CN-121971216-A - Intelligent restraint belt system for intensive care unit

CN121971216ACN 121971216 ACN121971216 ACN 121971216ACN-121971216-A

Abstract

The application discloses an intelligent restraint strap system for an intensive care unit, which comprises an intelligent restraint glove and a medical restraint strap risk assessment subsystem, wherein a buffer air bag is arranged on the wrist part of the intelligent restraint glove, a pressure sensor is arranged in the buffer air bag, doppler is arranged at a wrist compression position corresponding to the inner side of the intelligent restraint glove, the medical restraint strap risk assessment subsystem is used for receiving pressure data acquired by a film pressure sensor and physiological data acquired by Doppler, extracting pressure characteristics and physiological risk characteristics, carrying out data fusion on the pressure characteristics and the physiological risk characteristics, inputting the fusion characteristics into a three-level intelligent risk assessment model for comprehensive assessment to obtain an assessment result, and triggering corresponding-level acousto-optic alarm and nursing reminding if the assessment result exceeds a dynamically set risk threshold. The method can accurately predict the complications and the risk that the patient breaks loose from the restraint strap, and timely gives out early warning, so that potential safety hazards caused by over-tightening or over-loosening of restraint are effectively avoided.

Inventors

WU SIJIA
YANG LUO
LI YUTIAN
LIN JINMIN
GUO KAILU
WANG ZIYUAN
CHEN MINQI
WANG YIFEI

Assignees

浙江中医药大学

Dates

Publication Date: 20260505
Application Date: 20251230

Claims (10)

1. An intensive care unit intelligent restraint strap system, comprising: The intelligent constraint glove comprises a glove body, wherein a constraint belt body is arranged on the wrist part of the glove body, a buffer air bag is fixedly arranged at a key wrist pressed part of the constraint belt body, a thin film pressure sensor is fixedly arranged in the buffer air bag, doppler is fixedly arranged at a key wrist pressed part corresponding to the inner side of the constraint belt body, a main controller and a first communication module are fixedly arranged in the constraint belt body, and the thin film pressure sensor, the Doppler and the first communication module are electrically connected with the main controller; A medical restraint strap risk assessment subsystem comprising a second communication module communicatively connected to the first communication module, the medical restraint strap risk assessment subsystem for implementing the steps of the medical restraint strap risk assessment method; The medical restraint strap risk assessment method comprises the steps of receiving pressure data collected by the film pressure sensor and physiological data collected by Doppler, preprocessing the pressure data and the physiological data, extracting pressure features and physiological risk features respectively, fusing the pressure features and the physiological risk features to obtain fusion features, inputting the fusion features into a pre-built three-level intelligent risk assessment model for comprehensive assessment to obtain an assessment result, and triggering corresponding-level audible and visual alarm and nursing reminding if the assessment result exceeds a dynamically set risk threshold.
2. The intensive care unit intelligent restraint strap system of claim 1, wherein the restraint strap body adopts a multi-layer structure, and the multi-layer structure comprises a skin-friendly cotton layer, a buffer sponge layer and a waterproof moisture permeable outer layer from inside to outside in sequence.
3. The intelligent restraint system of the intensive care unit according to claim 1, wherein the pressure characteristic is extracted through a convolutional neural network, and the convolutional neural network sequentially comprises three convolutional layers, two pooling layers and one full-connection layer.
4. The intelligent restraint system of the intensive care unit according to claim 1, wherein the physiological risk features are extracted through a hybrid time sequence neural network, and the structure of the hybrid time sequence neural network sequentially comprises a time sequence convolution layer and a long-short-term memory network layer.
5. The intensive care unit intelligent restraint system according to claim 1, wherein the three-level intelligent risk assessment model comprises an instant risk detection, a trend risk assessment and a comprehensive risk assessment, wherein the instant risk detection adopts a preset threshold value for instant assessment, the trend risk assessment adopts an instant assessment result and a time sequence model for risk development trend assessment, and the comprehensive risk assessment adopts a multi-mode risk assessment model for comprehensive risk assessment.
6. The intelligent intensive care unit restraint system of claim 5, wherein the data weight of each modality is dynamically adjusted according to individual constitution and distance confidence score of the patient when the multi-modality risk assessment model performs comprehensive risk assessment.
7. The intensive care unit intelligent restraint system of claim 1, wherein the doppler employs color doppler.
8. The intensive care unit intelligent restraint strap system of claim 1, wherein the restraint strap body is fixedly provided with a velcro or quick release buckle on an outer surface thereof.
9. The intensive care unit intelligent restraint strap system of claim 1, wherein the first communication module is a bluetooth low energy module.
10. The intensive care unit intelligent restraint strap system of claim 1, further comprising a lithium battery fixedly disposed within the restraint strap body and electrically connected to the main controller.

Description

Intelligent restraint belt system for intensive care unit Technical Field The application relates to the technical field of medical equipment, in particular to an intelligent restraint strap system for an intensive care unit. Background In an Intensive Care Unit (ICU), there is a risk of unplanned tube drawing, self-injury, etc. in patients with agitation or consciousness disturbance, so that the prior art often adopts a traditional cloth or leather restraint strap to fix the parts of the patient such as the wrist. Although the restraint bands are widely applied clinically, the technology is stagnated for a long time, and the requirements of modern ICU on safety, precision and high-efficiency nursing are difficult to meet. Early-stage focusing material design optimization and intelligent monitoring technology development in European and American developed countries and searching alternatives are adopted to reduce constraint. In clinical application, there are methods of using molecular materials to reduce the risk of pressure sores and frictional damage, integrating sensors to monitor various data, and using pressure feedback systems to prevent tissue ischemia, and there are also studies currently being conducted on iterative optimization of constraining bands to make them more in line with clinical demands. However, the conventional restraint bands still have the following problems: 1. Passive hysteresis, namely, the struggling or break-off behavior of a patient is discovered by depending on regular patrolling of nurses, the alarm cannot be actively given out in real time, a safe blind area exists, response is delayed, and unexpected danger is easily caused by the behavior in ICU wards. 2. Complications cannot be pre-warned, and for complications such as local pressure sores and oedema caused by the restraint strap, the traditional restraint strap cannot detect early physiological signals (such as local tissue blood flow changes) of the restraint strap, and the early physiological signals can be found only after injury is seen by naked eyes, so that a good preventive opportunity is missed. 3. Lack of quantification, tightness of the restraint strap is judged according to the experience of nurses, and for the nurses with lack of experience, skin pressure sores, nerve injuries and unsmooth blood circulation are easily caused by too tight restraint, and restraint failure is possibly caused by too loose restraint. 4. The efficiency is lower, frequent manual inspection, adjustment and record, have increased medical personnel's work burden obviously. Disclosure of Invention Therefore, the application provides an intelligent restraint strap system for an intensive care unit, which aims to solve the problem that the restraint strap in the prior art cannot accurately predict complications and break loose risks in advance. In order to achieve the above object, the present application provides the following technical solutions: An intensive care unit intelligent restraint strap system, comprising: The intelligent constraint glove comprises a glove body, wherein a constraint belt body is arranged on the wrist part of the glove body, a buffer air bag is fixedly arranged at a key wrist pressed part of the constraint belt body, a thin film pressure sensor is fixedly arranged in the buffer air bag, doppler is fixedly arranged at a key wrist pressed part corresponding to the inner side of the constraint belt body, a main controller and a first communication module are fixedly arranged in the constraint belt body, and the thin film pressure sensor, the Doppler and the first communication module are electrically connected with the main controller; A medical restraint strap risk assessment subsystem comprising a second communication module communicatively connected to the first communication module, the medical restraint strap risk assessment subsystem for implementing the steps of the medical restraint strap risk assessment method; The medical restraint strap risk assessment method comprises the steps of receiving pressure data collected by the film pressure sensor and physiological data collected by Doppler, preprocessing the pressure data and the physiological data, extracting pressure features and physiological risk features respectively, fusing the pressure features and the physiological risk features to obtain fusion features, inputting the fusion features into a pre-built three-level intelligent risk assessment model for comprehensive assessment to obtain an assessment result, and triggering corresponding-level audible and visual alarm and nursing reminding if the assessment result exceeds a dynamically set risk threshold. Preferably, the restraint strap body adopts a multilayer structure, and the multilayer structure sequentially comprises a skin-friendly cotton layer, a buffer sponge layer and a waterproof moisture-permeable outer layer from inside to outside. Preferably, the pressure characteristic is extracted thro