.g., units 4c and 4d), which are arranged at intermediate points of the chain structure 10, so that measurements can also be taken at intermediate time instants (e.g., at 12 minutes and 17 minutes).

[0058] In summary, Figure 3 In the embodiment of the present invention, the analysis module M of the system 1 comprises a chain structure 10, which can have, for example, eighty-nine slots (mash) in which the tubes P are inserted and which are free to rotate at their joints. The chain 10 rotates clockwise in the analysis module by two traction wheels 10t driven by a motor unit 10m, transferring the tubes P to the shaking unit and then to the photoelectric unit.

[0059] As previously mentioned, before taking a reading of the blood sample, the tube is shaken by the shaking element 3. Figure 4 As shown, the shaking element 3 may include a guide 3g that engages with an engaging part or element (such as a side rail) of the chain structure 10, which, as shown above, is constructed with parts connected to each other and is constructed to rotate around an axis YY parallel to the direction of advancement of the pipe p. In addition, there is a suitable motorized device 3m, which is configured to move the guide 3g through a suitable gear system (or through a belt / pulley system) so that the part of the chain structure 10 that engages with the guide 3g rotates. In essence, a guide pad system is thus created that allows the pipe p to be shaken. As shown Figure 4 As shown in the non-limiting example of , the guide pad 3g is formed by two flanges 3f (for example circular) which are integrally moved by a motorized device 3m.

[0060] In some alternative embodiments, the support 2 and the rocking device 3 can also be the same component that performs both functions, the present invention not being limited in fact to a specific structural configuration of said components.

[0061] Reference now Figure 5 In some embodiments, the system 1 can further comprise a gripper 6, which is moved by a suitable device 6m (controlled by a suitable control device) and is configured to grip the tube P from a corresponding rack (not shown in the figure) and arrange the tube P in the support 2 on the chain structure 10.

[0062] There can also be a receiving area for racks containing the various tubes P to be analyzed, and there can also be control means (for example included in the processing unit C, but also as a separate unit) configured to verify the correct positioning of the various racks in said receiving area. There are also image detectors (for example photo / video cameras) that acquire images of the racks, and after processing said images, the processing unit C is able to verify the presence of the tubes in the racks and their positions, so as to be able to communicate this information to the control means of the gripper 6. Thus, the gripper is moved precisely to the position where the tubes P are located, so as to grip the tubes P effectively.

[0063] In any case, as described above, and whatever the support and the mode of movement of the tube P, the invention envisages processing, by means of a processing unit C, the signal from the detection unit 4 (indicated by the reference numeral Sgn), which contains information about the radiation absorbed by the blood sample, for the creation and subsequent processing of a reading curve.

[0064] Specifically, based on the signal Sgn, the processing unit C is configured to first create a reading curve (indicated by reference numeral L) corresponding to the real absorption of the radiation emitted by the detection unit 4. As mentioned above, the reading curve L comprises a plurality of n measurement points, which are obtained according to the relative movement between the detection unit 4 and the tube P (e.g. Figure 2 It is obtained by changing the stepping motion of the motorized trolley 4m).

[0065] As mentioned above, the reading curve L is taken immediately after shaking the tube and then after a certain sedimentation time (up to 20 minutes) and thus indicates the sedimentation process in the tube.

[0066] Due to the sedimentation process that occurs after shaking the tube P, the reading curve L shows the change in the measured radiation intensity during the transition from plasma to sediment, specifically, the higher absorption of radiation at the sediment, resulting in a decrease in the detected light intensity. Then, as Figure 6 As shown in the example of , there is a (more or less constant) plateau down to the bottom of the tube, Figure 6 An example of a reading curve L obtained by the detection unit 4 according to an embodiment of the present invention is shown.

[0067] More specifically, as shown above, the reading curve L is obtained immediately after shaking the tube (refer to Figure 6 left curve), followed by a specific settling time (up to twenty minutes, Figure 6 By comparing the two, the information about the sedimentation process of the blood sample in the tube P can be shown.

[0068] More specifically, in the readings after the reference reading, i.e. after the deposition time, it can be noted that the decrease in intensity due to the absorption of radiation occurs at t0 after the point t0 at which the first reference reading occurs due to the deposition formed after a certain period of time. 1 Point (such as Figure 6 As shown in the example, Figure 6 It can be seen that the plateau formed by blood cell absorption is shifted due to sedimentation, that is, after a certain period of time, separated from the rest of the plasma in which no significant absorption of radiation occurs), thus providing useful information for estimating the erythrocyte sedimentation rate.

[0069] Therefore, the control unit C comprises processing means (eg suitable software modules) for processing the required measured values ​​according to a predefined protocol, thus allowing the blood sample to be analyzed.

[0070] Based on this behavior of the reading curve L, the applicant has conducted research to develop an innovative method for analyzing this curve, which involves defining an ideal curve T of the trapezoidal type, which can appropriately approximate the real behavior, based on the reading curve L. Due to the above-mentioned trends, the use of the trapezoidal curve is very useful because it is an excellent starting point for analyzing the acquired reading curve.

[0071] Obviously, the term "trapezoidal" or "trapezoidal type" is to be understood in a broad sense and can denote a complete trapezoid or only a portion of a trapezoid (for example comprising only one of the two hypotenuses), thus generally referring to a shape (trend) comprising two sides that are substantially parallel and at least one hypotenuse between the two sides, according to a Cartesian reference system; in other words, an ideal curve of the trapezoidal type can be a function having two portions that are substantially constant relative to the ordinate and at least one portion having a certain slope between them (in the following example, a case will be presented in which there will be a trapezoidal function with two opposite hypotenuses representing a decrease and an increase in the detected intensity).

[0072] Therefore, the calculation process of the present invention is based on the use of trapezoidal trend T (such as Figure 7 The function of (as shown) approximates the reading curve L, and the trapezoidal trend T represents the ideal trend of the curve.

[0073] In other words, once a reading curve L containing n points (i.e. Li, i=0, ..., n-1) is obtained experimentally, a curve T is defined that also contains n points (i.e. Ti, i=0, ..., n-1). As mentioned before, the discretization of the measurement points is due to the number of steps of the trolley that moves the detection unit 4: each step corresponds to a measurement point.

[0074] The calculation process performed by the processing unit C for determining the best approximation of the reading curve L is based on a least squares optimization algorithm; thus, it provides for the determination of an ideal approximation curve T, which is used as the starting point for an optimization process, after which an excellent trapezoidal curve (indicated by the reference numeral T') is generated that better represents the real reading curve L. Finally, only said excellent trapezoidal curve T' is used (instead of the original reading curve L) to calculate the reading parameters.

[0075] The processing unit C thus comprises a module for reading and analyzing the reading curve L, susceptible of a code for analyzing said absorption curve by means of the above-mentioned approximation and subsequent optimization, outputting for example the identification of two basic values ​​(also called deflection points) corresponding to the level of the blood sample and to the level of the associated sediment deposited at the bottom of the tube P, as well as an estimate of the validity of the measurements and analyses performed.

[0076] The rest of this instruction manual will explain all the details of this process.

[0077] As mentioned above, in this calculation process, a preliminary ideal trapezoidal curve T is first defined. More specifically, the processing unit C is configured to define the ideal curve T of the trapezoidal type to be optimized by creating an initial parameter vector V according to the Cartesian reference system xy, the initial parameter vector V containing two ordinate values ​​(in Figure 7 and Figure 8 y 1 and 2 ) and four horizontal coordinate values ​​(called x 1 、x 2 、x 3 and x 4 ).

[0078] In other words, six parameters are identified: the ordinates of the two horizontal sections and the abscissas of the four vertices of the trapezoid. Figure 7 As shown, these parameters form a vector V of 6 elements, namely Therefore, the parameter x 1 and x 4 represents the coordinates of the utmost point of the ideal trapezoid, and the parameter x 2 and x 3 The processing unit C is therefore programmed to fill the vector V with the above four abscissa values ​​and two initial ordinate values, processing the real reading curve L obtained by optical measurement of the blood sample in the tube P.

[0079] Even more specifically, when defining the theoretical curve T that initially approximates the true curve, the initial values ​​of the ordinates (i.e., y1 and y2) are calculated as the minimum and maximum values ​​of the reading curve L, respectively, and in order to avoid artifacts at the beginning and end of the reading curve L, the extreme measured values ​​of the reading curve L may be discarded, especially when calculating y2. As an example only, it is possible to define the value identified as " head "and" tail " has two parameters equal to 10 and 20 respectively. Consider removing the first one head element and the last tail The part of the curve L where the reading is obtained after the element. In the formula: , where i is between 0 and N, , where i is greater than tail and i is less than N-tail; then define , that is, the width of the variation between the maximum and minimum values, such as Figure 8 As shown, this is an exemplary graph for calculating the initial value of the theoretical curve T starting from the reading curve L.

[0080] In addition, the initial abscissa values ​​(x1 , x 4 ) are calculated as the first and last points of the reading curve L that are less than or equal to a certain threshold value s. More preferably, it is found that point x 1 is the curve L below The first point of the given threshold. Similarly, point x 4 is the last point on the curve whose value is less than s.

[0081] To determine the point x 2 and x 3 , defines two quantities: the displacement along the horizontal axis , and the value along the ordinate (also called the “slope”) Then determine the descending part where the reading curve L is The range has dropped to several points In the formula, find the first i The value of i points), so the formula is . In this way, we find i , which may be increased by any value (in the non-limiting example, it is equal to 5), defines Figure 8 Points in x a , thus we get A similar process is performed on the right side of the curve to find the last (rightmost) i The value of (the reversal of sign is due to the fact that the curve increases), then define (that is, similarly to what was done for the left side of the curve, minus an arbitrary value chosen to be equal to 8, but not limited to this) and finally obtaining .

[0082] In other words, to define the point x that represents the lower vertex of the trapezoid 2 and x 3 , the difference in ordinate values ​​(i.e. slope) representing the increase / decrease phase (rise / fall) of the reading curve L is predefined ) and the difference between the horizontal coordinate value is the displacement ), according to the above definition, the horizontal coordinate value x 2 and x 3 are calculated as the first point of the curve that satisfies the above expression i ,Right now:

[0083] parameter and The optimal selection criterion for can be defined as follows: the ratio Represents the minimum slope of the curve necessary to identify a decreasing or increasing portion of the same curve. Therefore, a too high value of the ratio will result in an inability to determine the slope, since no part of the curve will have the required slope. Conversely, a too low value will result in the erroneous identification of local oscillations that occur at the beginning or end of the same curve. Specifically, for A value of , too low a value will result in the detection of local noise effects on the curve, while a too high value requires (in order to have enough ratio value ) cannot be achieved within the variable range of the curve Therefore, The value of and ratio determination.

[0084] parameter and The above parameters are stored in the memory unit MEM of the processing unit C with default values ​​and can possibly be modified to improve the process. In any case, it can be observed that the above parameters do not have a negative impact on the final result if they are chosen within a fairly large specific range.

[0085] In addition, the parameter for discarding extreme values ​​of the reading curve L head and tail The default values ​​are also stored in the memory unit MEM, and they can also be adjusted as required.

[0086] Then execute to verify x 1 , x a , x b , x 4 These four points are in ascending order. If not verified, the points are immediately x a and x b Repositioned in x 1 The left side and x 4 If the above control is performed, x a Still greater than x b , then both points are relocated to x 1 and x 4 The midpoint between.

[0087] Fig. 9The overlap between the starting theoretical curve T calculated above and the reading curve L is shown. The ordinate values ​​correspond to the change in light intensity, and the abscissa values ​​correspond to the displacement of the LED, for example to the number of steps of the trolley 4 m that moves said LED; in particular, in this non-limiting example, the reading (and the related approximate curve) starts from a level where there is no blood in the tube (upper flat line), and then the intensity changes from level L 0 It begins to descend with a specific slope, reaches a plateau which is more or less regular and indicative of the absorption of sediment, continues down to the bottom of the tube, and then rises again, as before.

[0088] Once the theoretical starting curve T has been defined in the above-described manner, the processing unit C performs an optimization procedure of this theoretical curve by adapting the ideal curve T as accurately as possible to the real reading curve L, the optimization being based on a least squares solution.

[0089] Specifically, the above optimization first involves the cost functional C s The cost function to be minimized is calculated as the weighted sum of the sum of the squares of the differences between the theoretical curve T and the reading curve L (as shown above, some measurement points may have been removed from the reading curve L in advance).

[0090] in, For the curve L i and T i The weight associated with each point in .

[0091] like Fig.10 As highlighted in FIG. 1 , the cost function Cs is connected to the area A included between the curve T and the curve L.

[0092] From a mathematical point of view, the measurement of area A is actually equal to However, minimizing a functional cost involving absolute values ​​has numerical problems, so it is preferred to minimize the quadratic cost function represented by the above formula. In addition, the weight The addition of also allows areas of the curve to be weighted differently if desired.

[0093] Specifically, the weight associated with each of the elements of the sum is defined in terms of the width of the curve, scaled to a value between zero and one. More specifically, the weight is calculated based on the following expression:

[0094] Among them, p 2 It is easy to define a single point of the reading curve Li The weight parameter, ε is an optional correction factor, value 10 -9 is assigned to this correction factor to avoid S In the definition of and may occur).

[0095] For the other variable parameters seen above, p 2 It is also a parameter stored in the memory unit MEM of the processing unit C with a default value, p 2 Can be modified to improve optimization of the curve.

[0096] According to an embodiment of the present invention, the cost function C is minimized by the Levenberg-Marquardt algorithm. S , thereby obtaining the optimal value of the above-mentioned vector V (that is, obtaining the optimized vector V'), thereby providing an optimized trapezoidal theoretical curve, represented as T'.

[0097] In other words, the processing unit C is configured to perform an optimization of the starting trapezoidal curve, for example based on the Levenberg-Marquardt procedure, in order to provide an optimized theoretical curve T' and thus obtain the desired optimization parameters .

[0098] It can also be noted that according to the aforementioned weights The definition of p 2 When it is equal to 1, we get , that is, all errors are weighted in the same way, and when p 2 When the value of is less than one, the smaller values ​​of the curve are considered to be more important in the minimization process ( ,therefore ).when Li With values ​​close to the upper limit, ,therefore .

[0099] Generally speaking, between 0 and 1 p 2 The value of is determined according to the relative importance people attach to the errors in the upper and lower parts of the curve; by changing the parameter p 2 , it is indeed possible to observe deformation of the hypotenuse of the trapezoid, especially if the curve is irregular (having a varying slope along the decreasing or increasing parts). In this case, the hypotenuse tries to follow one or the other of the two slopes based on the same parameters. As mentioned earlier, for very low p 2For values ​​close to 1, the optimized theoretical curve will be close to the lower part of the curve L, ignoring the upper part of the curve, while for values ​​close to 1, the trapezoid will also follow the reading curve L in the upper part.

[0100] Fig.11A and Fig. 11B shows an example of the behavior of the system when the curve on the right has a change in slope. Note that for small p 2 The optimized theoretical curve T′ follows the area in the lower region of the graph with high accuracy, and T′ moves away from the reading curve L when the reading curve L approaches its maximum value. On the contrary, when p 2 When the value of is equal to one, all errors are weighted in the same way, so the trapezoid follows the average slope of the detected increases.

[0101] According to the invention, the estimation of the output value is carried out starting from the calculated trapezoidal function. Based on the behavior of the curve, it is possible to envisage obtaining the output value (i.e. the value of the calculated ESR) based on the hypotenuse of the trapezoid, i.e. on the descent of the curve, as can be seen, after a certain settling time, the descent occurs at a different moment than the reference curve.

[0102] Specifically, calculate the point on the first hypotenuse of the trapezoid, that is, The corresponding vertical coordinate The corresponding value of x is called xo, which is the output value of the program. Therefore, p 1 The value of allows you to select the height of the point at the top vertex of the trapezoidal hypotenuse (when p 1 = 0) and the lower vertex (when p 1 =1) is variable. Fig. 9 and Fig.10 In the figure, two dots represent p 1 The value of is equal to the output value corresponding to 0.6 and 0.8, while the square represents p 1 = 0.9. Observe how these points are located on the descending side of trapezoid T (rather than on curve L). In other words, the parameter p 1 Used to position the reference point at the most appropriate point on the bevel.

[0103] In other words, the parameter p 1The percentage of travel along the y-axis of the curve (the hypotenuse of the trapezoid) that represents the detected blood level, which can move the point defining the "result". Therefore, the output value of the algorithm (connected to the ESR) therefore corresponds to the coordinate x of a point contained between the extreme values ​​of the inclined layer. p 1 =0, the output corresponds to an extreme value, when p 1 = 1, the output corresponds to the other extreme value. From a practical point of view, this parameter can be calibrated and calculated as needed to obtain a value that can be compared with other values. Since the reading curve L is usually smoother than a trapezoid (see Fig.10 ), considering that taking the lower vertex of the inclined layer as the output value will cause a smaller error, the parameter is introduced p 1 , which takes into account the existence of area A.

[0104] Hence, in an embodiment of the invention, the processing unit C is further configured to provide an estimate of the quality of the performed measurement, for example based on a comparison with a reference.

[0105] In summary, after the optimization procedure of the ideal curve T, an optimized vector V' is output, which includes two ordinate values ​​and four optimized abscissa values, wherein the optimized ordinate values ​​(y1', y2') correspond to the blood levels L in the tube P, respectively. 0 and the settlement level L 1 : The determination of these two quantities allows, through known correlations, to determine the ESR or, in general, a value of interest.

[0106] Thus, based on the above-described process, it is possible to generate an indication of the sedimentation rate of the erythrocytes of the blood sample contained in the tube P. In other words, the invention allows, through the aforementioned process of fitting the curve with a suitable approximation of the real absorption curve, to obtain the essential parameters related to the sedimentation process in order to estimate the ESR in a very precise manner; the calculated values ​​can be processed through known correlations, as described above, in order to obtain the ESR.

[0107] In an embodiment of the present invention, the system 1 further comprises an output device (indicated by reference numeral Out) configured to output the measurement results based on the generated values. It should be noted that the present invention is not limited to a specific output device, and the output device can include any suitable device from a display (e.g., a touch screen display) for displaying the measurement results to a printer for generating a paper report. For example, the calculated ESR or a value related thereto can be displayed.

[0108] In view of the above, it is obvious that the present invention also relates to a method for measuring the erythrocyte sedimentation rate in a blood sample, and a computer program product for measuring the erythrocyte sedimentation rate in a blood sample, the method comprising all the above steps of the calculation process, and the computer program product comprising a code portion that is easy to implement the above method.

[0109] In summary, the present invention is therefore able to brilliantly overcome technical problems, provide the above-mentioned system and method, and solve all the shortcomings of the prior art.

[0110] Advantageously, the method innovatively and efficiently processes experimental results obtained by optical absorption measurements in blood samples, providing a highly accurate diagnostic tool and method for estimating ESR.

[0111] In order to implement the calculation process of the invention, it is sufficient to provide the processing unit with a limited number of parameters, which relate to the initial estimation of the theoretical curve (i.e. the definition of the extreme values ​​and the slope of the reading curve), the optimization procedure (i.e. the weights associated with each point on the curve), and the position of the reference point on the theoretical curve. Said parameters are set in the system memory with default values, but can be modified according to specific needs.

[0112] All technical results described in this article were obtained with a simple experimental system and fast measurements (both from the point of view of analysis time (the last reading took place after approximately 20 min) and from the point of view of calculation time), in which measurement errors were greatly reduced, if not completely eliminated.

[0113] The calculation process of the present invention does have great advantages, and comparison with the Westergreen standard method shows very high correlation (eg, Pearson correlation coefficient and Spearman correlation coefficient are equal to r=0.909 and r=0.914, respectively), thus proving the effectiveness of the scheme.

[0114] Obviously, in order to meet accidental and specific needs, those skilled in the art can make various modifications and changes to the above-mentioned systems and methods, and these modifications and changes are all included in the protection scope of the present invention defined by the claims.

Claims

1. A system (1) for measuring erythrocyte sedimentation rate in a blood sample, comprising: - a support (2) for at least one tube (P) suitable for containing a blood sample to be analyzed; - a shaking element (3) configured to shake the tube (P); - at least one detection unit (4) configured to perform at least one optical measurement on said blood sample in said tube (P); - a movement device (4m) configured to cause a relative movement between the detection unit (4) and the tube (P) during the optical measurement; and a processing unit (C) adapted to process the signal (Sgn) coming from the detection unit (4), wherein, based on the signal (Sgn), the processing unit (C) is configured to: - creating a reading curve (L) corresponding to the variation of the absorption of the radiation emitted by the detection unit (4) with the relative movement between the detection unit (4) and the tube (P); - based on said reading curve (L), defining an ideal curve (T) of trapezoidal type suitable for approximating said reading curve (L); - performing an optimization procedure on said ideal curve (T) so as to generate an optimized ideal curve (T'); and - generating, based on said optimization procedure, at least one value indicative of the erythrocyte sedimentation rate of said blood sample contained in said tube (P), The system (1) further comprises an output device (Out), which is configured to output a measurement result based on the generated value.

2. The system (1) according to claim 1, wherein: The processing unit (C) is configured to perform an optimization of the ideal curve (T) by a least squares solution.

3. The system (1) according to claim 2, wherein: The processing unit (C) is configured to perform the optimization procedure according to the Levenberg-Marquardt algorithm, wherein the quantity to be minimized is calculated according to the following expression: in, L i represents the reading curve (L) comprising a plurality of discrete points, T i represents the ideal curve (T) comprising a plurality of discrete points, and is the weight associated with each point.

4. The system according to claim 3, wherein: The weight Calculated according to the following expression: where p2 is a parameter suitable for defining the weight of a single point of the reading curve (L), and ε is the correction factor, and Therein, p2 is a parameter stored in a memory unit (MEM) of the processing unit (C) with a default value and is susceptible to being modified to improve the optimization of the curve.

5. System (1) according to any one of the preceding claims, wherein: The processing unit (C) is configured to define an ideal curve (T) of the trapezoidal type to be optimized by creating a parameter vector (V) according to a Cartesian reference system (x, y), the parameter vector (V) comprising: - at least two ordinate values ​​(y1, y2) which facilitate identification of the two parallel bases of said trapezoid; and - at least four abscissa values ​​(x1, x2, x3, x4) which easily identify the 4 vertices of the trapezoid, The processing unit (C) is further configured to: - filling said vector (V) with two initial abscissa values ​​and four initial ordinate values ​​calculated by processing the obtained reading curve (L); and - after the optimization procedure of the ideal curve (T), an optimized vector (V') is provided, said optimized vector (V') comprising optimized abscissa values ​​(x1', x2', x3', x4') and optimized ordinate values ​​(y1', y2'), Wherein, the optimized ordinate values ​​(y1', y2') correspond to the plasma level in the tube (P) and the sedimentation level in the tube (P), respectively.

6. The system (1) according to claim 5, wherein: The processing unit (C) is configured to: - calculating the initial ordinate values ​​(y1, y2) as the maximum and minimum values ​​of the reading curve (L), respectively, possibly discarding extreme measured values; - respectively calculating the initial abscissa values ​​(x1, x4) corresponding to the vertices on the trapezoid as the first point and the last point of the reading curve (L) that are less than or equal to a certain threshold value (s); and - Allows to define the difference (Δy) in the ordinate values ​​of the rising / falling phase of the reading curve (L) s ) and the difference between the horizontal coordinate value (Δx s ), and based on the definition, calculate the initial abscissa values ​​(x2, x3) used to calculate the lower vertex of the trapezoid as the first point of the curve that satisfies the following expressions: Among them, the difference in the ordinate value (Δy s ) and the difference between the abscissa values ​​(Δx s ) are parameters stored in a memory unit (MEM) of the processing unit (C) at default values ​​and which are susceptible to being modified to improve the optimization of the curve.

7. System (1) according to any one of the preceding claims, wherein: The detection unit (4) comprises at least one emitter (4') and a corresponding detector (4"), which are arranged to illuminate the tube (P) and to collect the radiation after it has passed through the tube (P), wherein the emitter (4') is an LED configured to emit white light or infrared radiation.

8. System (1) according to any of the preceding claims, wherein: The detection unit (4) is arranged on the moving device (4m), which is configured to move the detection unit (4) along the longitudinal axis (HH) of the pipe (P), thereby allowing a plurality of measurement points to be acquired along the longitudinal axis (HH).

9. System (1) according to any one of the preceding claims, wherein: The support (2) of the tube (P) is included in a chain structure (10), the chain structure (10) is movable and defines a closed path for the tube (P), the chain structure (10) comprises a plurality of supports for a corresponding plurality of tubes, the tubes being movable integrally with the chain structure (10).

10. The system (1) according to claim 9, comprising four detection units arranged along the chain structure (10), such that each of the four detection units is configured to analyze the tube (P) moved by the chain structure (10) at a respective time instant.

11. The system (1) according to claim 9 or 10, further comprising: - a receiving area for racks, each rack being adapted to receive a tube to be analyzed (P); - grippers (6) configured to pick up the tubes (P) from the respective racks and arrange them in supports (2) on the chain structure (10); and - an image detector configured to acquire an image of the racks within the receiving area, The processing unit (C) is configured to process the image acquired by the image detector, and based on the processing, detect the presence of the tube (P) and the position of the tube (P) in the rack, and transmit the information to the control device of the gripper (6).

12. The system (1) according to any one of claims 9 to 11, wherein: The shaking element (3) includes a guide (3g) engaged with an engaging element of the chain structure (10), the chain structure (10) being constructed into a plurality of parts, the plurality of parts being connected to each other and being configured to rotate around an axis (YY) parallel to the advancing direction of the pipe (P), the shaking element (3) including a moving device (3m), the moving device (3m) being configured to move the guide (3g) and thereby rotate the part of the chain structure (10) engaged with the guide (3g).

13. A method for measuring erythrocyte sedimentation rate in a blood sample, comprising the steps of: - shaking the tube containing the blood sample to be analyzed (P); - performing at least one optical measurement of the blood sample in the tube (P) by means of at least one detection unit (4), the optical measurement involving a relative movement between the detection unit (4) and the tube (P); - creating a reading curve (L) corresponding to the variation of the absorption of the radiation emitted by the detection unit (4) with the relative movement between the detection unit (4) and the tube (P); - based on said reading curve (L), defining an ideal curve (T) of trapezoidal type suitable for approximating said reading curve (L); - performing an optimization procedure on said ideal curve (T) so as to generate an optimized ideal curve (T'); - generating, based on said optimization procedure, at least one value indicative of the erythrocyte sedimentation rate of said blood sample contained in said tube (P), and - outputting a measurement result based on said generated value.

14. The method according to claim 13, wherein: The optimization of the ideal curve (T) is performed by least squares solution according to the Levenberg-Marquardt algorithm, wherein the minimization quantity is calculated according to the following expression: in, L i represents the reading curve (L) comprising a plurality of discrete points, T i represents the ideal curve (T) comprising a plurality of discrete points, and is the weight associated with each point.

15. The method according to claim 13 or 14, wherein: The ideal curve (T) of the trapezoidal type to be optimized is defined by creating a parameter vector (V) according to a Cartesian reference system (x, y), said parameter vector (V) comprising: - at least two ordinate values ​​(y1, y2) suitable for identifying two parallel bases of said trapezoid; and - at least four abscissa values ​​(x1, x2, x3, x4) suitable for identifying the four vertices of the trapezoid, The method comprises the following steps: - filling said vector (V) with two initial abscissa values ​​and four initial ordinate values ​​calculated by processing the obtained reading curve (L); and - after the optimization procedure of the ideal curve (T), an optimized vector (V') is provided, said optimized vector (V') comprising optimized abscissa values ​​(x1', x2', x3', x4') and optimized ordinate values ​​(y1', y2'), Wherein, the optimized ordinate values ​​(y1', y2') correspond to the plasma level in the tube (P) and the sedimentation level in the tube (P), respectively.

16. A computer program product for measuring erythrocyte sedimentation rate in a blood sample, the computer program product comprising code portions adapted to carry out the method according to any one of claims 13 to 15.

Images