Medical ward volume regulation system and method

Abstract

This invention discloses a volume control system and method for medical wards, relating to the field of medical environment management technology. The system includes: a patient status monitoring unit for outputting patient status signals; a sound acquisition and recognition unit for identifying ambient sound types; a scene marker input unit for receiving medical scene markers; a central control unit for fusing the patient status, sound type, and medical scene information, and outputting intervention instructions based on a pre-stored decision rule table; and a reminder execution unit for executing the instructions. The method includes corresponding steps. This invention changes the existing technology's single-mode alarm based solely on a fixed volume threshold. Through multi-source information fusion and intelligent decision-making, it can distinguish between necessary medical sounds and unnecessary interference noise, and dynamically adjust management strategies according to the patient's real-time status, achieving precise, tiered, and non-invasive volume control. This ensures patient rest while preventing interference with necessary medical activities.

Description

Technical Field

[0001] This invention relates to the field of medical environment management technology, specifically to a volume control system and method for medical wards. Background Technology

[0002] Maintaining a quiet environment in medical wards, especially inpatient wards and intensive care units, is crucial for patients' rest and recovery. Excessive noise can affect patients' sleep, increase psychological stress, and delay recovery.

[0003] Currently, noise management in hospital wards mainly relies on manual reminders and automated devices. Manual reminders are limited by the workload and subjective judgment of medical staff, making it difficult to achieve continuous, timely, and consistent responses. Regarding automated devices, the patent with publication number CN120126292A, entitled "A Ward Noise Monitoring and Reminder Device and Method Based on Customized Noise Monitoring," represents a typical approach in the prior art. This device collects noise by deploying a sensor array in the ward, calculates the noise value at the nurses' station using a weighted equivalent continuous sound level algorithm via an edge processor, and compares it with a preset threshold (e.g., 65 dB during the day and 55 dB at night). If the threshold is exceeded, an audible and visual alarm is triggered.

[0004] However, these existing technological solutions still have significant shortcomings and cannot meet the needs of refined clinical management:

[0005] 1. Rigid threshold settings fail to adapt to individual patient differences: The alarm thresholds switch only based on time (day / night), rather than on the patient's real-time physiological state (such as deep sleep, light sleep, or wakefulness). A postoperative patient in deep sleep and a patient reading while awake have completely different tolerances to noise, but the existing device cannot distinguish between them and still uses a uniform threshold for judgment.

[0006] 2. Inability to identify the nature of sound, posing a risk of interfering with medical activities: This device only monitors volume (decibels) and cannot identify the nature and content of sound. It treats necessary medical communication (such as emergency doctor consultations or instructions containing keywords like "rescue") the same as unnecessary chatter, noise from objects colliding, etc. This leads to the risk of the system interfering with necessary medical activities and makes it unable to distinguish control priorities in complex medical environments.

[0007] 3. The reminder method is singular and intrusive: The reminder method is mainly a uniform sound and light alarm, which itself constitutes new noise and causes additional sound pollution to the ward environment. It lacks specificity and humanization. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a medical ward volume control system and method based on patient status and scene recognition. This system aims to achieve an intelligent balance between ensuring patient rest and maintaining normal medical operations by distinguishing between "necessary sounds" and "avoidable noises" based on intelligent decision-making logic through multi-source information fusion, and dynamically adjusting management strategies according to "patient's current status" and "ongoing medical activities."

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A volume control system for medical wards includes a patient status monitoring unit, a sound acquisition and recognition unit, a scene marker input unit, a central control unit, and a reminder execution unit.

[0011] The core of the system lies in the fact that the central control unit can integrate three types of information from the patient status monitoring unit, the sound acquisition and recognition unit, and the scene labeling input unit—namely, the patient's real-time physiological state, the type of ambient sound, and the current medical scene—and make intelligent decisions, ultimately outputting differentiated and non-invasive interventions through the reminder execution unit.

[0012] The specific structure and function of each unit are as follows:

[0013] Patient Status Monitoring Unit: This unit monitors the patient's physiological signals in real time and categorizes them into actionable status categories, such as deep sleep, light sleep or rest, and wakefulness. It can be implemented using a non-contact sleep monitoring pad or mattress pressure sensor deployed on the hospital bed, and its analysis results can be obtained by connecting to existing bedside monitors.

[0014] Sound Acquisition and Recognition Unit: Used to collect environmental sounds in the ward and classify them. Specifically, it distinguishes between: equipment alarm sounds (such as alarms from monitors and infusion pumps), medical-related human voices (dialogues containing specific emergency keywords), and non-medical-related human voices (ordinary conversations, entertainment sounds, etc.). The recognition of medical-related human voices can be achieved through a speech recognition module containing a preset emergency medical keyword database (such as "rescue," "heart rate," and "doctor").

[0015] Scene labeling input unit: Provides a human-computer interaction interface for medical staff to quickly label the current medical activity scene, such as ward rounds, treatment, nursing, visits, or rest. This unit includes a touch screen electronic doorplate at the ward entrance, or a mobile terminal held by medical staff (which automatically associates the scene by scanning the patient's wristband and name tag).

[0016] Central Control Unit: This is the core of the system's intelligent decision-making. It contains a pre-stored decision rule table. This unit receives and integrates the information input from the three units mentioned above (patient status, voice type, medical scenario), and outputs corresponding intervention instructions by querying the decision rule table. The decision rule table uses patient status, medical scenario, and voice type as three query dimensions, and its rules reflect the priority intelligence of clinical management, mainly including (but not limited to):

[0017] When the patient is in a deep sleep and non-medical voices are detected, regardless of the current situation, a reminder instruction will be output to ensure the patient's rest with the highest priority.

[0018] When an alarm is detected, a non-intervention command is output regardless of the patient's condition or the medical setting, to ensure that medical safety information takes absolute priority.

[0019] Depending on the combination of different states and scenarios, different levels of reminder instructions can be output to achieve refined management.

[0020] The reminder execution unit executes tiered and flexible reminders based on instructions from the central control unit. Intervention instructions can include different levels such as no intervention, Level 1 reminder, and Level 2 reminder. Level 1 reminders can trigger light prompts (such as flashing indicator lights at the nurse station) or portable vibration reminders (worn by medical staff). Level 2 reminders, in addition to the above, also trigger a directional sound reminder device, emitting a locally audible alert sound in a specific direction to minimize secondary interference.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. Personalized and precise volume management: By using the patient's real-time physiological state (such as deep sleep, light sleep, and wakefulness) as the core decision variable, the system can distinguish the different tolerance levels of different patients to noise, realizing a shift from a "responsible for a uniform environment" to a "service for specific patients" management model, making the management strategy more individualized and targeted.

[0023] 2. Enhanced functionality and reliability of sound recognition: The system can distinguish between equipment alarm sounds, emergency medical communications, and unnecessary noise. This ensures the absolute priority of transmitting critical medical acoustic signals (such as equipment alarms and instructions containing emergency keywords), fundamentally avoiding the risk of false alarms interfering with normal medical activities and ensuring medical safety.

[0024] 3. Enhanced system context awareness and adaptability: By introducing medical scenario markers, the system can perceive the current type of medical activity (such as ward rounds, treatment, and visits). The volume management strategy can be dynamically adjusted according to specific scenarios. For example, in the "visiting" scenario, the management of non-medical voices can be appropriately relaxed, making the control more in line with the actual clinical workflow.

[0025] 4. Optimized the humanization and effectiveness of intervention methods: Adopted tiered, targeted, and non-invasive reminder methods (such as light prompts, directional sounds, and localized vibrations) to replace the traditional single sound and light alarm. This method achieves effective reminders while minimizing "secondary interference" to the ward environment, thus improving the acceptance of medical staff and patients. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall architecture and connection of the system of the present invention.

[0027] Figure 2 This is a schematic representation of the three-dimensional decision-making rules of the central control unit of the present invention.

[0028] Figure 3 This is a system workflow diagram of the present invention. Detailed Implementation

[0029] The technical solution of the present invention will now be described in detail and completely with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0030] Example 1

[0031] refer to Figure 1 This embodiment provides a volume control system for medical wards in general hospital wards. The system specifically includes: a patient status monitoring unit, a sound acquisition and recognition unit, a scene marker input unit, a central control unit, and a reminder execution unit.

[0032] The composition, connection, and operation of each unit are as follows:

[0033] Patient status monitoring unit: A commercially available non-contact sleep monitoring mattress (e.g., a mattress based on fiber optic pressure sensors or piezoelectric films) is placed under the bed sheet. This mattress reliably distinguishes between "deep sleep," "light sleep / rest," and "awake" states by analyzing the patient's body movement, respiration, and heart rate signals, and transmits these signals via Wi-Fi or Zigbee protocols. As a more economical alternative, it can connect to the existing multi-functional bedside monitor in the ward via a data interface module to obtain the patient's sleep status as determined by the monitor's built-in algorithm.

[0034] Sound Acquisition and Recognition Unit: Each bed unit is equipped with an intelligent audio module integrating a microphone array and a voice processing chip. This module has built-in acoustic event detection and keyword recognition algorithms, which can realize two core functions through local computation: (a) Voiceprint Classification: Through a pre-trained model, the sound is initially classified into "continuous regular sound" (such as equipment operation sound), "sudden alarm sound", or "human voice dialogue". (b) Keyword Recognition: When "human voice dialogue" is recognized, a lightweight keyword recognition engine is activated. This engine only listens to a small preset emergency medical keyword library (such as: "rescue", "emergency", "low heart rate", "low blood pressure", "call a doctor"). If any keyword is recognized, the voice segment is marked as "medical-related voice"; otherwise, it is marked as "non-medical-related voice". Specifically, after detecting the voice frequency band signal, the keyword recognition engine quickly compares the real-time audio stream with the preset emergency medical keyword library; if any keyword is matched within the set time window, the voice segment is immediately marked as "medical-related voice"; otherwise, it is marked as "non-medical-related voice".

[0035] Scene Marking Input Unit: A touchscreen electronic doorplate is installed at the entrance of each ward. When medical staff enter the ward to perform different operations, they click the corresponding button on the screen, such as "Ward Rounds," "Infusion / Treatment," "Measure Vital Signs," "Visit," or "Rest." The doorplate sends the scene marking information to the central control unit via the local area network. Alternatively, nurses are equipped with handheld PDAs. Before performing an operation, they scan the patient's wristband QR code and their own employee ID QR code. The PDA automatically uploads the scene information: "Nurse XXX performs treatment operation on patient XXX."

[0036] Central Control Unit: This can be a small industrial control computer or server deployed at the nurses' station. It receives data from the aforementioned units via a local area network. Its core function is to: integrate patient status signals from the patient status monitoring unit, ambient sound types from the sound acquisition and recognition unit, and current medical scene marker information from the scene marker input unit in real time; and make decisions based on a pre-stored decision rule table, outputting corresponding intervention instructions. The decision rule table is implemented in software logic and stored in memory; some rule examples are shown below (for reference). Figure 2 ):

[0037] Rule 1: If the sound type is the device alarm sound, then regardless of the patient's condition and the scenario, the "do not intervene" command will be output.

[0038] Rule 2: If the patient's state is deep sleep and the voice type is a non-medical voice, then regardless of the scenario, a "Level 2 Reminder" instruction will be output.

[0039] Rule 3: If the patient's status is light sleep / rest, the sound type is non-medical related human voice, and the scenario is visiting, then output the "Level 1 Reminder" instruction.

[0040] Rule 4: If the patient's status is conscious, output the "Do not intervene" command for non-medical related human voices.

[0041] The reminder execution unit includes multiple execution terminals:

[0042] Directional sound alert device: This is a small directional speaker (using ultrasonic directional sound technology or parabolic reflection technology) installed on the ceiling or wall of the bed area, pointing towards common conversational areas in the ward. When triggered, it emits a gentle alert sound that can only be clearly heard from a specific direction (such as towards a door or corridor), such as a soothing "ding-dong" sound or a pre-recorded voice message "Please be quiet".

[0043] The lighting indicator system consists of two parts. First, each bed has a corresponding LED status light (usually green) on the central monitoring screen at the nurses' station. Second, a ring-shaped LED light strip is added to the electronic doorplate at the ward entrance. When a reminder is needed, the LED light changes to amber and flashes slowly.

[0044] Portable vibration alert device: A smart bracelet or name tag vibrator equipped for healthcare workers, connected to a central control unit via Bluetooth. When it is necessary to alert healthcare workers in a specific area, their bracelet or name tag will generate a short vibration.

[0045] Combination Figure 3 Taking a nighttime nurse rounds scenario as an example, this demonstrates how the system implements volume control:

[0046] S301: System initialization, all units begin operation. The sleep monitoring mattress for patient 2 shows that the patient has entered a "deep sleep" state.

[0047] S302: Nurse A clicks "Measure vital signs" on the electronic doorplate at the ward entrance to mark the current scene.

[0048] S303: Nurse A enters the ward to measure the blood pressure of patient in bed 1. During this time, she speaks quietly with the conscious patient in bed 1. The voice acquisition and recognition unit captures the voice, but it does not contain any emergency keywords, so it is marked as "non-medical related voice".

[0049] S304: The central control unit receives input: patient status (bed 2 = deep sleep), scenario (measuring vital signs), and sound type (non-medical related human voice). The central control unit integrates the above three types of information and queries the decision rule table: according to rule 1 (non-medical human voice in deep sleep), the system determines that intervention should be performed.

[0050] S305: The central control unit further determines that the conversation is taking place in the area of ​​bed 1, but may affect the patient in bed 2. Since the patient is in a deep sleep, the system decides to issue a "secondary alert" instruction.

[0051] S306: Remind the execution unit to execute the "Level 2 Reminder" combination instruction:

[0052] The directional sound alert device installed near bed 2 is triggered, emitting a soft directional alert sound in the direction of nurse A.

[0053] The status light for bed 2 on the large screen at the nurses' station turned amber and began to flash.

[0054] Nurse A's smart bracelet vibrated slightly once.

[0055] S307: Nurse A receives a combined alert (hears a directional prompt tone and experiences wristband vibration), realizes that her conversation may disturb the sleeping patient in bed 2, and immediately lowers her volume or pauses the conversation. The sound acquisition and recognition unit 102 continues to monitor, and when the central control unit 104 determines that the real-time sound type no longer meets the triggering conditions of rule 1 (such as the human voice stopping or the volume significantly decreasing), the alert is lifted, and the status light returns to green.

[0056] The decision rule table in this embodiment can be pre-configured and adjusted later according to the specific needs of different wards. For example, in the pediatric ward, "children's crying" can be listed as a separate sound type, and different processing rules can be set (such as no intervention or only light prompts). The rule table can be visually edited through the management software interface of the nurse station, increasing the system's flexibility and adaptability.

[0057] Example 2

[0058] In another embodiment, the sound acquisition and recognition unit can be implemented in a simpler way. For example, a circuit module capable of distinguishing between "steady-state noise" and "transient noise" can be used, combined with a filter that only recognizes a specific frequency range (such as common frequencies of monitor alarm sounds). When an alarm sound of a specific frequency is detected, it is directly classified as "device alarm sound"; when a human voice frequency band is detected and its duration exceeds a set threshold, it is classified as "human voice" and is defaulted to "non-medical related human voice," unless the scenario is marked as an extreme case such as "rescue." This method is less costly, and although the accuracy is slightly reduced, it still achieves the core differentiation function.

[0059] Example 3

[0060] In another implementation of the reminder execution unit, the directional sound reminder device can be omitted. When a "secondary reminder" is needed, the system can control the intelligent lighting system in the ward, causing the lights to slowly change brightness several times (similar to a breathing light effect), while simultaneously coordinating with the nurse station light prompts and wristband vibration. This purely visual and tactile reminder method may be more gentle at night.

[0061] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A medical ward sound volume regulation system, characterized in that, include: The patient status monitoring unit is used to monitor the patient's physiological signals and output the patient status signals; A sound acquisition and recognition unit is used to acquire ambient sound signals and identify the type of the ambient sound signals; The scene labeling input unit is used to receive the current medical scene labeling information input by medical staff; The central control unit is communicatively connected to the patient status monitoring unit, the sound acquisition and recognition unit, and the scene marker input unit. The central control unit is used to fuse the patient status signal, the type of the environmental sound signal, and the current medical scene marker information, and output the corresponding intervention command. The reminder execution unit is communicatively connected to the central control unit and is used to execute the intervention command.

2. The system of claim 1, wherein, The patient status signals include at least a deep sleep state, a light sleep or rest state, and a conscious state.

3. The system of claim 2, wherein, The patient status monitoring unit is a non-contact sleep monitoring pad, a mattress pressure sensor, or an interface module connected to a bedside monitor.

4. The system according to claim 1, characterized in that, The types of ambient sound signals include at least device alarm sounds, medical-related human voices, and non-medical-related human voices.

5. The system of claim 4, wherein, The sound acquisition and recognition unit includes a speech recognition module, which is used to distinguish between medical-related voices and non-medical-related voices by recognizing preset emergency medical keywords.

6. The system of claim 1, wherein, The central control unit has a pre-stored decision rule table, which is used to query and output the intervention command based on a combination of the patient status signal, the type of the environmental sound signal, and the current medical scene marker information.

7. The system of claim 6, wherein, The decision rule table is configured to: output a reminder instruction when the patient is in a deep sleep state and the sound type is a non-medical related human voice; output a no-intervention instruction when the sound type is a device alarm sound.

8. The system of claim 1, wherein, The reminder execution unit includes at least one of a directional sound reminder device, a light reminder device, and a portable vibration reminder device.

9. A method for regulating sound volume in a medical ward, characterized by, Includes the following steps: Patient status signals are acquired through a patient status monitoring unit; The type of environmental sound signal is collected and identified by the sound acquisition and recognition unit; The current medical scene marking information input by medical staff is obtained through the scene marking input unit; The central control unit integrates the patient status signal, the type of the environmental sound signal, and the current medical scene marker information, and outputs the corresponding intervention command; The execution unit is reminded to execute the intervention command.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method of claim 9.

Images