Peripheral venous indwelling needle model selection intelligent recommendation system
By using an intelligent recommendation system to select the model of peripheral intravenous catheters, and by employing decision models and multi-parameter fusion, the problem of inaccurate model selection in existing technologies has been solved, achieving higher decision accuracy and safety.
Patent Information
- Application Number
- CN202510716883.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-30
- Publication Date
- 2025-10-21
AI Technical Summary
In the existing technology, the selection of peripheral intravenous catheter models mainly relies on the visual experience of medical staff, resulting in inaccurate assessment and may cause adverse medical events such as vascular damage, leakage, phlebitis and catheter blockage.
An intelligent recommendation system for selecting peripheral intravenous catheter models is adopted. The system inputs decision parameters through an input module, performs logical decision-making using a pre-stored decision model, outputs matching decision results, integrates an alarm module for early warning, and combines multi-parameter fusion and historical data optimization to reduce human error.
This improved the accuracy of model selection, reduced adverse events such as leakage, phlebitis, and catheter blockage, and enhanced the accuracy and safety of decision-making.
Smart Images
Figure CN120824039A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical equipment, and in particular to an intelligent recommendation system for selecting peripheral venous catheter models. Background Art
[0002] An intravenous catheter is a common infusion tool used clinically. It primarily consists of a stainless steel core needle, a soft outer cannula positioned outside the core needle, and a plastic needle holder. During puncture, the outer cannula and core needle are inserted into the blood vessel together. Once the cannula is inserted, the core needle is withdrawn, leaving the soft outer cannula in the vessel for infusion. Currently, the selection of peripheral intravenous catheter size (e.g., 18G, 20G, 22G, 24G, etc.) in clinical practice relies primarily on visual judgment and experience by medical personnel. This presents the following challenges: It is highly subjective, with individual choices potentially variably made by different medical personnel; and vascular assessment is imprecise: Determining vessel diameter, elasticity, and blood flow solely through visual inspection or palpation can easily lead to size mismatch, potentially damaging small vessels or compromising infusion speed. Furthermore, incorrect selection can lead to adverse medical events such as leakage, phlebitis, and catheter clogging. Therefore, developing a peripheral intravenous catheter size selection device to overcome these existing challenges is a topic for further research by those skilled in the art. Summary of the Invention
[0003] The purpose of the present invention is to provide an intelligent recommendation system for peripheral venous catheter model selection, which can improve the accuracy of peripheral venous catheter model selection and reduce the occurrence of adverse medical events such as phlebitis.
[0004] The present invention discloses an intelligent recommendation system for selecting a peripheral venous catheter model, which includes:
[0005] An input module, wherein the input module is used to input decision parameters;
[0006] a decision module, the decision module being signal-connected to the input module and having a pre-stored decision model therein, the decision module being configured to input the decision parameters into the pre-stored decision model, perform a logical decision operation based on the decision model, and generate a decision result that matches the decision parameters;
[0007] An output module is connected to the decision module by signal and is used to output the decision result.
[0008] This technical solution allows users to select a peripheral IV catheter model by feeding relevant data from the input module into the decision module. The module then uses a pre-stored decision model to replace human experience in making decisions and outputs the results. This avoids the inaccurate assessments often associated with visual judgment by medical staff, thereby reducing the incidence of adverse medical events such as leakage, phlebitis, and catheter blockage.
[0009] Preferably, in the above-mentioned intelligent recommendation system, the decision parameters include a reference array composed of human body reference data and treatment reference data; the human body reference array includes a blood vessel diameter parameter, a blood vessel elasticity parameter, and a blood vessel depth parameter; the treatment reference data includes an infusion rate and a drug type;
[0010] The decision model is used to generate a decision result that matches the reference array; and to form a mapping relationship between the reference array and the decision result.
[0011] This technical solution enables a multi-parameter fusion-based decision-making process during execution, further improving the accuracy of decisions. By generating a decision result that maps one-to-one with a reference array composed of multiple parameters, it ensures that the decision result is targeted for complex data.
[0012] Preferably, the above intelligent recommendation system further includes:
[0013] An alarm module is pre-written with alarm logic, and the alarm module is signal-connected to the input module, and is used to read the decision parameters and generate and output an alarm prompt when the decision parameters match the alarm logic.
[0014] By adopting this technical solution:
[0015] Preferably, in the above intelligent recommendation system, the input module includes:
[0016] A manual input device, the manual input device is used to manually input the pre-collected decision parameters and transmit them to the decision module; the manual input device includes an input keyboard or is integrated into a portable touch screen;
[0017] A detection input device is used to realize on-site collection of the decision parameters and transmit the collected decision parameters to the decision module; the detection device includes at least one of a portable infrared sensor / a portable optical sensor.
[0018] Preferably, the above intelligent recommendation system further includes: a training module,
[0019] The training module is used to input intervention data and train the decision module based on the intervention data.
[0020] By adopting this technical solution: through the collection and input of intervention data, the decision-making module itself is iteratively optimized based on historical data feedback.
[0021] Preferably, in the above intelligent recommendation system, the intervention data includes a data set consisting of historical puncture success data and historical puncture failure data.
[0022] More preferably, the training module is based on the decision module trained by reinforcement learning / deep learning training method.
[0023] Preferably, the above intelligent recommendation system further includes: a communication module, wherein the communication module is configured to be signal-connected to the hospital's HIS system and used to access the hospital's HIS system.
[0024] By adopting this technical solution: This solution realizes remote access to the hospital HIS system through the communication module, thereby obtaining patient information such as gender, age, puncture side surgery history / trauma history, disease history, etc. based on the hospital HIS system, and combining it as an auxiliary judgment basis for medical staff on site, further improving accuracy.
[0025] Compared with the prior art, the present invention has the following advantages:
[0026] First, the present invention can make decisions based on pre-stored decision models instead of human experience, avoiding the problem of inaccurate assessments based on the visual experience of medical staff, thereby reducing the occurrence of adverse medical events such as leakage, phlebitis, and catheter blockage.
[0027] Secondly, the present invention can realize a decision-making process based on multi-parameter fusion during execution, further improving the accuracy of decision-making.
[0028] Third, the present invention can continuously iteratively optimize the decision model based on feedback from historical data.
[0029] Fourthly, the present invention can provide early warning for risk scenarios such as small blood vessels and those involving hypertonic solutions.
[0030] Finally, the present invention has a simple structure and is easy to prepare and operate. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 This is a functional module block diagram of Example 1.
[0032] Figure 2 This is a structural diagram of Example 1.
[0033] Figure 3This is a functional module block diagram of Example 2.
[0034] Figure 4 This is a functional module block diagram of Example 3.
[0035] Figure 5 This is a functional module block diagram of Example 4.
[0036] The corresponding reference numerals of the various components in the figure are as follows:
[0037] 100, input module; 200, decision module; 300, output module; 400, alarm module; 500, training module; 700, communication module; 600, outer shell; 610, display screen; 620, power-on button; 630, electronic input screen; 640, wiring port; 650, signal socket. DETAILED DESCRIPTION
[0038] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.
[0039] Example 1, please refer to Figure 1-2 :
[0040] An intelligent recommendation system for selecting peripheral intravenous catheter models includes: an input module 100, a decision module 200, and an output module 300.
[0041] The input module 100 is used to input decision parameters; the decision module 200 is connected to the input module 100 by signal, and the decision module 200 has a pre-stored decision model. The decision module 200 is used to input the decision parameters into the pre-stored decision model, perform logical decision operations based on the decision model, and generate a decision result that matches the decision parameters; the output module 300 is connected to the decision module 200 by signal and is used to output the decision result. Specifically:
[0042] The decision parameters include a reference array composed of human body reference data and treatment reference data; the human body reference array includes blood vessel diameter parameters, blood vessel elasticity parameters, and blood vessel depth parameters; the treatment reference data includes infusion rate and drug type; the decision model is used to generate a decision result that matches the reference array; and the reference array and the decision result form a mapping relationship.
[0043] In practice, the input module 100 may include at least one of a manual input device and a detection input device. The manual input device is used to manually input and transmit pre-collected decision parameters to the decision module 200. Manual input devices are generally used to input data such as vascular diameter parameters, vascular elasticity parameters, and vascular depth parameters. Vascular diameter parameters can be pre-collected through ultrasound or magnifying caliper measurement, vascular elasticity parameters can be pre-collected through palpation and grading, and vascular depth can be pre-collected based on ultrasound data. The detection input device is used to collect the decision parameters on-site and transmit the collected decision parameters to the decision module 200. The detection device may include at least one of a portable infrared sensor and a portable optical sensor. The portable infrared / optical sensor is used to measure vascular diameter parameters in real time. In this example, the input module 100, the decision module 200, and the output module 300 are integrated within an overall rectangular outer shell 600. The surface of the outer shell 600 is provided with a display screen 610, a power button 620, an electronic input screen 630, a wiring port 640, and a signal plug port 650. The display screen 610 is used to implement the output module 300's function of outputting the decision result. The electronic input screen 630 is used to implement the manual input device's function of manually inputting the pre-collected decision parameters. The power button 620 is used to control the power on / off of the input module 100, the decision module 200, and the output module 300. The wiring port 640 is configured to electrically connect the outer shell 600 to an external power source. The signal plug port 650 is configured to connect the outer shell 600 to the detection device (i.e., the portable infrared / optical sensor) for signals.
[0044] The operating process is as follows: When selecting a peripheral intravenous catheter model, data related to the catheter model is input into the decision module 200 via the input module 100. A decision is made based on a pre-stored decision model in the decision module 200, replacing human experience, and a decision result is generated. The decision result is then outputted via the output module 300. In practice, the decision module 200 is constructed based on the standards outlined in the "Clinical Guideline Logic" and can convert the decision result into a visual table format that includes a vascular image and a model comparison chart.
[0045] Example 2, please refer to Figure 2 、 3 :
[0046] Compared to Example 1, Example 2 further discloses an alarm module 400 based on the structure of Example 1. This alarm module 400 is also integrated into the outer housing 600. Alarm logic is pre-programmed into the alarm module 400, which is signal-connected to the input module 100. The alarm module 400 is configured to read the decision parameters and generate and output an alarm when the decision parameters match the alarm logic. The display screen 610 also implements the function of the output module 300 to output the decision result.
[0047] In practice, the alarm module 400 is mainly used to implement pre-alarms for clinical situations where central venous catheterization is required due to vascular stenosis, hyperosmotic fluid, etc. For example, the alarm logic can be implemented according to the following process: one or more items related to the alarm are decomposed from the decision parameters and logical operations are performed. Taking the vascular diameter parameter as an example, the corresponding vascular diameter alarm threshold is pre-written, and the vascular diameter parameter is subtracted from the vascular diameter alarm threshold. When the result of the operation is less than 0, an alarm is triggered and an alarm prompt is output. In this example, an alarm picture is pre-stored in the alarm module 400; the alarm prompt can be implemented by retrieving and displaying the pre-stored alarm picture on the display screen 610.
[0048] Example 3, please refer to Figure 2 、 4 :
[0049] Compared with Example 2, Example 3 further discloses:
[0050] The training module 500 is used to input intervention data and train the decision module 200 based on the intervention data. The training module 500 is also integrated into the outer shell 600 and is signal-connected to the decision module 200. The intervention data includes a dataset consisting of historical puncture success data and historical puncture failure data. The training module 500 trains the decision module 200 based on a reinforcement learning / deep learning training approach. In practice, the intervention data is also input via the input module 100. The training process is initiated by a medical professional inputting control commands via the electronic input screen 630.
[0051] Example 4, please refer to Figure 2 、 5 :
[0052] Compared with Example 3, Example 4 further discloses:
[0053] The communication module 700 is connected to the input module 100, the output module 300, and the hospital HIS system for signal communication, and is used to access the hospital HIS system. In practice, the communication module 700 can access the hospital HIS system through a wired / wireless signal connection. When the communication module 700 is connected to the hospital HIS system through a wired signal connection, it is connected to an external signal line through the wiring port 640. When the communication module 700 is connected to the hospital HIS system through a wired signal connection, it can be connected through a Wi-Fi module integrated in the outer shell 600.
[0054] When selecting the model of a peripheral venous catheter, medical staff can input relevant data into the decision module through the input module to obtain the decision result. At the same time, they can also remotely access the hospital's HIS system through the input module via the communication module, thereby obtaining patient information such as gender, age, puncture side surgical history / trauma history, and disease history based on the hospital's HIS system, and make a comprehensive judgment based on the obtained decision result, thereby further improving the accuracy of the human decision-making in the final model selection.
[0055] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the scope of protection of the present invention.
Claims
1. An intelligent recommendation system for peripheral venous catheter model selection, characterized in that: include: An input module, wherein the input module is used to input decision parameters; a decision module, the decision module being signal-connected to the input module and having a pre-stored decision model therein, the decision module being configured to input the decision parameters into the pre-stored decision model, perform a logical decision operation based on the decision model, and generate a decision result that matches the decision parameters; An output module is connected to the decision module by signal and is used to output the decision result.
2. The intelligent recommendation system according to claim 1, characterized in that: The decision parameters include a reference array composed of human body reference data and treatment reference data; the human body reference array includes blood vessel diameter parameters, blood vessel elasticity parameters, and blood vessel depth parameters; the treatment reference data includes infusion rate and drug type; The decision model is used to generate a decision result that matches the reference array; and to form a mapping relationship between the reference array and the decision result.
3. The intelligent recommendation system according to claim 2, characterized in that: Also includes: An alarm module is pre-written with alarm logic, and the alarm module is signal-connected to the input module, and is used to read the decision parameters and generate and output an alarm prompt when the decision parameters match the alarm logic.
4. The intelligent recommendation system according to claim 3, characterized in that: The input module includes: A manual input device, the manual input device is used to manually input the pre-collected decision parameters and transmit them to the decision module; the manual input device includes an input keyboard, or is integrated into a portable touch screen; A detection input device is used to realize on-site collection of the decision parameters and transmit the collected decision parameters to the decision module; the detection device includes at least one of a portable infrared sensor / a portable optical sensor.
5. The intelligent recommendation system according to claim 4, characterized in that: Also includes: A training module is used to input intervention data and train the decision module based on the intervention data.
6. The intelligent recommendation system according to claim 5, characterized in that: The intervention data includes a data set consisting of historical puncture success data and historical puncture failure data.
7. The intelligent recommendation system according to claim 6, characterized in that: The training module is based on the decision module trained by reinforcement learning / deep learning training method.
8. The intelligent recommendation system according to claim 7, characterized in that: It also includes: a communication module, which is configured to be signal-connected to the hospital HIS system and used to access the hospital HIS system.