TWM646121U - Auxiliary determination system for kidney dialysis vascular obstruction - Google Patents

Abstract

An auxiliary determination system for renal dialysis vascular access obstruction, suitable for signal connection to a hemodialysis machine. The auxiliary determination system for renal dialysis vascular access obstruction includes a connected host and an artificial intelligence model. The connection host retrieves a plurality of hemodialysis data for multiple hemodialysis treatments received by the hemodialysis machine through the renal dialysis vascular access. The artificial intelligence model is connected to the host computer. Based on the artificial intelligence model trained on hemodialysis data, when a kidney dialysis patient undergoes hemodialysis treatment, the artificial intelligence model is used to generate auxiliary judgments, and the display device can display whether the kidney dialysis vascular access is blocked to the point that intervention is required.

Description

Auxiliary determination system for renal dialysis vascular access obstruction

The present disclosure relates to an auxiliary determination system, particularly an auxiliary determination system for renal dialysis vascular access obstruction.

With the advancement of medical technology and improvement in the quality of care, patients with end-stage renal disease are extending their lives by receiving hemodialysis treatment. Hemodialysis requires the establishment of vascular access, and a common complication of vascular access is fistula obstruction. It can be seen that the function of the vascular access for renal dialysis often determines the effect and quality of hemodialysis. When the patient is undergoing renal dialysis, whether the vascular access needs to be intervened, especially the golden treatment time for vascular access obstruction is 48 hours. Toxin accumulation, electrolyte imbalance. When the vascular access is severely blocked and cannot be used, life will be in danger.

The existing technology uses methods to detect whether the vascular access is blocked. One is for the patient to perform routine self-detection, and the other is for the nurse to assess the condition of the vascular access through palpation during hemodialysis. However, these detection actions often rely on human experience, and It is more subjective, and there may be doubts about inaccurate judgment, which will cause doubtful patients to lose the golden period of rescue.

In addition, the existing technology requires additional equipment (such as an ultrasound device to listen to the blood flow sound at the vascular access site and convert it into an image for judgment, or a blood flow sound acquisition device to capture the blood flow sound and convert it into a frequency spectrum for judgment). This not only increases equipment costs, but also increases operating manpower.

The present disclosure provides an auxiliary determination system for renal dialysis vascular access obstruction, which can assist medical staff in determining whether the renal dialysis vascular access is blocked enough to require intervention.

One embodiment of the present disclosure provides an auxiliary determination system for renal dialysis vascular access obstruction, which is suitable for signal connection to a hemodialysis machine. The auxiliary determination system for renal dialysis vascular access obstruction includes a connected host and an artificial intelligence model. The connection host retrieves a plurality of hemodialysis data for multiple hemodialysis treatments received by the hemodialysis machine through the renal dialysis vascular access. The artificial intelligence model is connected to the host computer. An artificial intelligence model trained based on hemodialysis data is used to generate an auxiliary judgment reference when a kidney dialysis patient undergoes hemodialysis treatment.

In one embodiment, the above-mentioned auxiliary determination system for renal dialysis vascular access obstruction further includes a display device, and the display device is connected to the connection host. The display device displays the auxiliary judgment reference.

In one embodiment, the diagnostic reference result is that the obstruction requires interventional treatment or does not require interventional treatment.

In one embodiment, the above-mentioned connected host receives the hemodialysis data before an interventional processing action means is executed as the hemodialysis data during obstruction.

In one embodiment, the above-mentioned connected host receives the hemodialysis data executed by an interventional processing action means as unblocked hemodialysis data.

In one embodiment, the hemodialysis data includes a physiological detection value.

In one embodiment, the physiological detection values include a diastolic blood pressure value, a systolic blood pressure value, a pulse value, a respiration value, a blood flow value and a venous pressure value.

In one embodiment, the above-mentioned artificial intelligence model includes a model training data set. The model training data set includes a plurality of hemodialysis data during obstruction and a plurality of non-blockage hemodialysis data. The hemodialysis data during obstruction and non-blockage Hemodialysis data includes corresponding A diastolic blood pressure value, a systolic blood pressure value, a pulse value, a respiration value, a blood flow value and a venous pressure value.

In one embodiment, the number of hemodialysis data during obstruction is three.

In one embodiment, the number of unblocked hemodialysis data is five.

In one embodiment, the number of the hemodialysis data is eight.

In one embodiment, the amount of unblocked hemodialysis data is greater than the amount of blocked hemodialysis data.

In one embodiment, the connected host uses a type of neural network algorithm to modify the artificial intelligence model.

In one embodiment, the neural network-like algorithm is a random forest algorithm.

Based on the above, in the auxiliary determination system for renal dialysis vascular access obstruction proposed in this disclosure, this disclosure only needs to use the hemodialysis data collected by the hemodialysis machine and the artificial intelligence model to provide medical staff with the assistance to determine whether the renal dialysis vascular access is blocked. The blockage requires intervention. There is no need to use additional external equipment (such as an ultrasound device to listen to the blood flow sound of the dialysis vascular access, or an acquisition device to capture the blood flow sound and convert it into a frequency spectrum) to make a judgment.

In order to make the present disclosure more obvious and understandable, embodiments are given below and described in detail with reference to the attached drawings.

50:Hemodialysis machine

60: Renal dialysis vascular access

100: Auxiliary determination system for renal dialysis vascular access obstruction

110:Connect to the host

120:Artificial intelligence model

122: Model training data set

130:Display device

A: The blockage requires intervention.

B: No need for intervention

BP1: blood from arteries

BP2: blood sent to veins

BPS: Hemodialysis information

G1: Hemodialysis data during obstruction

G2: Unobstructed hemodialysis data

TN: physiological test value

TN1: diastolic blood pressure value

TN2: systolic blood pressure value

TN3: pulse value

TN4: breathing value

TN5: blood flow value

TN6: venous pressure value

RS: auxiliary judgment reference

Figure 1 is a schematic diagram of one embodiment of the disclosed auxiliary determination system for renal dialysis vascular access obstruction. Figure.

Figure 2 is a schematic diagram of an embodiment of the hemodialysis data of the present disclosure.

Figure 3 is a schematic diagram of the model training data set being the hemodialysis data during obstruction and the hemodialysis data without obstruction.

Embodiments are enumerated below and described in detail with reference to the drawings. However, the embodiments provided are not intended to limit the scope of the present disclosure. In addition, the drawings are for illustrative purposes only and are not drawn to original size. To facilitate understanding, the same elements will be identified with the same symbols in the following description.

The terms "including", "including", "having", etc. mentioned in this disclosure are all open terms, which means "including but not limited to".

In the description of various embodiments, when terms such as “first” and “second” are used to describe elements, they are only used to distinguish these elements from each other and do not limit the order or importance of these elements.

In the description of various embodiments, the so-called "coupling" or "connection" may refer to two or more elements making direct physical or electrical contact with each other, or indirectly making physical or electrical contact with each other, and "coupling" "Connect" or "connect" can also refer to the mutual operation or action of two or more components.

Figure 1 is a schematic diagram of an auxiliary determination system for renal dialysis vascular access obstruction according to an embodiment of the present disclosure. Please refer to Figure 1 . The disclosed auxiliary determination system 100 for renal dialysis vascular access obstruction is suitable for signal connection to the hemodialysis machine 50 . The hemodialysis machine 50 is connected to the patient's dialysis vascular access 60 to perform hemodialysis treatment. There is an artificial kidney in the hemodialysis machine 50, and the dialysis blood vessels are open The blood BP1 from the artery in the path 60 can detoxify the toxins after flowing through the artificial kidney, and then the detoxified blood BP2 sent to the veins returns to the veins in the dialysis vascular passage 60 .

In this embodiment, the auxiliary determination system 100 for renal dialysis vascular access obstruction includes a connection host 110, an artificial intelligence model 120, and a display device 130. The connection host 110 is connected to the hemodialysis machine 50, for example, through the Internet of Things (IOT). The connection host 110 can receive a plurality of hemodialysis data BPS from the hemodialysis machine 50 that receives multiple hemodialysis treatments HS from the dialysis vascular access 60 . The hemodialysis data BPS includes physiological detection values TN. As shown in Figure 2, the physiological detection values TN include, for example, a diastolic blood pressure value TN1, a systolic blood pressure value TN2, a pulse value TN3, a respiration value TN4, a blood flow value TN5, and 1. Venous pressure value TN6 and other information.

The artificial intelligence model 120 is connected to the connected host 110, and the artificial intelligence model 120 is trained based on the connected host 110 acquiring a large number of people's hemodialysis data BPS over a long period of time. The connection host 110 uses a type of neural network algorithm to modify the artificial intelligence model. The neural network-like algorithm is, for example, a random forest algorithm.

Compared with human judgment, this disclosure uses machine learning to make the prediction ability of the artificial intelligence model 120 after training more objective, which can be used by medical staff to assist in judgment. Further, the disclosed connected host 110 will continuously record and collect hemodialysis data BPS, and input and update the artificial intelligence model 120, thereby continuously updating the artificial intelligence model 120.

In the above machine learning process, the number of hemodialysis data BPS is, for example, eight, that is, every eight hemodialysis data BPS are trained and updated through the random forest algorithm 120 of the artificial intelligence model. These hemodialysis data BPS are collected into the artificial intelligence model. The model training data set 122 (as shown in Figure 3) in the model 120 is used for training.

Please refer to Figure 3, the model training data set 122 is divided into hemodialysis during obstruction Data G1 and unobstructed hemodialysis data G2. The hemodialysis data BPS before the interventional treatment action method is executed is used as the hemodialysis data G1 during obstruction. On the other hand, the hemodialysis data BPS after the interventional processing action means is executed is regarded as the unblocked hemodialysis data G2. Each hemodialysis data G1 during obstruction and each non-obstruction hemodialysis data G2 respectively include the corresponding diastolic blood pressure value TN1, systolic blood pressure value TN2, pulse value TN3, respiratory value TN4, and blood flow as shown in Figure 2. Value TN5 and venous pressure value TN6 and other information.

For example, there were patients who had previously experienced obstruction of the renal dialysis vascular access 60 and underwent vascular access interventional surgery. For example, after performing the vascular access interventional surgery on May 5, 2023, five days after May 6, 2023 The renal dialysis vascular access 60 is in a patent state, as the unobstructed hemodialysis data G2, and the renal dialysis vascular access 60 in the first three days or the first four times on May 5, 2023 is in a blocked state, as the hemodialysis data when blocked Data G1.

In this disclosure, the three hemodialysis data BPS before the execution of the interventional processing action means are regarded as the hemodialysis data G1 during obstruction, and the five hemodialysis data BPS after the execution of the interventional processing action means are regarded as the unobstructed hemodialysis data G2. Through the first The artificial intelligence model 120 is shown to assist in determining whether the renal dialysis vascular access 60 is blocked enough to require interventional treatment.

The display device 130 is connected to the connected host 110 . In this way, when the patient undergoes hemodialysis treatment again, the connection host 110 captures the time when the hemodialysis machine 50 receives the hemodialysis vascular access 60 and obtains the hemodialysis data BPS. At the end of the kidney dialysis, 5 hemodialysis data BPS during the kidney dialysis process are retrieved and used through the artificial intelligence model 120 to generate the auxiliary judgment reference RS. The display device 130 displays the auxiliary judgment reference RS, such as blockage requiring interventional treatment A, or In the state of no need for interventional treatment B, this auxiliary determination reference RS can provide medical staff with the ability to determine whether the renal dialysis vascular access 60 is blocked to the extent that interventional treatment is required.

In summary, in the auxiliary determination system for renal dialysis vascular access obstruction proposed in this disclosure, this disclosure only needs to use the hemodialysis data collected by the hemodialysis machine and the artificial intelligence model to provide medical staff with the auxiliary determination of renal dialysis vascular access obstruction. Is the passage blocked enough to require intervention? There is no need to use additional external equipment (such as an ultrasound device to listen to the blood flow sound of the dialysis vascular access, or an acquisition device to capture the blood flow sound and convert it into a frequency spectrum) to make a judgment.

The auxiliary judgment reference generated by this disclosure can be used as a reference for medical staff to judge whether the renal dialysis vascular access is blocked to the extent that interventional treatment is required.

Compared with human judgment, this disclosure uses machine learning to make the prediction ability of the artificial intelligence model after training more objective, which can be used to assist medical staff in making judgments.

Although the disclosure has been disclosed above through embodiments, they are not intended to limit the disclosure. Anyone with ordinary knowledge in the technical field may make slight changes and modifications without departing from the spirit and scope of the disclosure. Therefore, The scope of protection of this disclosure shall be determined by the scope of the patent application attached.

50:Hemodialysis machine

60: Renal dialysis vascular access

100: Auxiliary determination system for renal dialysis vascular access obstruction

110:Connect to the host

120:Artificial intelligence model

130:Display device

A: The blockage requires intervention.

B: No need for intervention

BP1: blood from arteries

BP2: blood sent to veins

BPS: Hemodialysis information

RS: auxiliary judgment reference

Claims (14)

An auxiliary determination system for renal dialysis vascular access obstruction, suitable for signal connection to a hemodialysis machine. The auxiliary determination system for renal dialysis vascular access obstruction includes: a connection host, which retrieves a renal dialysis blood vessel received by the hemodialysis machine A plurality of hemodialysis data for multiple hemodialysis treatments; and an artificial intelligence model connected to the connection host, wherein the artificial intelligence model trained based on the hemodialysis data, when a kidney dialysis patient undergoes hemodialysis treatment At this time, an auxiliary judgment reference is generated through the artificial intelligence model. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1 further includes: a display device connected to the connection host, and the display device displays the auxiliary determination reference. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 2, wherein the auxiliary determination reference is whether the obstruction requires interventional treatment or does not require interventional treatment. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein the connection host receives the hemodialysis data before an interventional processing action is performed as the hemodialysis data during obstruction. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein the connection host receives the hemodialysis data executed by an interventional processing action method as unobstructed hemodialysis data. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein each hemodialysis data includes a physiological detection value. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 6, wherein the physiological detection value includes a diastolic blood pressure value, a systolic blood pressure value, a pulse value, a respiration value, a blood flow value and a venous pressure value . The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein the artificial intelligence model includes a model training data set, and the model training data set includes multiple hemodialysis data during obstruction and multiple unobstructed blood The dialysis data, the hemodialysis data during obstruction and the hemodialysis data without obstruction respectively include a corresponding diastolic blood pressure value, a systolic blood pressure value, a pulse value, a respiration value, a blood flow value and a vein value. pressure value. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 8, wherein the number of hemodialysis data during obstruction is three. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 8, wherein the number of unobstructed hemodialysis data is five. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 8, wherein the number of unblocked hemodialysis data is greater than the number of blocked hemodialysis data. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein the number of the hemodialysis data is eight. The auxiliary determination system for renal dialysis vascular access obstruction as described in claim 1, wherein the connected host uses a type of neural network algorithm to modify the artificial intelligence model. As claimed in claim 13, the auxiliary determination system for renal dialysis vascular access obstruction, wherein the neural network algorithm is a random forest algorithm.

Images