This application is related to and claims the benefit of Italian Patent Application No. 102020000031391, filed on Dec. 18, 2020, the contents of which are herein incorporated by reference in their entirety.
The present disclosure relates to a blood filtering device, particularly for hemodialysis and/or haemofiltration apparatuses.
As is well known, there are many medical procedures, mostly therapeutic, during which a patient's blood is subjected to filtration to remove unwanted substances present in the blood or to separate certain components of the whole blood.
The two main mechanisms with which it is possible to purify the blood and thus remove substances and/or components that need to be removed are the hemodialysis, which mainly uses the diffusion principle, and the haemofiltration, which uses the convection principle.
In therapeutic hemodialysis treatments, the blood is taken from the patient, made to flow through a dialyzer filter inside which it comes into contact with a semi-permeable membrane that allows the selective passage, mainly by osmotic diffusion, of the toxic substances to be removed from the blood, and then it is returned to the patient.
In therapeutic haemofiltration treatments, such as therapeutic apheresis, it is provided for the blood taken from a patient in extracorporeal circulation to be filtered through a membrane that separates a specific blood component by convection thanks to a pressure difference, before being reintroduced into the patient, and if necessary supplemented with a solution compatible with the patient's own blood as a replacement for the removed component.
There are also so-called therapeutic haemodiafiltration treatments in which the two mechanisms of dialysis and filtration are adopted simultaneously.
However, the blood filtering devices used in hemodialysis and in haemofiltration are not free from drawbacks, one of which is that the incorrect functioning thereof can have very serious consequences for the patient undergoing the therapeutic treatment.
In addition, the blood filtering devices commonly used in the clinical setting do not allow a real-time monitoring of the performance of the filter itself, not only to ensure its integrity, but also to make sure that the therapeutic treatment is taking place effectively.
To date, the only check that, according to current clinical protocols, is carried out on hemodialysis and haemofiltration apparatuses is to detect traces of blood in the filtrate fraction and/or in the dialysis fluid coming out of the filter, as the presence of blood indicates a rupture of the filter membrane. With regard to this, there are in fact international standards, such as the IEC 60601-2-1 standard, which define a maximum blood threshold, expressed in terms of flow rate, which can be detected at the filter outlet. Beyond this threshold, an alarm is activated to interrupt the therapeutic treatment and restore the filter.
This threshold, in the case of hemodialysis, is currently equal to a flow rate of 0.35 ml/min of blood in flows of dialysis fluid having flow rates of 800 ml/min. This corresponds to a blood volume of 218 μL in 500 mL of dialysis fluid, i.e. a dilution ratio of 1 in 2285.
This check is currently carried out using optical sensors (spectrophotometers) to detect blood loss, so-called “BLD” sensors, from the acronym “BLOOD LEAK DETECTOR”, which are placed at the outlet of the hemodialysis or haemofiltration filter.
One of the main drawbacks that afflict BLD sensors of the known type consists in the fact they have a sensitivity that, although compatible with the minimum requirements imposed by the aforesaid international standards, is relatively low. In fact, these sensors are activated only when significant amounts of blood pass, as there must be a perceptible change in the absorption value of the effluent under examination at the characteristic wavelengths of the haemoglobin.
Another drawback consists in the fact that BLD sensors of the known type often generate false positive results, as the presence of traces of blood in the filtrate fraction and/or in the dialysis fluid is mistaken for the presence of other substances that cause colour changes in the effluent, such as bilirubin. Similarly, BLD sensors of the known type do not allow to distinguish whether the presence of traces of blood leaving the filter is due to a phenomenon of haemolysis, which is often independent of the integrity of the filter.
Furthermore, BLD sensors often malfunction when exposed to inadequate ambient lighting conditions.
As mentioned above, another drawback of the hemodialysis and haemofiltration apparatuses commonly used in the clinical setting consists in the fact that a real-time analysis of parameters indicative of the effectiveness of the therapeutic treatment to which the patient is subjected is not generally provided.
For example, in the case of hemodialysis, the effectiveness of dialysis treatment depends on the extent to which toxins are removed from blood. In this respect, blood urea concentration is commonly used as a measure of blood toxicity.
Currently, the standard clinical procedure for monitoring hemodialysis therapy is to collect pre- and post-dialysis blood samples from time to time, e.g. once a month during a dialysis session, which are analysed to quantify the urea content in a clinical laboratory. It is therefore clear that such an occasional analysis, in addition to being particularly laborious and costly, is not indicative of the effectiveness of the single hemodialysis treatment.
Similar considerations can also be made with regard to haemofiltration therapies.
The main task of the present disclosure is in realising a blood filtering device, particularly for hemodialysis and/or haemofiltration apparatuses, which obviates the drawbacks and overcomes the limits of the prior art by allowing the integrity and performance of the filter itself to be monitored in real time with high sensitivity and accuracy.
Within the scope of this task, the present disclosure provides a blood filtering device which detects the presence, even minimal, of traces of blood leaving the filter, minimising false positive results.
The disclosure further provides a blood filtering device which is capable of giving the greatest assurances of reliability and safety in use.
The disclosure also provides a blood filtering device that is easy to manufacture and economically competitive if compared to the prior art, as well as easy to integrate into commonly used hemodialysis and/or haemofiltration apparatuses.
The aforementioned task, as well as the aforementioned advantages and others which will become clearer hereinafter, are achieved by providing a blood filtering device, particularly for hemodialysis and/or hemofiltration apparatuses as claimed in claim 1.
Other features are provided in the dependent claims.
Further characteristics and advantages will become clearer from the description of two preferred, but not exclusive, embodiments of a blood filtering device, illustrated by way of non-limiting example with the aid of the attached drawings in which:
With reference to the above-mentioned figures, the blood filtering device, particularly for hemodialysis and/or haemofiltration apparatuses, globally indicated with reference number 1, comprises a filter 2 comprising a first compartment 30, adapted to allow the passage of blood 3, 3′, and a second compartment 40, separated from the first compartment 30 by means of a membrane 20 adapted to allow the passage of a filtered fraction F from the first compartment 30 to the second compartment 40. The filtering device 1 further comprises an outlet conduit 5′ where, at least, the filtered fraction F leaving the filter 2 is collected. The filtered fraction F flows along said outlet conduit 5′ along a flow direction S.
As shown in the accompanying figures, the filtering device 1 receives at its inlet a flow of blood to be filtered, indicated with 3. Inside the filter 2, a filtered fraction F passes from the first compartment 30, through the membrane 20, to the second compartment 40, being separated from the incoming blood 3 and being collected in the outlet conduit 5′ of the filtering device 1. Reference 3′ indicates the filtered blood leaving the filtering device 1.
According to the disclosure, the filtering device 1 comprises a first sensor 6 comprising:
at least one semiconductor laser source 60, 61 comprising a laser cavity 62 and adapted to generate a laser light beam 64 which strikes the outlet conduit 5′ along an irradiation direction R incident to the flow direction S;
at least one front photodiode 66, 68 placed along the irradiation direction R on a side opposite to the semiconductor laser source 60, 61 with respect to the outlet conduit 5′,
at least one lateral photodiode 67, 69 placed along a diffusion direction D substantially orthogonal to the irradiation direction R.
At least in correspondence of the semiconductor laser source 60, 61, of the at least one front photodiode 66, 68 and of the at least one lateral photodiode 67, 69, the outlet conduit 5′ is transparent to the laser light beam 64.
The at least one front photodiode 66, 68 generates a first electrical signal dependent on the modulation of the power of the laser light beam 64 operated, according to a retro-injection interferometry effect (so-called “self-mixing interferometry” effect), by suspended particles present within the filtered fraction F and moving along the outlet conduit 5′.
The at least one lateral photodiode 67, 69 instead generates a second electrical signal depending on the part 65 of the laser light beam 64 which is diffused by the filtered fraction F along substantially the diffusion direction D.
Finally, the filtering device 1 comprises a processing and control unit 8 programmed to process the first electrical signal, generated by the at least one front photodiode 66, 68, and the second electrical signal, generated by the at least one lateral photodiode 67, 69, and to generate, on the basis of said two electrical signals, a signal indicative at least of the quantity of the suspended particles moving along the outlet conduit 5′.
Advantageously, from the combination of the electrical signals obtainable due to the presence of the two front photodiodes 66 and lateral photodiodes 67 it is possible to generate a signal representative of the presence of blood in the filtered fraction F in a very wide measuring range, since the front photodiode 66, operating in self-mix, is also sensitive to the passage of the single red blood cell inside the filtrate fraction F, while the lateral photodiode 67 is able to operate correctly when the quantity of blood, i.e. red blood cells, inside the filtrate fraction F becomes preponderant. In this situation, the front photodiode 66 goes into saturation while the lateral photodiode 67 continues to detect the radiation diffused by the red blood cells.
Advantageously, if the membrane 20 of the filter 2 is adapted to prevent the passage of particles such as red blood cells, the possibility of detecting their presence and quantity in the filtrate fraction F provides direct indications about the integrity of the filter 2 itself.
Advantageously, the processing and control unit 8 is programmed to generate an alarm signal based on the signal indicative at least of the quantity of suspended particles moving along the outlet conduit 5′, when this signal exceeds a predefined threshold value.
In this respect,
The experiment was carried out by adding known gradually increasing quantities of blood to a base liquid consisting of plasma.
The graph shown in
The “AVR” signal shown on the ordinate in the graphs of
Advantageously, starting from the combination of the electrical signals obtainable thanks to the presence of the aforesaid two front 66 and lateral 67 photodiodes, it is possible to implement different types of processing and calculations to obtain a signal indicative of the presence of suspended particles moving along the outlet conduit 5′ of the filtering device 1.
The calculation proposed above for obtaining the “AVR” signal shown in the graphs in
The graph illustrated in
As is evident from the comparison of the two trends shown in the graph in
As also explained below, other types of processing and calculations can be implemented, starting from the electrical signals generated by the front 66 and lateral 67 photodiodes, on the basis of different applications and on the basis of different substances and/or components whose presence in the liquid leaving the outlet conduit 5′ is to be verified.
Advantageously, through the analysis of the so-called “self-mix” signal generated by the front photodiode 66 it is possible to have a very sensitive measurement of the quantity of particles present in the filtrate fraction F, as it is also possible to detect the presence of even very small single particles, such as red blood cells, which otherwise could not be detected by a normal photodiode in any other way. Therefore, as mentioned above, the filtering device 1 is extremely sensitive in detecting minimal traces of blood due to the presence of red blood cells in the filtrate fraction F.
The high sensitivity is also combined with a high accuracy in detecting the presence of red blood cells. In fact, thanks to the fact that the filtering device 1 takes into account both the self-mix signal detected by the front photodiode 66 and the signal related to the diffusion of the laser radiation detected by the lateral photodiode 67, it is possible to reduce the undesired effects due to the responses of the photodiodes in the presence of ambient light.
In a practical example, the sensor 6 can easily distinguish the presence of red blood cells in the filtrate fraction F, e.g. due to a rupture in the filter 2, from the presence of haemoglobin dissolved in the blood as a result of a haemolysis phenomenon independent of the integrity of the filter 2.
In the first case, the front photodiode 66, operating in self-mix, will detect the passage of red blood cells, while the lateral photodiode 67 will detect the part of laser radiation diffused by them. In the second case, the front photodiode 66 will not detect any passage of red blood cells, while the lateral photodiode 67 will still detect a laser radiation diffused by the presence of hemoglobin dissolved in the filtrate fraction F.
The expression “flow direction S” means the direction along which the filtrate fraction F flows within the outlet conduit 5′, with particular reference to the portion of said conduit 5′ which is struck by the laser light beam 64. In the case in which the conduit 5, in the portion thereof which is struck by the laser light beam 64, has a rectilinear course, said flow direction S coincides, or is parallel, with the central axis of the conduit 5. In the case in which the conduit 5′ has, precisely in the portion thereof which is struck by the laser light beam 64, a curvilinear course according to a curved line, the expression “flow direction S” means the direction tangent to the curved line near the area of the conduit 5 struck by the laser light beam 64.
The term “incident” means that the flow direction S and the irradiation direction R have a common point, that is, they intersect defining an angle greater than 0°.
Preferably the angle of incidence between the flow direction S and the irradiation direction R is substantially equal to 90°.
Advantageously, the sensor 6 comprises a monitor photodiode 13 arranged upstream of the laser cavity 62, adapted to generate an electrical signal also dependent on the modulation of the power of the laser light beam 64 (so-called self-mix signal). In this case, the processing and control unit 8 is programmed to also process the electrical signal generated by the monitor photodiode 13 to improve the signal-to-noise ratio of the signal indicative of the quantity of suspended particles moving along the outlet conduit 5′.
In fact, the front photodiode 66, 68 and the monitor photodiode 13 both measure the amplitude modulations of the laser light beam 64 induced by the self-mix effect. However, these modulations have opposite signs between them. Therefore, by calculating the difference between the two self-mix signals detected by the front photodiode 66, 68 and by the monitor photodiode 13, a gain of a factor of two is obtained on the amplitude of the self-mix signal, and also a subtraction of all the common disturbances is obtained, such as the noise and the disturbances of the supply of the laser source, as well as the “shot-noises” and the “1/f” noise of the laser itself.
Advantageously, the processing and control unit 8 is programmed to process the first and the second electrical signals to also generate a signal indicative of the type and/or quantity of solutes present in the filtrate fraction F, such as for example urea, hemoglobin, bilirubin.
Advantageously, the processing and control unit 8 comprises a programmable memory 80 configured to receive and store at least one reference signal associated with at least one specific type of solute present in a reference liquid. The processing and control unit 8 is then programmed to generate a signal indicative of the quantity of said type of solute present in the filtrate fraction F on the basis of a comparison with the signal reference associated with said type of solute stored in the programmable memory 80. This comparison can advantageously be made in real time in the processing and control unit 8, so that the signal indicative of the quantity of this type of solute present in the filtrate fraction F can be obtained in real time.
Advantageously, it is also possible to store a plurality of reference signals associated with a plurality of different types of solutes present in a reference liquid, so as to generate a plurality of signals indicative of the quantity of each of these solutes present in the filtrate fraction F.
In essence, it is possible to store, in the programmable memory 80 of the processing and control unit 8, a “fingerprint” of a reference liquid in relation to a plurality of different solutes whose presence in the filtrate fraction F is to be ascertained. In this way, the comparison of the signals generated by the processing and control unit 8 starting from the measurements of the photodiodes 66 and 67 with the aforesaid “fingerprint” of a reference liquid allows to estimate the amount of the solutes of interest in the filtrate fraction F.
Advantageously, the programmable memory 80 is configured to receive and store at least one reference signal associated with at least one specific type of solute present in a reference liquid consisting of the filtrate fraction F obtained by filtering the blood of a patient in a given therapeutic session. In this case, the “fingerprint” is taken on the patient's own filtrate fraction F, to be used in the analysis of subsequent therapeutic sessions in order to evaluate their effectiveness over time.
Advantageously, as illustrated in particular in
In this case, the filter 2 is a dialyzer filter comprising a first compartment 30, adapted to allow the passage of blood 3, 3′, and a second compartment 40, adapted to allow the passage of a dialysis fluid 4, 4′. The first compartment 30 and the second compartment 40 are separated by a semi-permeable membrane 20 which is selective to the crossing of the filtrate fraction F from the blood 3 to the dialysis fluid 4, according to an osmotic phenomenon. In this case the outlet conduit 5′ is adapted to collect a mixture 4′ of the filtrate fraction F and of the dialysis liquid.
Therefore, the first electrical signal generated by the front photodiode 66, 68 depends on the modulation of the power of the laser light beam 64 operated, according to a retro-injection interferometry effect, by suspended particles present within the mixture 4′ of the filtrate fraction F with the dialysis fluid moving along the outlet conduit 5′.
Correspondingly, the second electrical signal generated by the lateral photodiode 67, 69 depends on the part 65 of the laser light beam 64 which is diffused by the mixture 4′ of the filtrate fraction F with the dialysis fluid along substantially the diffusion direction D. For convenience, the mixture 4′ of the filtrate fraction F with the dialysis fluid will also be referred to simply as “dialysis fluid 4′ at the output”.
Advantageously, from the combination of the electrical signals obtainable thanks to the presence of the two front 66 and lateral 67 photodiodes, it is possible to generate not only a signal representative of the presence of blood in the dialysis fluid 4′ at the output of the filtering device 1, as described above, but also to generate a signal representative of the type and quantity of solutes present in said dialysis fluid 4′.
Advantageously, the processing and control unit 8 is programmed to process the first electrical signal, generated by the front photodiode 66, and the second electrical signal, generated by the lateral photodiode 67, to generate a signal indicative of the quantity of urea present in the mixture 4′ of the filtrate fraction F and of the dialysis fluid.
Advantageously, the processing and control unit 8 is programmed to perform an algorithm classifying one or more features of the first electrical signal and one or more features of the second electrical signal and to generate said signal indicative of the quantity of urea present in the mixture 4′ of the filtrate fraction F and of the dialysis fluid.
Advantageously, the aforesaid classifying algorithm may be defined starting from automated machine learning techniques, preferably starting from automated learning techniques based on the so-called “random decision forest” classification methods capable of identifying, among the many statistical features of two or more input signals, the main features that allow to robustly estimate a desired output signal, such as precisely a signal indicative of the amount of urea present in the dialysis fluid 4′ leaving the filtering device 1.
In particular, it is possible to store in the programmable memory 80 of the processing and control unit 8 a “fingerprint” of the urea consisting of the main statistical features that best describe the presence of urea in a reference fluid, so that the processing and control unit 8 can generate in real time the signal indicative of the quantity of urea present in the dialysis fluid 4′ leaving the filtering device 1.
In other words, once the machine learning algorithm has been trained to recognise the presence of urea in a reference fluid, this algorithm can be stored in the programmable memory 80 and be executed by the processing and control unit 8 which, by correlating the two signals generated by the two photodiodes 66 and 67, is able to estimate in real time the amount of urea present in the dialysis fluid 4′ and thus to allow the effectiveness, for the patient, of the current dialysis session to be known in real time.
Therefore, starting from the same signals generated by the front photodiode 66 and by the lateral photodiode 67, it is possible to derive different types of information on different properties of the filtrate fraction F and/or of the dialysis liquid 4′ leaving the filter 2 through different processing of the aforesaid two signals and performing different statistical calculations.
As indicated, it is possible, for example, to detect the presence and the quantity of blood in the dialysis fluid 4′ leaving a hemodialysis filter 2, distinguishing among other things the presence of red blood cells, which may indicate a rupture in the filter 2 itself, from the presence of haemoglobin dissolved in the dialysis fluid 4′, which may indicate a haemolysis phenomenon, or even the amount of urea present in the dialysis fluid 4′ leaving the filter 2, which provides a real-time indication of the progress of the hemodialysis therapy.
Advantageously, the at least one semiconductor laser source 60, 61 is adapted to generate a laser light beam 64 having a wavelength comprised between 600 and 850 nm, preferably comprised between 750 and 800 nm, and even more preferably equal to about 780 nm. Advantageously, the front 66 and lateral 67 photodiodes are operational at least in a working range compatible with the wavelength of the laser light beam 64.
With a semiconductor laser source 60 that is adapted to generate a laser light beam 64 having a wavelength of about 780 nm it has been possible to obtain information on the presence of blood in the plasma, as illustrated in the graphs shown in
Moreover, with the same experimental set-up, and in particular with a semiconductor laser source 60 generating a laser light beam 64 having a wavelength of about 780 nm, it has been possible to develop a classification algorithm capable of estimating, starting from the combination of the signals generated by the two photodiodes 66 and 67, the amount of urea present in a test liquid.
Advantageously, since the urea absorption spectral range is centred on 280 nm, the selection of a semiconductor laser source 60 generating a laser light beam 64 having a wavelength of about 280 nm is preferable for the purpose of identifying urea in the dialysis fluid 4′ leaving the filter 2.
Advantageously, as illustrated in
The filtering device 1 advantageously comprises a second sensor 9 comprising:
at least one semiconductor laser source 90 comprising a laser cavity 92 and adapted to generate a laser light beam 94 which strikes the inlet conduit 5 along an irradiation direction incident to the flow direction S;
at least one front photodiode 96 placed along the irradiation direction on a side opposite to said semiconductor laser source 90 with respect to the inlet conduit 5,
at least one lateral photodiode 97 placed along a diffusion direction D substantially orthogonal to the irradiation direction R.
At least in correspondence of the semiconductor laser source 90, of the at least one front photodiode 96 and of the at least one lateral photodiode 97, the inlet conduit 5 is transparent to the laser light beam 94.
The at least one front photodiode 96 generates an electrical signal dependent on the modulation of the power of the laser light beam 94 operated, according to a retro-injection interferometry effect (so-called “self-mixing interferometry” effect), by suspended particles possibly present within the dialysis fluid 4 and moving along the inlet conduit 5.
On the other hand, the at least one lateral photodiode 97 generates an electrical signal depending on the part of the laser light beam 94 which is diffused by the dialysis liquid 4 along substantially the diffusion direction D.
The processing and control unit 8 is in this case programmed to use the two electrical signals generated by the front photodiode 96 and by the lateral photodiode 97 in subtraction respectively of the two electrical signals detected by the front photodiode 66 and by the lateral photodiode 67 of the first sensor 6 to generate said signal indicative of at least the quantity of said suspended particles moving along the outlet conduit 5′ deprived of the disturbances common to the electrical signals of the first sensor 6 and of the second sensor 9.
Advantageously, the second sensor 9 is substantially a replica of the first sensor 6.
In particular, the second sensor 9 may have exactly the same components as the first sensor 6.
In this way, it is further ensured that the signals generated by the first sensor 6 can be used differentially from the signals generated by the second sensor 9 to eliminate all common mode disturbances, such as those of an electrical nature, those due to external ambient lighting conditions, and those due to particular physical/chemical features of the dialysis fluid 4.
Advantageously, the first sensor 6 comprises a first semiconductor laser source 60 and at least a further source 61, 61′ that is selectable between:
(i) at least a second semiconductor laser source 61 adapted to generate a laser light beam having a different wavelength with respect to the laser light beam 64 generated by the first semiconductor laser source 60, and
(ii) at least a radiation source 61′ adapted to generate a radiation that strikes the outlet conduit 5′ along an irradiation direction incident to the flow direction S.
Advantageously, the radiation source 61′, such as an LED, emits a radiation having a much wider emission spectrum than the emission spectrum of the laser source 60, which is adapted to emit a coherent radiation beam.
Thus, in a first example, the first sensor 6 comprises a first semiconductor laser source 60 and at least a further radiation source 61′, such as an LED.
In this case, the at least one front photodiode 66, 68 generates an electrical signal indicative of the transmittance of the radiation emitted by the radiation source 61′ through the filtrate fraction F (or through the dialysis fluid 4′ at the output), wherein the transmittance of said radiation depends on the quantity and/or type of solutes present in the filtrate fraction F, while the at least one lateral photodiode 67, 69 generates an electrical signal depending on the part of said radiation which is diffused by the filtrate fraction F (or by the dialysis liquid 4′ at the output) along substantially the diffusion direction D.
Advantageously, therefore, with the same set of photodiodes, i.e., with the same front photodiode 66 and with the same lateral photodiode 67, it is possible to operate the first sensor 6 as described above, i.e., also using the self-mix operation, and also as a spectrophotometric sensor.
It is in fact possible, for example, to synchronise the activation of the first laser source 60 in an alternating manner with respect to the activation of the further radiation source 61′, so that the same photodiodes 66, 67 are sensitive in an alternating manner in time to the two different types of radiation.
Advantageously, the radiation emitted by the further radiation source 61′ may present a spectrum of wavelengths which also include the wavelength of the radiation constituting the laser light beam 64 emitted by the first semiconductor laser source 60. Basically, the radiation emitted by the source 61′ and the laser emitted by the laser source 60 can overlap in terms of wavelength values.
Advantageously, it is also possible to provide for a plurality of different radiation sources 61′, such as different LEDs, centred on different wavelengths.
Advantageously, therefore, the first sensor 6 integrates, in a compact manner and with a limited number of components, both the possibility of carrying out an interferometric analysis and the possibility of carrying out spectrophotometry.
In a second example, the first sensor 6 comprises at least two semiconductor laser sources 60, 61, wherein a first semiconductor laser source 60 is adapted to generate a laser light beam 64 having a different wavelength with respect to the laser light beam generated by a second semiconductor laser source 61.
Advantageously, the sensor 6 comprises at least two front photodiodes 66, 68 placed along the irradiation direction R on a side respectively opposite to the first semiconductor laser source 60 and to the second semiconductor laser source 61 with respect to the outlet conduit 5′, and at least two lateral photodiodes 67, 69 placed along a diffusion direction D substantially orthogonal to the irradiation direction R.
The possibility of providing a plurality of laser sources 60, 61 operating at different wavelengths, and possibly a corresponding plurality of front 66, 68 and lateral 67, 69 photodiodes, makes it possible to improve the sensitivity of the sensor 6 to the detection of specific particles and/or solutes as a function of the relative absorption spectrum.
As mentioned above, a laser source at about 780 nm is preferably used to identify and quantify blood in the filtrate fraction F and can also be used to estimate the amount of urea in the filtrate fraction F or in the dialysis fluid 4′ at the output of the filter 2, by means of a classifying algorithm.
Preferably, the possibility of using laser sources 60 and 61 operating at different wavelengths allows to improve the selectivity of the sensor 6 in detecting different particles and/or solutes.
For example, a semiconductor laser source adapted to emit a laser light beam 64 at a wavelength comprised between 200 and 400 nm, preferably comprised between 200 and 300 nm, e.g. equal to about 280 nm, can be used for improving the selectivity of measurement of substances such as the urea present in the dialysis liquid 4′ at the output of a dialyzer filter 2.
Basically, the fact of providing a plurality of laser sources 60, 61 operating at different wavelengths, and possibly a plurality of front 66, 68 and lateral 67, 69 photodiodes if a single photodiode does not have an operating range sufficient to cover the overall range of radiations emitted by the different laser sources 60, 61, allows to increase the performance of the sensor 6 making it usable to detect, in a very sensitive manner, the presence and quantity of different types of particles and/or different types of solutes.
Advantageously, the first sensor 6, as well as, similarly, the second sensor 9, may comprise at least one control photodiode 55 adapted to intercept the laser light beam 64 directly emitted by the at least one semiconductor laser source 60 (i.e., adapted to intercept the laser light beam 64 in an area where it does not pass through, or has not yet passed through, the outlet conduit 5′). This control photodiode 55 generates an electrical control signal directly dependent on the laser light beam 64. Advantageously, the processing and control unit 8 is programmed to also process said electrical control signal in order to generate a signal indicative at least of the quantity of suspended particles along the outlet conduit 5′, or also a signal indicative of the type and/or the quantity of solutes present in the filtrate fraction F.
Advantageously, the filtering device 1 comprises, in addition to sensors 6 and 9, at least one spectrophotometric sensor 7, which comprises:
a radiation source 70 adapted to generate a radiation 72 which strikes the outlet conduit 5′ along a direction of radiation incident to the flow direction S,
a photodiode 74 placed along the irradiation direction on a side opposite to the radiation source 70 with respect to the outlet conduit 5′.
This photodiode 74 generates an electrical signal indicative of the transmittance of the radiation through the filtrate fraction F, which transmittance depends on the quantity and/or type of solutes present in the filtrate fraction F.
The processing and control unit 8 is programmed to process the aforesaid electrical signal to generate a signal indicative of the quantity and/or type of solutes present in the filtrate fraction (F).
On the basis of the type of spectrometric sensor 7 used, the filtering device 1 is capable of providing information about solutes of a different type from the types that the sensor 6 is instead capable of detecting, or even redundant information about the same solutes that the sensor 6 is capable of detecting, thus making the analysis of the performance of the filtering device 1 even more robust.
Advantageously, the radiation source 70 of the spectrophotometric sensor 7 is adapted to generate a radiation 72 having a wavelength comprised between 500 nm and 850 nm, wherein the transmittance of the radiation 72 depends on the amount of hemoglobin (and/or bilirubin) present in the filtrate fraction F. The processing and control unit 8 is in this case programmed to process the electrical signal generated by the photodiode 74 so as to generate a signal indicative of the quantity of hemoglobin and/or bilirubin present in the filtrate fraction F.
Advantageously, the radiation source 70 of the spectrophotometric sensor 7 is adapted to generate an ultraviolet radiation 72 or in the Near Infrared (NIR) range, wherein the transmittance of the radiation 72 depends on the amount of urea present in the filtrate fraction F, and in particular in the mixture 4′ of the filtrate fraction 4 with the dialysis fluid. The processing and control unit 8 is in this case programmed to process the electrical signal generated by the photodiode 74 so as to generate a signal indicative of the quantity of urea present in the filtrate fraction F.
In addition, a plurality of spectrophotometric sensors 7 placed in correspondence of the outlet conduit 5′ can be provided, each configured to detect the presence and the amount of a different type of solute within the same filtrate fraction F.
Advantageously, in the case of a filtering device 1 for hemodialysis apparatuses 100, an inlet conduit 5 is adapted to convey the dialysis fluid 4 inlet to the filter 2 dialyzer, wherein said dialysis fluid 4 flows along the inlet conduit 5 according to a flow direction S. A second spectrophotometric sensor 11 is advantageously present in correspondence of the inlet conduit 5, having technical characteristics corresponding to those of the spectrophotometric sensor 7 described above.
In this case, the second spectrophotometric sensor 11 placed at the inlet of the dialyzer filter 2 allows to generate information useful for a better processing of the information derivable from the spectrophotometric sensor 7 placed at the outlet of the dialyzer filter 2, for example to eliminate common mode disturbances, as well as to generate information related to the characteristics of the dialysis liquid 4 at the inlet of the same filter 2.
Advantageously, the filtering device 1 also comprises one or more of the following sensors:
The processing and control unit 8 is programmed to also process the signals generated by such sensors to generate a signal indicative of the quantity of suspended particles moving along the outlet conduit 5′, as well as the type and/or the amount of solutes present in the filtrate fraction F.
The present disclosure further relates to a hemodialysis and/or haemofiltration apparatus 100 comprising a blood filtering device 1 as described above.
For example,
As illustrated schematically in
The hemodialysis apparatus 100 then comprises a pump 111, of the peristaltic type, adapted to put part of the patient's blood into extracorporeal circulation, a system for introducing heparin 112 into the blood taken from the patient, a system for removing air 113 possibly present in the blood, before the re-introduction thereof into the patient. There are also provided some devices for detecting and displaying arterial pressure 114 and venous pressure 115, as well as devices 116 for detecting and displaying the flow of blood entering the dialyzer filter 2.
The present disclosure further relates to a process for detecting suspended particles and/or solutes present in a filtered fraction F coming out of a blood filtering device 1.
According to the disclosure, the process includes the steps of:
having a blood filtering device 1 as described above;
detecting a first electrical signal generated by the at least one front photodiode 66, 68 of the first sensor 6;
detecting a second electrical signal generated by the at least one lateral photodiode 67, 69 of the first sensor 6;
processing said first electrical signal and said second electrical signal to generate a signal indicative at least of the amount of suspended particles moving along the outlet conduit 5′ of the filtering device 1.
As described above, processing the electrical signals generated by the front photodiode 66 and the lateral photodiode 67 makes it possible to detect the presence of red blood cells and thus to know the amount of blood in the filtrate fraction F.
Advantageously, the process for the detection of suspended particles and/or solutes comprises the following steps:
processing said first electrical signal and said second electrical signal to generate a signal indicative of the quantity and/or type of solutes present in the filtrate fraction F at the output of the filtering device 1.
As described above, the processing of the electrical signals generated by the front photodiode 66 and by the lateral photodiode 67 also makes it possible to estimate the amount of particular solutes in the filtrate fraction F or in the dialysis fluid 4′ leaving the filtering device 1, such as for example urea, or haemoglobin, or bilirubin.
Advantageously, the process for the detection of suspended particles and/or solutes comprises the following steps:
comparing the signal indicative of the presence and/or quantity of the suspended particles with the signal indicative of the quantity and/or type of solutes present in the filtrate fraction F at the outlet of the filtering device 1;
processing a signal indicative of the composition of the filtrate fraction F.
As described above, the comparison between the electrical signals generated by the two different photodiodes 66 and 67 allows to estimate the amount of various solutes present in the filtrate fraction F or in the dialysis fluid 4′ leaving filter 2, such as urea, bilirubin, or haemoglobin dissolved in the filtrate fraction F for a hemodialysis effect.
Advantageously, the process for the detection of suspended particles and/or solutes comprises the following step:
generating an alarm signal when the signal indicating at least the quantity of suspended particles in movement along the outlet conduit 5′ exceeds a threshold value.
In practice it has been found that the blood filtering device, particularly for hemodialysis and/or haemofiltration apparatuses, according to the present disclosure, achieves the intended aim and objects as it is possible to monitor its integrity and performance in a highly sensitive and accurate manner.
Another advantage of the blood filtering device, according to the disclosure, relates to the fact of incorporating a “BLD” sensor—Blood Leak Detector—capable of generating an alarm signal in the presence of traces of blood in the filtrate fraction.
A further advantage relates to the fact that it is possible to distinguish, in the filtrate fraction, the presence of red blood cells from the presence of bilirubin and/or haemoglobin. In fact, it is generally the presence of red blood cells in the filtrate fraction that indicates that the filter has been damaged. Conversely, the detection of haemoglobin in the absence of red blood cells indicates the occurrence of a phenomenon of hemolysis which usually does not depend on the integrity of the filter.
Yet another advantage of the disclosure relates to the fact that the detection of traces of blood in the filtrate fraction is not affected by the surrounding ambient light conditions.
A further advantage of the filtering device, according to the disclosure, relates to the fact that the combination of the signals obtained from photodiodes working both in self-mix and in radiation absorption allows to verify the presence and estimate the quantity of different types of solutes present in the filtrate fraction at the output of the filtering device.
In particular, the possibility of estimating the presence of urea, moreover with the same sensor adapted to work as a BLD, makes it possible to know in real time the effectiveness of the therapeutic treatment, for example of hemodialysis, which is being carried out, being able to intervene accordingly, for example by interrupting or prolonging the therapeutic session when a desired purification of the blood is found, or by modifying the dialysis parameters during the session itself.
Yet another advantage of the disclosure relates to the fact that the combination of the signals obtained from photodiodes working both in self-mix and in radiation absorption allows obtaining very robust and accurate information about the presence of blood and/or solutes in the filtrate fraction.
A further advantage of the disclosure relates to the fact that it is inexpensive to manufacture and to fit into hemodialysis and/or haemofiltration apparatuses of known type.
The blood filtering device, particularly for hemodialysis and/or haemofiltration apparatuses thus conceived, is susceptible of numerous modifications and variations, all of which are within the scope of the inventive concept.
Furthermore, all the details can be replaced by other technically equivalent elements.
In practice, any materials can be used according to requirements, as long as they are compatible with the specific use, the dimensions and the contingent shapes.
Number | Date | Country | Kind |
---|---|---|---|
102020000031391 | Dec 2020 | IT | national |