Intelligent weighing feeding bowl and intelligent control method thereof
The intelligent weighing feeding bowl design solves the problem of accurately controlling the "one bite" during the eating process for patients with swallowing disorders. It enables real-time monitoring and individualized adjustment, reduces risks, and is suitable for safe feeding control in multiple scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TONGJI HOSPITAL ATTACHED TO TONGJI MEDICAL COLLEGE HUAZHONG SCI TECH
- Filing Date
- 2025-12-02
- Publication Date
- 2026-05-05
AI Technical Summary
Existing technologies lack precise control over the "one bite" portion when managing the feeding process for patients with dysphagia. Traditional tableware relies on experience and is inaccurate, while smart tableware suffers from unstable weighing and a lack of in-process safety warning mechanisms. High-end equipment is expensive and unsuitable, and cannot achieve individualized adjustments and real-time intervention.
A smart weighing feeding bowl was designed, comprising a suction cup structure, a weighing module, a main control unit, an alarm prompting unit, and a communication module. Through real-time weighing and algorithm analysis, it can achieve precise control over the amount and pace of food intake, and issue audible and visual alarms before dangerous behavior occurs. It also supports data synchronization with mobile terminals and personalized parameter settings.
It enables precise control over the feeding process, reduces the risks of aspiration, choking, and suffocation, provides objective data support and individualized adaptability, and is suitable for hospital and home rehabilitation scenarios, improving safety and practicality.
Smart Images
Figure CN121970977A_ABST
Abstract
Description
Technical Field
[0001] This specification relates to the field of medical care, and more specifically, this application relates to an intelligent weighing feeding bowl and its intelligent control method. Background Technology
[0002] Dysphagia is a common complication of stroke, Parkinson's disease, Alzheimer's disease, and other neurological disorders. Its main manifestations include delayed swallowing reflexes, impaired muscle coordination, and decreased pharyngeal control, leading to serious consequences such as aspiration, choking, and even suffocation during eating. For these patients, the key to safe eating lies in precisely controlling the amount of food put into the mouth each time, the so-called "one-bite." If the bite is too small, it may not effectively trigger the swallowing reflex, causing food to remain in the mouth or pharynx; while if the bite is too large, it can easily lead to aspiration, airway obstruction, and other high-risk events, which can result in suffocation or aspiration pneumonia in severe cases.
[0003] Currently, the methods used to control "one-time doses" in clinical and home care settings are still relatively primitive, and mainly have the following shortcomings: (1) Traditional tableware relies on experience: such as ordinary spoons and teaspoons, which are not precise in terms of capacity. Nursing staff or patients often rely on visual estimation to take food, which is greatly affected by subjective judgment and hand stability, making it difficult to achieve precise control. (2) Medical syringes or feeding devices are highly invasive: Although they can achieve a certain degree of quantitative control, their use is unnatural and can easily cause patients to feel disgusted or resistant. In addition, the flow rate adjustment is rough and they are not suitable for oral feeding training. (3) The feeding spoon or cup has a single function: Existing products usually only have a fixed capacity and lack dynamic control functions for feeding speed, interval time and swallowing rhythm, and cannot be intelligently adjusted according to the patient's condition; (4) High-end nasogastric feeding pump equipment has limited applicability: such equipment is bulky and expensive, and is mainly used for nasal or gastric tube feeding, and is not suitable for oral rehabilitation feeding scenarios; (5) The core problems of existing smart tableware have not been solved: Although some conceptual smart tableware integrates weighing or timing functions, they generally have two key defects: First, the bowl is easy to move, which leads to unstable weighing data; second, there is a lack of active safety warning mechanism in the process, which cannot intervene in time when "excessive" or "too fast" eating occurs.
[0004] Therefore, it is necessary to provide an intelligent weighing feeding bowl and its intelligent control method to at least solve some of the above-mentioned problems. Summary of the Invention
[0005] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. This summary section is not intended to limit the key and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.
[0006] Firstly, this application proposes an intelligent weighing feeding bowl, comprising: Bowl body and smart base; The aforementioned smart base includes a suction cup structure, a weighing module, a main control unit, an alarm notification unit, a communication module, and a power management module; The suction cup structure is used to fix the bowl to a smooth tabletop; the weighing module is used to detect the weight data of the food in the bowl in real time; the main control unit is used to process the weight data and generate alarm prompts; the alarm prompt unit includes a ring-shaped multi-color LED indicator and a buzzer, which are used to issue an audible and visual alarm when there is excessive food or when eating too quickly, based on the alarm prompts; the communication module is used to interact with the mobile terminal.
[0007] In one feasible implementation, the bowl is made of food-grade material, and the bowl is connected to the smart base via a detachable structure.
[0008] In one feasible implementation, the suction cup structure includes a flexible silicone suction cup and a mechanical locking mechanism to enhance fixation stability and prevent weighing errors.
[0009] In one feasible implementation, the alarm unit displays the current status using different colored indicator lights, where green indicates that it is safe to eat, yellow indicates that the amount is close to the safe amount, and red flashing accompanied by a buzzer indicates that the amount is excessive.
[0010] In one feasible implementation, the communication module uses Bluetooth or BLE technology to achieve wireless pairing, parameter setting, and data synchronization of the APP inside the mobile terminal.
[0011] Secondly, the present invention proposes an intelligent control method for the intelligent weighing feeding bowl described in any of the claims of the first aspect, comprising: Control the weighing module to collect the weight data of the food in the bowl; The control unit executes an over-limit judgment algorithm based on a preset single safety threshold. When the decrease in the weight data exceeds the threshold within a preset time window, an alarm command is generated and the alarm prompt unit is triggered to issue an audible and visual alarm. The control unit executes a feeding interval determination algorithm based on a preset swallowing interval time. When the interval between consecutive feedings is detected to be less than the threshold, the alarm unit is controlled to issue a prompt signal. The control module transmits the weight curve, alarm events, and timestamps recorded during the feeding process to the mobile terminal simultaneously.
[0012] In one feasible implementation, it further includes: When executing the above-mentioned overload determination algorithm, the main control unit combines the slope of the weight curve collected by the weighing module, food type parameters and real-time ambient temperature, and uses an adaptive correction factor to dynamically adjust the single safety threshold.
[0013] In one feasible implementation, it further includes: When executing the feeding interval determination algorithm, the main control unit uses a sliding time window to calculate the distribution characteristics of the number of consecutive feedings and the swallowing interval, forming a rhythm risk index. The rhythm risk index is automatically divided into normal zone, attention zone and danger zone according to the feeding frequency and fluctuation trend. When the aforementioned rhythm risk index enters the attention zone, the aforementioned alarm unit will flash blue to indicate this. When the aforementioned rhythm risk index enters the danger zone, the aforementioned alarm unit flashes red and blue alternately and sends an alarm message to the aforementioned mobile terminal.
[0014] In one feasible implementation, it further includes: While outputting an audible and visual alarm, the alarm unit also sends a tactile feedback command to the mobile terminal via the communication module. Upon receiving the aforementioned feedback command, the wearable device connected to the mobile terminal generates a slight vibration or sound signal to remind the caregiver to perform intervention or pause the feeding operation.
[0015] In one feasible implementation, it further includes: After receiving the feeding data, the mobile terminal uses an incremental learning algorithm based on the historical feeding sample set to update the single safety threshold and swallowing interval parameters online. The update weight is adaptively adjusted according to the number of times the patient eats, the alarm frequency, and the rehabilitation assessment results.
[0016] In summary, the intelligent weighing feeding bowl proposed in this embodiment demonstrates significant improvements and technical effects in structural design, functional integration, and safety control. Firstly, regarding structural stability and weighing accuracy, this embodiment achieves secure fixation of the bowl on a smooth tabletop by incorporating a suction cup structure beneath the intelligent base, supplemented by a mechanical locking design. This effectively prevents slippage or tipping caused by hand tremors or accidental contact during the feeding process. The stable support formed by the suction cup and the base not only improves the mechanical stability of the weighing process but also reduces noise interference from external vibrations. This allows the weighing module to accurately acquire weight signals at a high sampling rate, thereby obtaining a stable, continuous, and vibration-free weight curve, providing a reliable data foundation for subsequent algorithm analysis. Compared to traditional tableware that relies on manual control or experience-based judgment, this embodiment achieves dual guarantees of physical stability and data accuracy. Secondly, regarding safety and intelligent early warning, this embodiment introduces proactive intervention safety control logic. The main control unit analyzes the changes in the weighing curve in real time and can automatically trigger an audible and visual alarm when it detects a weight loss exceeding a preset threshold within a short period. This allows the system to provide immediate warnings and interventions at the moment a dangerous behavior occurs, thus advancing the safety precautions to the critical moment "before the food enters the mouth." When the system detects that the interval between consecutive feedings is too short, it issues a rhythm reminder to guide the operator to slow down the feeding pace. This proactive intervention mechanism based on real-time monitoring significantly improves the safety of patients during feeding compared to existing passive recording devices or devices only used for data playback. It reduces the probability of high-risk events such as aspiration, choking, and suffocation, achieving a functional leap from "post-event observation" to "in-event protection." Furthermore, in terms of individualization and intelligent adaptation, the main control unit in this embodiment, combined with parameters set on the mobile terminal, can flexibly adjust the single-session safety threshold and minimum swallowing interval based on different patients' swallowing abilities, food viscosity, and recovery stages. Nursing staff or rehabilitation therapists can set and send parameters through the mobile terminal interface to achieve precise control over different patients and different meals. Meanwhile, the system can also gradually optimize threshold settings through algorithmic self-learning to better match the patient's recovery pace. This personalized dynamic adaptation capability makes this embodiment suitable not only for hospital environments but also for stable operation in home rehabilitation scenarios. Furthermore, in terms of data management and rehabilitation assessment, the communication module of this embodiment uses Bluetooth or BLE technology to connect with the mobile terminal in real time, enabling the synchronous uploading of weight curves, alarm events, and timestamps during the eating process. The system automatically generates detailed eating reports after meals, including multi-dimensional indicators such as total food intake, average single scoop volume, eating rhythm distribution, and number of violations, providing objective and quantitative data support for medical staff. Unlike traditional tableware that relies solely on experience for observation, this embodiment achieves digital recording and visual analysis of eating behavior, providing doctors with precise evidence to assess the progress of swallowing function recovery and adjust training plans.Finally, regarding practicality and energy efficiency management, this embodiment implements sleep and wake-up control through a power management module, entering a low-power mode when not in use, thereby extending battery life and reducing maintenance costs. Simultaneously, the interactive button design supports one-click reset, Bluetooth pairing, and device start / stop operations, facilitating rapid deployment and standardized use. The overall structure is lightweight, easy to assemble and disassemble, and convenient to clean, meeting both the stability and safety requirements of hospital nursing processes and the ease of use and comfort needs of home users.
[0017] Other advantages, objectives and features of this application will be apparent in part from the description which follows, and in part from what those skilled in the art will understand through study and practice of this application. Attached Figure Description
[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a structural schematic diagram of an intelligent weighing feeding bowl provided in an embodiment of this application; Figure 2 This is a schematic diagram of another intelligent weighing feeding bowl structure provided in an embodiment of this application; Figure 3 This is a flowchart illustrating an intelligent control method provided in an embodiment of this application. Figure 4 A schematic diagram illustrating weight changes and overfeed alarms during the feeding process, provided as an embodiment of this application; Figure 5 This is a schematic diagram of a mobile smart eating assistant interface provided in an embodiment of this application. Detailed Implementation
[0019] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. The technical solutions of the embodiments of this application will now be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.
[0020] Please see Figure 1 and Figure 2 The diagram below illustrates the structure of an intelligent weighing feeding bowl provided in this application embodiment, which may specifically include: Firstly, this application proposes an intelligent weighing feeding bowl, comprising: Bowl body 101 and smart base 102; The aforementioned smart base 102 includes a suction cup structure 1021, a weighing module 1022, a main control unit 1023, an alarm prompting unit 1024, a communication module 1025, and a power management module 1026. The suction cup structure 1021 is used to fix the bowl 101 to a smooth tabletop; the weighing module 1022 is used to detect the weight data of the food in the bowl in real time; the main control unit 1023 is used to process the weight data and generate an alarm prompt command; the alarm prompt unit 1024 includes a ring-shaped multi-color LED indicator and a buzzer, and is used to issue an audible and visual alarm when there is excessive food or eating too quickly based on the alarm prompt command; the communication module 1025 is used to interact with a mobile terminal.
[0021] For example, the intelligent weighing feeding bowl consists of a bowl body 101 and an intelligent base 102. The intelligent base 102 integrates a suction cup structure 1021, a weighing module 1022, a main control unit 1023, an alarm display unit 1024, a communication module 1025, and a power management module 1026. The suction cup structure 1021 is located on the lower surface of the base, using negative pressure to firmly fix the bowl body 101 to a smooth tabletop, preventing weighing errors and food spillage caused by the bowl slipping during feeding by the patient or caregiver. This provides stable mechanical boundary conditions for subsequent weight acquisition and algorithmic judgment. The weighing module 1022 is positioned above the suction cup structure 1021 and is electrically connected to the main control unit 1023. It is used to acquire the weight signal of the food inside the bowl body 101 with high sensitivity and short cycle and output a weight change curve. The alarm display unit 1024 has a ring-shaped multi-color LED indicator around the side wall of the base, and works with a buzzer to present three states: safe, warning, and danger, using different colors and flashing rhythms. The aforementioned communication module 1025 establishes a synchronization channel with the mobile terminal using Bluetooth or BLE protocol for parameter transmission and data reporting. The aforementioned power management module 1026 includes a rechargeable battery and a voltage regulator / sleep circuit to ensure portability and low-power operation.
[0022] In actual use, the caregiver first fixes the smart base 102 to a smooth surface such as a bedside table or dining table using the suction cup structure 1021, assembles the bowl 101, and places the target food inside. Then, in the settings interface of the mobile terminal, the caregiver sets the "maximum allowable amount to be scooped at one time" (e.g., 10g) and the "minimum swallowing interval" (e.g., 15s) for the individual patient, and completes pairing with the communication module 1025. Once ready, the system begins real-time monitoring: the weighing module 1022 outputs a weight sequence at a high sampling rate, and the main control unit 1023 performs noise reduction and first-order slope analysis on the sequence to identify feeding events of "sudden weight drop-plateau period" and calculates the amount scooped at one time and the time interval between two adjacent feedings. When a weight drop exceeding a preset threshold is detected within a short time window (e.g., 2-3s), the main control unit 1023 immediately generates an alarm command, driving the alarm prompting unit 1024 to rapidly flash a red LED and activate a buzzer, prompting the operator to reduce the amount of food in the spoon or stop feeding. When the interval between two consecutive valid feedings is detected to be less than the set safe swallowing time, the alarm prompt unit 1024 switches to a rhythmic reminder (e.g., flashing blue) to guide a slower pace. During normal feeding, the ring LED remains solid green; it flashes yellow when approaching the threshold for early intervention.
[0023] Throughout the feeding process, the main control unit 1023 continuously records the weight curve generated by the weighing module 1022, the overfeeding events and overfeeding events determined by the algorithm, and their timestamps. This data is then synchronized to the mobile terminal's visual interface via the communication module 1025. A feeding report is automatically generated after the meal, including indicators such as total food intake, average amount scooped per feeding, feeding duration, swallowing interval distribution, and number of violations. This information is used by medical staff to assess changes in the patient's swallowing function and optimize rehabilitation plans. Through the mechanical stability provided by the suction cup fixation and the proactive safety mechanism of in-process audio-visual reminders, this embodiment achieves a synergistic effect in four aspects: "avoiding measurement deviations caused by displacement," "preemptively blocking the risk of overfeeding," "standardizing the feeding rhythm," and "providing objective quantitative data." Unlike conventional tableware and measuring spoons that only passively record data, this embodiment enables real-time, reliable, and patient-centered safety management in oral feeding scenarios.
[0024] It should be noted that the thresholds and intervals can be individually adjusted via mobile terminals to match different food viscosities and patient stages. The sleep / wake-up strategy of the aforementioned power management module 1026 can also reduce standby power consumption during non-working periods. One-click reset (pepidocalization), Bluetooth pairing, and device start / stop can also be achieved via interactive buttons, facilitating standardized nursing procedures. These optional features, working in conjunction with the aforementioned core components, ensure that the system meets both the stability requirements of clinical nursing and the ease of use and maintainability needs of home settings.
[0025] In summary, the intelligent weighing feeding bowl proposed in this embodiment demonstrates significant improvements and technical effects in structural design, functional integration, and safety control. Firstly, regarding structural stability and weighing accuracy, this embodiment achieves secure fixation of the bowl on a smooth tabletop by incorporating a suction cup structure beneath the intelligent base, supplemented by a mechanical locking design. This effectively prevents slippage or tipping caused by hand tremors or accidental contact during the feeding process. The stable support formed by the suction cup and the base not only improves the mechanical stability of the weighing process but also reduces noise interference from external vibrations. This allows the weighing module to accurately acquire weight signals at a high sampling rate, thereby obtaining a stable, continuous, and vibration-free weight curve, providing a reliable data foundation for subsequent algorithm analysis. Compared to traditional tableware that relies on manual control or experience-based judgment, this embodiment achieves dual guarantees of physical stability and data accuracy. Secondly, regarding safety and intelligent early warning, this embodiment introduces proactive intervention safety control logic. The main control unit analyzes the changes in the weighing curve in real time and can automatically trigger an audible and visual alarm when it detects a weight loss exceeding a preset threshold within a short period. This allows the system to provide immediate warnings and interventions at the moment a dangerous behavior occurs, thus advancing the safety precautions to the critical moment "before the food enters the mouth." When the system detects that the interval between consecutive feedings is too short, it issues a rhythm reminder to guide the operator to slow down the feeding pace. This proactive intervention mechanism based on real-time monitoring significantly improves the safety of patients during feeding compared to existing passive recording devices or devices only used for data playback. It reduces the probability of high-risk events such as aspiration, choking, and suffocation, achieving a functional leap from "post-event observation" to "in-event protection." Furthermore, in terms of individualization and intelligent adaptation, the main control unit in this embodiment, combined with parameters set on the mobile terminal, can flexibly adjust the single-session safety threshold and minimum swallowing interval based on different patients' swallowing abilities, food viscosity, and recovery stages. Nursing staff or rehabilitation therapists can set and send parameters through the mobile terminal interface to achieve precise control over different patients and different meals. Meanwhile, the system can also gradually optimize threshold settings through algorithmic self-learning to better match the patient's recovery pace. This personalized dynamic adaptation capability makes this embodiment suitable not only for hospital environments but also for stable operation in home rehabilitation scenarios. Furthermore, in terms of data management and rehabilitation assessment, the communication module of this embodiment uses Bluetooth or BLE technology to connect with the mobile terminal in real time, enabling the synchronous uploading of weight curves, alarm events, and timestamps during the eating process. The system automatically generates detailed eating reports after meals, including multi-dimensional indicators such as total food intake, average single scoop volume, eating rhythm distribution, and number of violations, providing objective and quantitative data support for medical staff. Unlike traditional tableware that relies solely on experience for observation, this embodiment achieves digital recording and visual analysis of eating behavior, providing doctors with precise evidence to assess the progress of swallowing function recovery and adjust training plans.Finally, regarding practicality and energy efficiency management, this embodiment implements sleep and wake-up control through a power management module, entering a low-power mode when not in use, thereby extending battery life and reducing maintenance costs. Simultaneously, the interactive button design supports one-click reset, Bluetooth pairing, and device start / stop operations, facilitating rapid deployment and standardized use. The overall structure is lightweight, easy to assemble and disassemble, and convenient to clean, meeting both the stability and safety requirements of hospital nursing processes and the ease of use and comfort needs of home users.
[0026] In one feasible implementation, the bowl is made of food-grade material, and the bowl is connected to the smart base via a detachable structure.
[0027] For example, the bowl is made of food-grade high-temperature resistant material, and the inner surface can be coated with an anti-slip or easy-to-clean coating to ensure both safety and hygiene when holding liquid or semi-liquid foods. The bowl and the smart base are connected by a detachable structure. After disassembly and assembly, the main control unit can automatically or via a one-button operation to reset the tare weight, ensuring accurate zero-point weighing upon reuse. This reduces cumulative errors caused by long-term use and facilitates daily cleaning and replacement of bowls of different capacities or shapes.
[0028] In one feasible implementation, the suction cup structure includes a flexible silicone suction cup and a mechanical locking mechanism to enhance fixation stability and prevent weighing errors.
[0029] For example, the suction cup structure includes a flexible silicone suction cup and a mechanical locking mechanism: the flexible silicone suction cup is located at the bottom of the smart base, and negative pressure is created by pressing to achieve initial fixation; the mechanical locking mechanism is located in the annular pressure-bearing area above the suction cup, and a secondary locking is achieved by rotation or pressing after adsorption, ensuring the base remains stable when subjected to lateral forces on the table. When releasing the fixation, the negative pressure is released first through a pressure relief structure or unlocking action before the mechanical locking mechanism is released, avoiding the impact caused by roughly pulling out the weighing module. This significantly reduces the weighing error and food spillage risk caused by bowl slippage while ensuring rapid installation / disassembly efficiency.
[0030] In one feasible implementation, the alarm unit displays the current status using different colored indicator lights, where green indicates that it is safe to eat, yellow indicates that the amount is close to the safe amount, and red flashing accompanied by a buzzer indicates that the amount is excessive.
[0031] For example, the alarm unit described above is equipped with a ring of multi-color LED indicator lights around the outer periphery of the smart base, and works in conjunction with a buzzer to distinguish different safety levels: when the system is in a feedable state, the indicator light is solid green to indicate "feeding is permitted"; when the system detects that the amount of food scooped up at one time is close to a preset threshold or that the feeding pace is too fast, the indicator light switches to yellow or blue and flashes rhythmically to provide a prior warning; when the weight loss exceeds the threshold within a preset time window, the indicator light flashes red rapidly and is accompanied by a buzzer, prompting the operator to immediately reduce the amount of food or stop feeding, thereby moving safety intervention to the critical moment "before the food enters the mouth", reducing the risk of aspiration and choking.
[0032] In one feasible implementation, the communication module uses Bluetooth or BLE technology to achieve wireless pairing, parameter setting, and data synchronization of the APP inside the mobile terminal.
[0033] For example, the aforementioned communication module uses Bluetooth or BLE technology to establish an encrypted connection with the mobile terminal APP, supporting initial pairing, remote parameter distribution, and real-time data transmission. Caregivers can set individualized parameters such as single-session safety thresholds and swallowing intervals in the APP, and view information such as weight curves, event markers, and cumulative intake during meals. After meals, the APP automatically generates reports based on the uploaded time series, including indicators such as total food intake, average single-serving volume, eating duration, rhythm distribution, and number of violations. Simultaneously, it can synchronize device logs and battery status for maintenance and management, thereby achieving a closed loop of hardware monitoring and software evaluation.
[0034] The second aspect, such as Figure 3 As shown, this invention proposes an intelligent control method for the intelligent weighing feeding bowl described in any of the first aspects, comprising: S210. Control the above-mentioned weighing module to collect the weight data of the food in the bowl; S220. The control unit executes an over-limit judgment algorithm based on a preset single safety threshold. When the decrease in the weight data exceeds the threshold within a preset time window, an alarm command is generated and the alarm prompt unit is triggered to issue an audible and visual alarm. S230. The main control unit executes a feeding interval determination algorithm based on a preset swallowing interval time. When the interval between consecutive feedings is detected to be less than the threshold, the alarm prompting unit is controlled to issue a prompt signal. S240. Control the above-mentioned communication module to synchronously transmit the weight curve, alarm events and timestamps recorded during the eating process to the above-mentioned mobile terminal.
[0035] For example, in step S210, the weighing module is controlled to continuously collect the weight data of the food in the bowl in a short period of time and output the weight change curve in real time. The weighing module is electrically connected to the main control unit to ensure that the collected weight signal can be processed and judged in real time, providing basic data for subsequent identification of "single feeding event" and "swallowing interval".
[0036] In step S220, the main control unit performs an over-limit judgment based on a preset "maximum permissible amount to scoop in a single instance" threshold: if a significant drop in the weight curve is detected within a preset short time window (e.g., 2-3 seconds), and the drop exceeds the threshold, an alarm command is generated and the alarm prompt unit (ring-shaped multi-color LED and buzzer) is triggered to issue an audible and visual alarm, thereby intervening in the process before the food is placed in the mouth, reminding the operator to reduce the amount of food in the spoon or pause feeding. This step, through the combination of real-time weighing and immediate alarm, moves the safety defense line to the critical node "before the entrance".
[0037] In step S230, the main control unit performs feeding rhythm determination based on a preset minimum swallowing interval. The algorithm identifies a "weight drop-plateau" time pattern as a valid feeding; starting from the point of the drop, if a new round of "drop" is detected again within the set safe interval, it is determined that the feeding is too fast, and the alarm prompt unit outputs a rhythm reminder (such as changing the indicator light color and flashing rhythm) to guide the operator to slow down the feeding frequency and extend the swallowing interval, thereby reducing the risk of choking and aspiration and standardizing the nursing rhythm.
[0038] In step S240, the weight curve, overfeed alarm events, and excessively fast alert events, along with their timestamps, generated throughout the entire eating process are recorded by the main control unit and synchronized with the mobile terminal via a Bluetooth / BLE channel established through the communication module. The mobile terminal visualizes the entire process in the form of a graph and automatically generates a report after the meal containing indicators such as total food intake, average single scoop volume, eating duration, rhythm distribution, and number of violations, providing objective evidence for medical staff to assess changes in swallowing function and formulate rehabilitation prescriptions. Through the above-mentioned data recording and visualization analysis, compared with conventional tableware that only passively records data, this embodiment can stably, accurately, and quantifiably achieve closed-loop management of "fixed-monitoring-judgment-early warning-recording".
[0039] In one feasible implementation, it further includes: When executing the above-mentioned overload determination algorithm, the main control unit combines the slope of the weight curve collected by the weighing module, food type parameters and real-time ambient temperature, and uses an adaptive correction factor to dynamically adjust the single safety threshold.
[0040] For example, to reduce false alarms / false negatives caused by different food forms and environmental fluctuations affecting the determination of a fixed threshold, the main control unit introduces an adaptive correction mechanism when executing the overload determination algorithm: First, the weighing module outputs a discrete sequence of weight over time at a high sampling rate. The main control unit performs noise reduction and first derivative calculation on this sequence to obtain the slope of the weight curve and the short-term drop. The basic threshold is preset by the nursing staff on the mobile device according to an individualized prescription (e.g., "maximum allowable amount to be scooped at one time" 10 g, time window 2~3 s), and sent to the device via Bluetooth / BLE as the initial values of T0 and Δt. Subsequently, the main control unit combines the feeding speed reflected by the slope of the weight curve, food type parameters (e.g., labeled as "liquid / semi-solid / solid" or "low / medium / high viscosity" in the APP), and real-time ambient temperature to calculate the adaptive correction factor κ = f(v, c, τ), dynamically generating the working threshold T = κ·T0. When high viscosity or slow feeding speed is detected, the threshold range is widened (the value of κ is increased), while when low viscosity, fast feeding speed, or high ambient temperature (increased fluidity) is detected, the threshold range is tightened (the value of κ is decreased). To avoid frequent fluctuations, the system employs a sliding window and upper / lower limit clamping strategy for κ (e.g., κ ∈ [0.7, 1.3]), and displays a short yellow flash indicating "threshold has been adaptively adjusted" when the threshold change exceeds the set range, ensuring the operator has an intuitive understanding of the current judgment sensitivity.
[0041] In the specific judgment process, the main control unit still follows the core rule of "if the significant decrease within a short time window is greater than the working threshold T, an alarm will be triggered": when Δw(Δt) > T, an alarm command is immediately output, driving the ring-shaped multi-color LED to switch to rapid red flashing and triggering a buzzer to sound, prompting the operator to reduce the amount of data or pause the process. If Δw(Δt) is close to T (e.g., ≥ 0.8T), a warning state is entered, and the indicator light flashes yellow rhythmically to intervene in advance. The above adaptive threshold only changes the "trigger boundary value" and does not change the basic criterion of "short time window detection," thus maintaining the interpretability and verifiability of the algorithm while taking into account sensitivity.
[0042] To ensure traceability and tunability, the system synchronizes metadata such as the "basic threshold T0, κ at each time point, actual trigger threshold T, corresponding decrease Δw, time window Δt, and alarm / warning timestamp" along with the weight curve to the mobile device after each meal. When generating the eating report, the mobile device displays not only the total food intake, average single serving size, rhythm distribution, and number of violations, but also overlays the adaptive trajectory of the threshold over time. This facilitates medical staff in reviewing the sensitivity of judgments under different food and environmental conditions and further fine-tuning individualized prescription parameters. This design maintains consistency with the existing framework of "parameters set and distributed on the APP" and "full-process curve and event visualization recording," achieving a smooth expansion from fixed thresholds to adaptive thresholds.
[0043] In one feasible implementation, it further includes: When executing the feeding interval determination algorithm, the main control unit uses a sliding time window to calculate the distribution characteristics of the number of consecutive feedings and the swallowing interval, forming a rhythm risk index. The rhythm risk index is automatically divided into normal zone, attention zone and danger zone according to the feeding frequency and fluctuation trend. When the aforementioned rhythm risk index enters the attention zone, the aforementioned alarm unit will flash blue to indicate this. When the aforementioned rhythm risk index enters the danger zone, the aforementioned alarm unit flashes red and blue alternately and sends an alarm message to the aforementioned mobile terminal.
[0044] For example, to further quantify the overall safety of the eating rhythm beyond simply judging whether it is too fast, the main control unit introduces a sliding time window mechanism when executing the eating interval judgment algorithm: using the most recent N (e.g., N=5~10) effective eating events as a window, it calculates the time interval sequence {Δt1,Δt2,…,Δt_N} between two adjacent eating events and extracts its distribution characteristics (such as mean, standard deviation / coefficient of variation, short-term upward or downward trend, etc.) to form a rhythm risk index R. The basic "effective eating event" is still identified by the "sudden drop-plateau" pattern of the weight curve. The interval is determined by timing from the point of sudden drop, and the set minimum swallowing interval is used as the safety baseline. When a new sudden drop is detected within the baseline, it is judged as too fast. This event is also included in the window statistics to update the calculation of R, so that the rhythm assessment is consistent with the single violation judgment.
[0045] The rhythm risk index R is automatically divided into three levels based on feeding frequency and fluctuation trends: "Normal Zone" is marked when the average interval is significantly higher than the safety baseline and the fluctuation is low (stable rhythm); "Payment Zone" is marked when the average interval is close to the baseline or there are frequent, slight advances in a short period (e.g., multiple feedings near the baseline) and the fluctuation increases; and "Danger Zone" is marked when there are consecutive excessively fast events or the average interval is significantly lower than the baseline and accompanied by high fluctuation (rhythm disorder). To ensure immediate and perceptible feedback, the alarm prompt unit is linked to this classification: when entering the attention zone, the ring-shaped multi-color LED flashes blue rhythmically, prompting the operator to slow down the feeding frequency. When entering the danger zone, the LED flashes red and blue alternately and a buzzer sounds, providing a stronger indication that immediate intervention or pausing feeding is necessary. These color / sound and light prompts are consistent with the system's original multi-color indicators and buzzer strategy, allowing the operator to quickly understand the current rhythm status without checking the mobile device.
[0046] At the data closed-loop level, the rhythm risk index R, as well as the timestamps of each interval and excessively fast events within the window, are synchronized to the mobile terminal along with the weight curve through the communication module. The front-end interface displays the data visually by overlaying the time series curve with "attention / danger" segments. Based on this, caregivers can replay rhythm changes, identify triggers (such as a sudden increase in feeding speed), and adjust parameter configurations (such as appropriately increasing the minimum swallowing interval or enabling stricter reminder strategies during high-risk periods).
[0047] In one feasible implementation, it further includes: While outputting an audible and visual alarm, the alarm unit also sends a tactile feedback command to the mobile terminal via the communication module. Upon receiving the aforementioned feedback command, the wearable device connected to the mobile terminal generates a slight vibration or sound signal to remind the caregiver to perform intervention or pause the feeding operation.
[0048] For example, to simultaneously alert caregivers in addition to the patient's audio-visual cues, the alarm unit, upon detecting overeating or eating too quickly, not only drives the ring-shaped multi-color LED and outputs the corresponding audio-visual alarm (e.g., rapid red flashing and buzzing when overeating, flashing a warning color when approaching the threshold or eating too quickly), but also sends a "tactile feedback command" to the mobile terminal via the communication module. This enables tactile or acoustic alerts to be triggered by the wearable device on the terminal side. The audio-visual cues, color semantic correspondence, and buzzer coordination are all consistent with the system's existing alarm strategy, ensuring that caregivers can intuitively assess the risk level and react quickly without having to look at the terminal screen.
[0049] On the communication link, the communication module establishes an encrypted connection with the mobile terminal's APP using Bluetooth / BLE. This connection is used for parameter setting and data transmission, as well as for pushing control or notification messages when an event is triggered. This embodiment reuses this link, sending "haptic feedback commands" as a type of event notification to the mobile terminal. The mobile terminal then forwards these commands to paired wearable devices (such as wristbands or clip-on vibrators) to generate slight vibrations or short beeps, reminding caregivers to immediately reduce the amount of food, slow down the pace, or pause feeding briefly. Since this link is consistent with the existing "wireless pairing, parameter setting, and data synchronization" channel, no new hardware is required. Expansion can be achieved simply by adding message types at the application layer, maintaining system consistency and low power consumption.
[0050] In the collaborative workflow, after the main control unit completes the event determination, it first drives the sound and light alarm locally, and then reports the event and tactile feedback command simultaneously through the communication module. After receiving the command, the mobile terminal immediately triggers the vibration of the wearable device and highlights the current event and timestamp on the front-end interface. Then it continues to record the weight curve, event type and processing action to provide evidence for post-event review.
[0051] In one feasible implementation, it further includes: After receiving the feeding data, the mobile terminal uses an incremental learning algorithm based on the historical feeding sample set to update the single safety threshold and swallowing interval parameters online. The update weight is adaptively adjusted according to the number of times the patient eats, the alarm frequency, and the rehabilitation assessment results.
[0052] For example, after receiving the weight curve, overfeed / overspeed events, and timestamps synchronized from the communication module, the mobile terminal merges the new meal data with the existing historical eating sample set, and performs small-step adaptive updates to the two core parameters, "single-time safety threshold" and "minimum swallowing interval," based on an online / incremental learning framework. The mobile terminal first extracts statistical and temporal features (such as single-mouth volume distribution, short-window slope of eating speed, rhythm stability index, and alarm occurrence rate) for each meal, and uses these features along with the phased rehabilitation assessment labels (such as swallowing training grades) from the healthcare provider as learning input. The algorithm uses the current prescription parameters as initial values and minimizes the weighted objective of "unnecessary alarm rate and missed alarm risk" through an error-driven iterative strategy (such as constrained linear regression or online gradient updates with a forgetting factor), thereby obtaining fine-tuned individualized parameters. To ensure clinical safety and interpretability, the system sets dual constraints on parameter updates: firstly, the update magnitude of the threshold and interval is constrained by session-level upper and lower bounds (e.g., ±10%) and a moderate learning rate is used. Secondly, when abnormal fluctuations occur continuously or alarm frequencies surge briefly, a conservative rollback strategy and manual review prompts are automatically triggered to ensure that parameters do not drift excessively due to occasional data noise. The above learning and adjustment process is seamlessly integrated with the existing architecture of "parameters set and distributed by the APP" and "full-process data visualization and report generation," which utilizes historical data to achieve individualization while maintaining the stability and auditability of the device-side judgment logic.
[0053] Regarding the weighting mechanism, the incremental learning update weights are adaptively adjusted based on the patient's cumulative eating frequency, alarm frequency, and external rehabilitation assessment results: when the patient eats more frequently and the recent alarm frequency decreases and the rhythm becomes more stable, the system appropriately increases the influence of new data on the parameters to accelerate convergence towards a more lenient but still safe threshold. When there are frequent alarms or medical assessments indicate a phased decline in swallowing function, the system reduces the weight of new data and tightens the threshold / lengthens the interval to prioritize ensuring the safety boundary. Each updated parameter and corresponding evidence (time window, feature summary, update magnitude, and reason) are recorded in the post-meal report and device log, facilitating playback, verification, and one-click rollback when necessary by medical staff. Simultaneously, all updates are sent to the device via the existing Bluetooth / BLE channel, ensuring subsequent judgments operate under the new individualized prescription.
[0054] like Figure 4The diagram shows the weight change curve during the feeding process of this invention. The vertical axis represents the real-time weight in the bowl, and the horizontal axis represents time. At the initial moment (0s), the weight of the food in the bowl is approximately 130g. After the patient finishes the first feeding, the weight in the bowl drops to approximately 112g, indicating that the amount consumed in a single bite is approximately 18g. This amount is less than the system's preset safety threshold (e.g., 20g), so no alarm is triggered. Subsequently, the system enters the swallowing interval timing stage, where the main control unit begins to determine whether the time interval meets the safe rhythm. When the interval reaches the set value (e.g., 15s), the system restores the green indicator, prompting that feeding can continue. During the second feeding, a weight decrease of approximately 12g is detected within a short time window, exceeding the set threshold. Therefore, the main control unit immediately generates an alarm command, triggering the alarm prompt unit, causing the ring LED indicator to flash red and accompanied by a buzzer, marking the "overfeeding alarm point" shown in the diagram. This curve clearly shows that the system can accurately identify single feeding behaviors during real-time monitoring, distinguish between normal and overfeeding events, and provide visual feedback on the safety status. This dynamic curve is not only used for real-time monitoring, but also serves as the basis for subsequent data analysis.
[0055] like Figure 5 The image shows a schematic of the "Smart Feeding Assistant" interface on a mobile terminal. This interface is divided into two parts: a parameter setting area and a real-time data area. In the parameter setting area, caregivers can input the safe single-serving amount (e.g., 10g), swallowing interval (e.g., 15s), and alarm sensitivity level (e.g., medium) according to the patient's condition, and click "Start Monitoring" to establish a Bluetooth connection with the smart bowl. The real-time data area displays the currently monitored weight (e.g., 85g) and status (e.g., "Eating normally"), and simultaneously plots the data against the monitor's readings. Figure 4 The system displays the corresponding feeding curve. As feeding progresses, the system automatically calculates the total amount of food consumed (e.g., 150g), the number of overfeeding incidents (e.g., 2 times), and the feeding duration (e.g., 12 minutes) at the bottom of the interface. If overfeeding or excessively rapid feeding occurs, a corresponding alarm will flash or vibrate on the mobile interface to alert the caregiver to intervene.
[0056] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A smart weighing feeding bowl, characterized in that, include: Bowl body and smart base; The intelligent base includes a suction cup structure, a weighing module, a main control unit, an alarm notification unit, a communication module, and a power management module; The suction cup structure is used to fix the bowl to a smooth tabletop; the weighing module is used to detect the weight data of the food in the bowl in real time; the main control unit is used to process the weight data and generate alarm prompts; the alarm prompt unit includes a ring-shaped multi-color LED indicator and a buzzer, used to issue an audible and visual alarm when the food is overeaten or eaten too quickly based on the alarm prompt; the communication module is used to interact with the mobile terminal.
2. The intelligent weighing feeding bowl according to claim 1, characterized in that, The bowl is made of food-grade material, and the bowl is connected to the smart base via a detachable structure.
3. The intelligent weighing feeding bowl according to claim 1, characterized in that, The suction cup structure includes a flexible silicone suction cup and a mechanical locking mechanism to enhance fixation stability and prevent weighing errors.
4. The intelligent weighing feeding bowl according to claim 1, characterized in that, The alarm unit displays the current status using different colored indicator lights. Green indicates that it is safe to eat, yellow indicates that the amount is close to the safe amount, and red flashing accompanied by a buzzer indicates that the amount is excessive.
5. The intelligent weighing feeding bowl according to claim 1, characterized in that, The communication module uses Bluetooth or BLE technology to enable wireless pairing, parameter setting, and data synchronization of the APP within the mobile terminal.
6. An intelligent control method for the intelligent weighing feeding bowl according to any one of claims 1 to 5, characterized in that, include: The weighing module is controlled to collect the weight data of the food in the bowl; The control unit executes an over-limit judgment algorithm based on a preset single-time safety threshold. When the decrease in weight data exceeds the threshold within a preset time window, an alarm command is generated and the alarm prompt unit is triggered to issue an audible and visual alarm. The main control unit executes a feeding interval determination algorithm based on a preset swallowing interval time. When the interval between consecutive feedings is detected to be less than the threshold, the alarm prompting unit is controlled to issue a prompt signal. The control module synchronously transmits the weight curve, alarm events, and timestamps recorded during the eating process to the mobile terminal.
7. The intelligent control method according to claim 6, characterized in that, Also includes: When executing the overload determination algorithm, the main control unit combines the slope of the weight curve collected by the weighing module, food type parameters, and real-time ambient temperature, and uses an adaptive correction factor to dynamically adjust the single safety threshold.
8. The intelligent control method according to claim 6, characterized in that, Also includes: When the main control unit executes the feeding interval determination algorithm, it uses a sliding time window to calculate the distribution characteristics of the number of consecutive feedings and the swallowing interval to form a rhythm risk index. The rhythm risk index is automatically divided into normal zone, attention zone and danger zone according to the feeding frequency and fluctuation trend. When the rhythm risk index enters the attention zone, the alarm prompt unit will flash blue to indicate this. When the rhythm risk index enters the danger zone, the alarm prompt unit flashes red and blue alternately and sends an alarm message to the mobile terminal.
9. The intelligent control method according to claim 6, characterized in that, Also includes: While outputting an audible and visual alarm, the alarm notification unit sends a tactile feedback command to the mobile terminal through the communication module. The wearable device connected to the mobile terminal generates a slight vibration or sound signal after receiving the feedback command to remind the caregiver to perform intervention or pause the feeding operation.
10. The intelligent control method according to claim 6, characterized in that, Also includes: After receiving the feeding data, the mobile terminal uses an incremental learning algorithm based on the historical feeding sample set to update the single safety threshold and swallowing interval parameters online. The update weight is adaptively adjusted according to the number of times the patient eats, the alarm frequency, and the rehabilitation assessment results.