Intelligent reservation number calling system and method thereof

Abstract

The invention discloses an intelligent reservation number calling system and a method thereof, and belongs to the technical field of medical information. The system comprises a data acquisition module, an intelligent estimation module, a resource collaborative scheduling module and a patient service module. By collecting static and dynamic treatment data of a patient, personalized dynamic estimation of the treatment duration is realized; generating an optimized daily scheduling plan based on multi-dimensional constraints such as mold type collaboration, equipment load and human state; and the queuing information and the equipment state are pushed to the patient in real time. According to the invention, the problems of uncertain treatment duration, non-cooperative resource scheduling and non-transparent information of the radiotherapy center are solved, and comprehensive improvement of accurate reservation, efficient scheduling and patient experience is realized.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and in particular to an intelligent appointment and queuing system and method for radiotherapy centers. Background Technology

[0002] Radiotherapy is one of the main methods of cancer treatment. As a core treatment device, the operational efficiency of linear accelerators and the level of patient management directly affect a hospital's diagnostic and treatment capabilities and the patient experience. With the increasing number of cancer patients, radiotherapy centers need to receive a large number of patients daily, placing higher demands on the intelligence level of appointment and queuing systems.

[0003] Currently, several general-purpose intelligent queuing systems exist. For example, patent document CN116884133A discloses an intelligent appointment queuing method for the financial sector, which manages queues through online check-in and geolocation verification; patent document CN112184980B discloses an intelligent queuing system for hospitals, which verifies patient identity through a verification module to prevent erroneous queuing. However, these systems have significant limitations when applied to the specific scenario of a radiotherapy center: Lack of personalized and dynamic treatment duration estimation: Existing systems typically assign patients fixed time slots based on average estimates. However, a single radiotherapy session is complex, involving multiple steps such as patient preparation, positioning, scanning, registration, and beam exit. Its total duration is influenced by factors such as the patient's self-care ability, the therapist's skill level, and the specific treatment plan, exhibiting significant individual differences and dynamic variations. A generic fixed-duration appointment system cannot accommodate this complexity, leading to frequent discrepancies between the actual treatment process and the plan, with cumulative delays severely impacting overall efficiency.

[0004] Ignoring the coordinated scheduling of specific physical resources: During radiotherapy, patients require specific positioning fixation molds (such as body membranes, vacuum pads, etc.), and different types of molds may require changing the base plate of the treatment bed. Existing systems do not consider this critical physical resource, resulting in patients with different types of molds being frequently interspersed in the scheduling, increasing the non-therapeutic workload of therapists, reducing efficiency, and accelerating equipment wear.

[0005] Lack of transparency and system foresight: Patients cannot obtain accurate real-time treatment progress and machine status (such as malfunctions or maintenance), which can easily lead to anxiety and blind waiting. At the same time, the system cannot predict the risk of machine overload based on historical data and real-time load, making it difficult to issue timely warnings to management personnel and proactively prevent the increased risk of malfunctions due to continuous high-load operation of equipment and complaints caused by excessively long waiting times.

[0006] Therefore, there is an urgent need in this field for a dedicated intelligent appointment and queuing system and method that can dynamically and accurately predict treatment duration, coordinate and optimize the scheduling of treatment resources, and improve information transparency, tailored to the characteristics of radiotherapy centers. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent appointment and queuing system and method. This system and method can solve the problems of uncertain treatment time for patients in radiotherapy centers, lack of coordination in physical resource scheduling, and lack of transparency in system status, thereby achieving accurate appointment, efficient scheduling, and optimized patient experience.

[0008] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent appointment and queuing system, comprising: Data acquisition module: used to acquire the patient's static characteristic data and dynamic treatment data; the static characteristic data includes the patient's radiotherapy prescription, treatment plan, type of body positioning device, type of imaging scan, and self-care ability rating; the dynamic treatment data includes the actual time taken for each stage of the patient's historical treatment. Intelligent prediction module: connected to the data acquisition module, used to make an initial prediction based on the static feature data, and integrate the dynamic treatment data of the patient's first two beam exit treatments, dynamically output and continuously correct the predicted duration of the patient's single treatment through a preset algorithm model; Resource Coordination Scheduling Module: Connected to the intelligent prediction module, it is used to generate an optimized daily scheduling plan with the goal of maximizing the daily treatment volume, and with multi-dimensional constraints such as mold type coordination, equipment load limit, and therapist working status, and to generate early warning information for situations that approach or exceed the load limit. Patient service module: Connects to the resource collaborative scheduling module and is used to push the patient's estimated queue number, treatment progress and real-time status of treatment equipment to the patient terminal in real time.

[0009] Secondly, the present invention provides an intelligent appointment and queuing method, applied to the above-mentioned system, comprising the following steps: Data acquisition steps: Obtain the patient's static characteristic data and historical dynamic treatment data; Intelligent prediction step: Based on the static feature data and the last two dynamic treatment data, dynamically calculate and output the predicted duration of the patient's next treatment; Resource coordination scheduling steps: Based on the estimated duration and model type of all scheduled patients, schedule optimization is performed with the goal of resource coordination and load balancing, generating schedule reminders and warnings; Information synchronization and queuing process: Based on the scheduling plan, queuing information and equipment status are synchronized to the patient's terminal, and a call reminder is given at an appropriate time. The patient's terminal and the waiting area display show the treatment progress and equipment status. The treatment process is divided into stages, such as positioning - image verification - treatment. The duration and progress are displayed accordingly.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention introduces a personalized, adaptive dynamic prediction model for treatment duration, making the scheduling plan closer to the actual treatment process. This significantly reduces queue chaos and cumulative delays caused by inaccurate time estimation, and improves equipment utilization and the reliability of the schedule.

[0011] This invention innovatively uses "mold type" as a key constraint for scheduling. Through a mold similarity priority algorithm, it effectively reduces the number of physical resource switching during the treatment process, reduces the workload of therapists, increases the effective treatment throughput of a single device, and achieves collaborative optimization of multi-dimensional resources.

[0012] The system of this invention achieves real-time transparency of treatment progress and equipment status through the patient service module, greatly improving the patient experience. At the same time, the system's load early warning function provides decision support for managers, enabling them to proactively intervene and prevent risks, thus forming a complete optimized closed loop from intelligent scheduling and status synchronization to early warning management, and constructing a transparent and forward-looking management closed loop. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the module structure of an intelligent appointment and queuing system in one embodiment of the present invention.

[0016] Figure 2 This is a flowchart illustrating an intelligent appointment and queuing method according to an embodiment of the present invention.

[0017] Figure 3 This is a detailed flowchart of the resource collaborative scheduling steps in one embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0020] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" and "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0021] Example 1 like Figure 1 As shown, this embodiment provides an intelligent appointment and queuing system, which mainly includes a data acquisition module, an intelligent prediction module, a resource collaborative scheduling module, and a patient service module. Each module can communicate and exchange data via the hospital's internal network or a dedicated data bus.

[0022] The data acquisition module is the data foundation of this invention's system, responsible for automatically, accurately, and in real-time collecting all key data related to patient radiotherapy appointments and treatment procedures from multiple heterogeneous information systems within the hospital. These heterogeneous information systems include hospital information systems, radiotherapy planning systems, and treatment room terminals, among other data sources. Specifically, it is configured to acquire static characteristic data and dynamic treatment data of patients. Static characteristic data refers to attribute data that is determined before or at the beginning of the patient's treatment cycle and does not change frequently in the short term. This data provides the system with a background framework for patient treatment and initial estimation basis. Dynamic treatment data refers to time-series data generated during each treatment execution, reflecting the actual operation. This data is crucial for the system's adaptive learning and precise correction.

[0023] Furthermore, static characteristic data includes data such as the patient's radiotherapy prescription, treatment plan, type of imaging scan, and self-care ability rating, specifically: Radiation therapy prescription and treatment plan data: This is core medical data jointly developed by a team of radiation oncologists and physicists. It precisely specifies the macroscopic and microscopic parameters of the treatment, including the following data items: Total Dose: The unit is gray (Gy), which refers to the total radiation dose delivered to the target area throughout the entire treatment plan.

[0024] Fractional dose: The unit is gray (Gy), which refers to the dose given in a single treatment.

[0025] Total Fractions: The total number of treatments required throughout the entire treatment course.

[0026] Beam-on Time: A precise and fixed radiation exposure time calculated by the Treatment Planning System (TPS) based on parameters such as plan complexity and machine hop count. This is one of the few components of the treatment duration that can be precisely determined in advance.

[0027] Treatment techniques include intensity-modulated radiotherapy (IMRT), volumetric modulated radiotherapy (VMAT), and stereotactic body radiotherapy (SBRT). Different techniques have different requirements for positioning and verification.

[0028] Target site location: such as head and neck, chest, abdomen, pelvis, etc., directly affects the positioning complexity and scanning range.

[0029] The data collection process for radiotherapy prescriptions and treatment plans is as follows: The data acquisition module interfaces with the radiotherapy planning system and the hospital information system / radiotherapy information system through standard medical data interfaces (such as HL7, DICOM RT) or APIs provided by the hospital; When a doctor completes plan verification and signs a prescription in TPS, or completes appointment registration in HIS / RIS, the system generates a structured data package containing the aforementioned data items.

[0030] The data acquisition module listens to or periodically polls these systems. Once a new plan or plan update is detected, it automatically captures and parses the data packet, storing the key fields in the system's database.

[0031] Image scan type data: This refers to the type of image verification technology used to confirm patient positioning before each treatment, such as CBCT (cone-beam computed tomography), MVCT (megavolt computed tomography), and 2D orthogonal X-rays. Different types of scans have different time requirements. This information is usually embedded in DICOM RT plans or HIS / RIS medical orders as part of the treatment plan or prescription. The acquisition module extracts the corresponding fields by parsing the data source of the above systems.

[0032] Self-care ability rating data: This refers to a standardized assessment level of a patient's mobility and cooperation level. This rating system is derived from the analysis and learning of massive amounts of historical treatment data. Through machine learning models, the system correlates patient movement characteristics (such as whether assistance is needed, use of assistive tools, and movement speed) with the key dynamic data of "patient preparation time," thereby constructing a standard non-technical preparation time range corresponding to different ratings. In clinical application, nurses or doctors assess and select patients based on this data-driven rating standard when admitting them or developing a treatment plan. The system can then automatically match and output a scientifically objective initial estimated preparation time based on historical big data learning. Ratings are typically categorized into data-driven levels such as "fully self-sufficient," "requires minimal assistance," "requires wheelchair," and "bedridden / anesthesia," providing accurate initial input for subsequent personalized dynamic duration estimations. Specific data items (standardized ratings) include: Level I (Completely Self-Reliant): The patient can independently and quickly complete actions such as getting on and off the treatment bed and dressing and undressing.

[0033] Level II (Minor Assistance Required): The patient moves slightly slower or requires verbal guidance or minor assistance from the therapist when a postural fixation device is placed.

[0034] Level III (requires wheelchair): The patient needs to come in a wheelchair and requires assistance from a therapist to get in and out of the treatment bed.

[0035] Category IV (Bedridden / Anesthesia): The patient is completely unable to move, or is a child or other uncooperative patient requiring anesthesia, necessitating multiple medical staff to move the patient or provide anesthesia monitoring.

[0036] Self-care ability rating data is generated by the head nurse or physician during the patient's initial simulation positioning or planning, using a dedicated assessment form within the hospital information system or radiotherapy information system. The data acquisition module synchronizes the assessment results from the HIS / RIS. To ensure data consistency, this system also provides a standardized rating dictionary for use by the source system.

[0037] Furthermore, dynamic treatment data records the start and end times of each key node in the entire treatment process in the form of timestamps. The main steps include: Patient preparation time: This refers to the time from when the patient enters the treatment room until the patient has assumed a preliminary position on the treatment bed and the therapist is ready to begin scanning and verification. This time includes the duration of all non-technical preparatory activities, such as the patient getting lost, dressing and undressing, and taking out and putting in the positioning mold.

[0038] Therapist positioning time: from the start of manually or with laser light-assisted adjustment of the patient's position, to the point where the position reaches the initial position required by the treatment plan, until the therapist returns to the operating room, closes the shielded door, and is ready to start taking verification images.

[0039] Image scan time: refers to the cumulative time for effective image acquisition by image-guided equipment (such as CBCT). Timing begins with the scan initiation command and ends when the system completes image acquisition. If the scan is interrupted due to patient movement, equipment adjustment, or other reasons, timing pauses; it resumes once the scan restarts. The system ultimately records the net scan time, excluding interruptions.

[0040] Image registration time: From the completion of the scan, when the therapist compares the acquired images with the planned images on the workstation and calculates the offset, to the completion of registration and approval of the shift parameters. If adaptive radiotherapy is performed, this stage also includes the process of redrawing the target area, redesigning the physical plan, and re-verifying the images, which significantly extends the time required.

[0041] Accelerator beam exit time: In the scheduling and prediction stages, the system uses the planned beam exit time, pre-calculated and set by the Treatment Planning System (TPS), as a key component in estimating the total duration of a single treatment session, primarily for initial scheduling reservations. During actual treatment, the system collects and records the actual clinical beam exit time from the start of accelerator beam emission to the end of radiation. This actual time, based on feedback from the accelerator control system, is used to verify and analyze deviations from the planned value, and is entered into a dynamic database as part of the actual total treatment duration, providing data support for subsequent intelligent prediction and process optimization.

[0042] Patient departure and room clearing time: This timeframe records the duration from the end of one treatment session to the time the treatment room is cleared to receive the next patient. The calculation is based on the arrival of the next patient to more accurately reflect the room's turnaround efficiency. The system employs a multi-source data fusion strategy for accurate data collection, which primarily includes: The core timing method based on the status of the shielded door (preferred): The system takes the first opening of the shielded door after treatment as the starting point of timing and the closing of the shielded door again after the therapist has positioned the next patient and left the treatment room as the ending point of timing. This method can reliably capture the complete "patient departure - treatment room preparation" cycle by directly monitoring the status signal of the shielded door.

[0043] Assisted Validation and Handling of Complex Scenarios: To address complex clinical situations (such as potential interference from family members entering and exiting the premises), the system employs additional methods for cross-validation and judgment, including: Facial recognition: While protecting privacy, an identification terminal is deployed at the exit of the treatment room to confirm the identity of the previous patient and record the time of their departure, serving as an auxiliary verification point.

[0044] Infrared sensing and logic interpretation: By combining infrared sensor data at the access control point, the algorithm results feedback (such as identifying continuous entry and exit signals in a short period of time) can distinguish the movement of patients, therapists and their families, ensuring that the timing is not interfered with by non-therapist personnel.

[0045] Therapist terminal operation log: The operation log is compared with the operation log of the touch screen terminal in the treatment room (such as records of key nodes such as the end of treatment and the start of positioning) in time sequence to form a logical closed loop.

[0046] The above data acquisition methods work together to accurately capture the time interval from the end of the previous treatment cycle to the readiness of the treatment room, providing crucial data for evaluating room turnaround efficiency and optimizing scheduling. The system algorithm will prioritize the most reliable and interference-resistant data source and perform multi-source verification in case of ambiguity to ensure the quality of dynamic treatment data.

[0047] The acquisition of dynamic data relies on terminal devices and IoT technology within the treatment room to achieve automated or semi-automated recording. For example, a touchscreen all-in-one machine can be deployed in each treatment room as a terminal, connected to the linear accelerator control system and imaging system via a network. Automatic recording can be achieved through automatic triggering. Patient entry / exit: To accurately identify patient entry and exit events, the system employs an intelligent recording scheme combining physical triggering, identity binding, and logical filtering. Physical event triggering: Deploy infrared sensors or magnetic switches at the shielded door of the computer room to detect the opening and closing status of the door in real time and generate the original door status change event sequence.

[0048] Identity binding and event labeling: To accurately associate door status change events with specific patient treatment cycles, the system mandates or strongly recommends performing identity binding operations at key nodes. Therapist terminal confirmation: On the control terminal in the operating room, the therapist clicks "Patient Has Entered" after the patient actually enters the machine room but before preparation begins; and clicks "Patient Has Left" after the patient has completely left the machine room. This operation directly binds the current timestamp to the patient ID, which is the most reliable event annotation method in the system.

[0049] Patient self-service QR code scanning (optional): Set up a QR code scanner in a suitable location in the machine room. After entering, the patient (or assisted by the therapist) scans the unique QR code on their appointment form or wristband, and the system automatically records it as an "entry" event; before leaving, they scan the code again and record it as a "departure" event.

[0050] Intelligent logic filtering of interference: The system has a built-in rule engine that compares and cleans up raw sensor events with manually confirmed events. If a therapist performs a patient entry / exit confirmation operation within a reasonable time window before or after a door status change event, then that manual confirmation time will be adopted as the authoritative record.

[0051] For door status changes that are not associated with any manually confirmed events, the system will intelligently determine whether they are family members passing through, therapists temporarily entering or leaving, or other interference based on the context (such as the current treatment stage and adjacent event sequences), and mark or filter them, and not use them as the basis for calculating the patient's dynamic time.

[0052] This system, while utilizing sensors for automatic event capture, incorporates a manual verification process and intelligent filtering logic to ensure accuracy. This allows for reliable acquisition of critical moments of patient entry and exit even in complex real-world environments (such as frequent family visits), laying a solid foundation for subsequent time calculations.

[0053] Image Scan Start / End: The system actively acquires precise operation commands and status streams from the imaging system (such as OBI) by calling its application programming interface (API) or directly reading its shared memory state. When the imaging system begins executing the scan sequence, this system synchronously captures the "scan start" command and records a timestamp; when the imaging system reports the completion of the scan sequence, this system synchronously captures the "scan end" signal and records a timestamp. This method achieves high-precision synchronization between the image operation status and the system's recording time, ensuring the accuracy and reliability of the dynamic data in this stage.

[0054] Accelerator beam start / end: ​​Automatically recorded by listening to the status signals of the accelerator control system (such as "Beam On", "Beam Off").

[0055] For data collection in key areas such as treatment positioning and image registration, the system adopts a combination of "structured operation process guidance and multi-source intelligent verification," which improves the automation of data collection while fundamentally enhancing clinical safety.

[0056] Structured operating procedure guidance and security checks: The system presents therapists with a visual operating interface (such as a flowchart) on the touchscreen terminal in the treatment room, which is strictly synchronized with the treatment process. Therapists must follow the interface guidance and click the corresponding confirmation button (such as [Start Positioning], [Registration Complete]) at key nodes (such as Positioning Start, Positioning End, Registration Start, Registration Complete).

[0057] Before or simultaneously with clicking each key step button, the system will force or intelligently trigger a "three checks and seven verifications" security check. For example: Before clicking "Start Positioning," the interface will automatically display the current patient information (name, medical record number, treatment area, and photo taken on the day). The therapist must verify the patient and information on-site in the treatment room before confirming and starting the timer.

[0058] This confirmation process not only records the time point, but its operation log itself (operator, time, and verification items) constitutes an electronic security check record, ensuring the traceability of clinical safety procedures.

[0059] Intelligent fusion and cross-validation with automated sensor data: The system deeply integrates and logically verifies the therapist's manual operations with automatically collected sensor data to improve data accuracy and handle complex scenarios. Time logic verification: For example, the time point of "positioning begins" should be later than the sensor record of "patient enters," and the time point of "registration complete" should be earlier than the system signal of "accelerator beam exit begins." The system performs automatic logic verification and issues prompts for abnormal sequences.

[0060] The system uses the following data to calculate and verify the time between patient departure and room clearing: The primary basis for calculation is the time interval between the opening and closing of the shielding door after the therapist clicks "End Treatment," ensuring the signal's stability and reliability.

[0061] Auxiliary verification: Using facial recognition deployed at the exit under the premise of protecting privacy, the time when the previous patient left is confirmed as an auxiliary verification point.

[0062] Intelligent filtering: By analyzing the trigger count and pattern of the infrared sensor of the shielded door, combined with the context of the treatment process, the system intelligently filters out interference signals caused by family members or therapists entering and exiting alone, ensuring that the calculated "clearing time" reflects the true turnover efficiency.

[0063] Through the above process, the system transforms the therapist's necessary clinical operations (safety checks and process advancement) into high-value structured data. This data, along with IoT sensor data, corroborates each other, forming an intelligent data collection closed loop that ensures patient safety, accurately collects the duration of each step, and assesses the efficiency of the data center.

[0064] The treatment room terminal records a sequence of timestamped events and transmits it in real time to the data acquisition module via the hospital's intranet. The acquisition module parses, cleans, and integrates these raw logs to generate a complete, time-based, dynamic treatment record for each patient's treatment, and stores it in the database.

[0065] The intelligent prediction module, connected to the data acquisition module, is used to perform initial predictions based on the static feature data and integrate dynamic treatment data from the patient's previous two treatment sessions. It then dynamically outputs and continuously corrects the predicted duration of each treatment session using a preset algorithm model. The intelligent prediction module transforms static and dynamic data into accurate duration predictions, including: Initial Prediction Model: For patients undergoing treatment for the first time, the system uses a pre-trained regression model to predict the duration of treatment based on their static feature data. The model's training data comes from a large number of historical patients' static features and final average treatment durations. For example, a patient diagnosed with conventional nasopharyngeal carcinoma who requires positional immobilization using a "head, neck, and shoulder thermoplastic membrane combined with expanding foam," if their self-care ability rating is "fully self-sufficient," CBCT imaging verification is used, and the planned beam exit time is 2 minutes, the system's initial predicted treatment duration might be 12 minutes. This predicted duration is significantly longer than the baseline beam exit time because the model has learned from historical patterns regarding the relatively complex positioning and registration under head and neck immobilization in nasopharyngeal carcinoma, and the potential increase in preparation time due to the use of expanding foam.

[0066] Dynamic Correction Mechanism: Starting from the patient's second treatment, the module initiates dynamic correction. Its core logic is to continuously calibrate the estimated treatment duration based on the patient's latest historical performance, dynamically approximating the actual treatment time and providing a more reliable basis for subsequent scheduling. Specifically, the system calculates the average of the actual total time (from entering the treatment room to leaving) of the patient's "first two treatments after exiting the treatment." The system then weights and merges this average with the initial estimated value, or directly uses it as the primary basis for subsequent treatment estimates. Through this rolling correction process, the system can establish and continuously update a personalized treatment duration model for each patient. As the number of treatments increases, the estimated duration will increasingly closely match the patient's actual treatment level. This provides a precise data foundation for dynamically adjusting and scheduling appointments for newly admitted patients, thus ensuring the stability and efficiency of the overall scheduling. It should be noted that "the first two" are used because the first treatment usually involves repositioning and initial imaging verification, excluding treatment after exiting the treatment, and the patient is unfamiliar with the procedure, making the time unrepresentative. Taking the average of the first two treatments better reflects the patient's current routine level.

[0067] Example: Patient A's initial estimated treatment time was 25 minutes. Their second treatment actually took 28 minutes, and their third treatment actually took 26 minutes. Therefore, the system will correct the estimated treatment time for their fourth treatment to (28+26) / 2 = 27 minutes.

[0068] The core task of the intelligent prediction module is to achieve accurate and personalized prediction of the duration of a single treatment session for a patient. Its operation is a dynamically evolving, self-optimizing intelligent loop, mainly divided into two stages: Phase 1: Initial Prediction (applicable to initial treatment and overall patient appointment time and location control). When a patient is receiving treatment for the first time, due to the lack of personal historical data, the module will activate a pre-trained machine learning model. This model is trained based on massive amounts of historical patient data and can analyze the current patient's static characteristics (such as illness, treatment plan, physical condition, etc.) and output an initial estimated duration based on group patterns, providing a scientific basis for the initial scheduling.

[0069] Phase Two: Dynamic Correction (Applicable to Subsequent Treatments). Starting from the patient's second treatment, the module activates its core dynamic correction mechanism. This mechanism follows the principle that "individual history is the best predictor of the future," and the dynamic correction process is as follows: First, the module will focus on the actual time taken for the patient's "first two actual treatments". Selecting "first two" is to exclude abnormal data caused by unfamiliarity with the process during the first treatment, thereby capturing the patient's routine treatment rhythm more quickly and reliably.

[0070] The module then calculates the average of the first two actual treatment times and uses this result as the primary basis for predicting the duration of subsequent treatments. In practice, the system either directly uses this historical average to overwrite the initial estimate or performs a weighted fusion with it, causing the prediction result to quickly shift from "generality of the group" to "specificity of the individual".

[0071] Secondly, continuous rolling optimization: After each new treatment is completed, its actual data is immediately incorporated into the calculation to update the next prediction. This allows the estimated duration to "snowball," continuously approaching the patient's actual condition as treatment progresses, achieving increasingly accurate time control and better ensuring the final outcome of scheduling the appropriate number of patients while maintaining clinical quality.

[0072] The resource collaborative scheduling module, connected to the intelligent prediction module, is used to generate an optimized daily scheduling plan with the goal of maximizing daily treatment volume, and with multi-dimensional constraints including mold type coordination, equipment load limits, therapist work status, patient feelings, and complaints. It also generates early warning information for situations approaching or exceeding the load limit. The system, through the resource collaborative scheduling module, formulates the optimal daily scheduling plan, including: 1. Optimization goal: To more rationally control the number of appointments and arrange patients for treatment within fixed working hours, thereby maximizing the number of patients treated per day.

[0073] 2. Multidimensional constraints, including: Mold Type Collaboration: The system incorporates a "mold type similarity priority" algorithm. When generating schedules, this algorithm attempts to group patients using the same type of positional fixation mold (such as a "head, neck, and shoulder thermoplastic film") together on a timeline. Specifically, during scheduling, the system calculates a "mold switching cost" for each time period. Given a fixed number of patient appointments, it estimates the total treatment time for all patients that day, analyzes reasonable shifts, and assesses the feasibility of the target number of appointments, providing comprehensive recommendations (e.g., "The current total appointment time is nearing shift saturation; it is recommended not to add any more appointments"; or "Based on mold type distribution, concentrating patients using mold A in the morning could increase efficiency by X%, potentially accommodating an additional Y patients"). This effectively reduces the number of times therapists need to change the scaffold baseboard.

[0074] Equipment Load Cap: The system sets a reference limit for the daily treatment duration or number of patients treated for each linear accelerator. This threshold is set based on equipment performance, maintenance cycle, and clinical experience. During scheduling, the system calculates the estimated total load in real time and compares it with this limit. When the scheduling result approaches or exceeds the load threshold, the system issues a warning to administrators and therapists and analyzes potential causes (e.g., excessive patient concentration due to mold coordination at a certain time). The system provides a data-driven risk warning and optimization suggestion (e.g., "The current schedule is close to the equipment's saturation load, which may increase subsequent patient waiting time or equipment operation risks"), rather than a mandatory constraint. Therapists and administrators can combine this warning information, the urgency of newly admitted patients, individual differences in actual treatment, and other comprehensive clinical factors to conduct a final review and flexible adjustment of the scheduling plan to ensure that efficiency is improved while guaranteeing the timeliness and safety of treatment.

[0075] Therapist work status: The system combines patient complaint rates, changes in the treatment rate during therapists' working hours, and the probability of rising adverse events to assess the reasonable workload and shift arrangements for therapists, avoiding overwork.

[0076] 3. Scheduling Output: The module ultimately outputs a detailed daily schedule, including the appointment time slot for each patient, the machine to be used, and the mold type identifier.

[0077] The resource coordination and scheduling module serves as the "command center" of the entire system, responsible for transforming the intelligent output of preceding modules into an efficient, feasible, and robust daily operational plan. Its workflow is not simply task sequencing, but a multi-objective decision-making process seeking the optimal solution under multiple complex constraints. The specific workflow is as follows: 1. Input and Initialization: When the intelligent data collection module starts, it first obtains the static basic information (such as mold type) of all patients to be scheduled from the data acquisition module, and obtains the accurate and personalized estimated treatment duration for each patient from the intelligent prediction module. These intelligently corrected durations are the fundamental guarantee of the reliability of the scheduling plan.

[0078] 2. Optimization and Trade-offs: Seeking the optimal solution under constraints. The module's core objective is to "maximize the number of treatments within a fixed timeframe," employing mathematical optimization models (such as genetic algorithms and integer programming) for scheduling calculations. In this process, it must intelligently balance the following three key multidimensional constraints: Mold Type Collaboration (Efficiency Constraint): The system incorporates a "mold similarity priority" algorithm. During scheduling, it intelligently clusters patients using the same type of mold along the timeline. This effectively reduces the time spent on frequent "mold frame switching" during scheduling, directly improving therapists' efficiency and equipment throughput.

[0079] Equipment load limit (capacity constraint): The system strictly adheres to the maximum daily load capacity of each linear accelerator. When the system attempts to add new patients, causing the total load to approach or exceed the preset threshold, it will automatically trigger an early warning mechanism to alert the administrator of the overload risk, thereby preventing an increase in failure rate and a decrease in treatment quality due to equipment overload.

[0080] Therapist work status (human resource constraints): The system regards therapists as an important resource, taking into account their shifts, reasonable rest time, and workload balance, avoiding tight schedules that may lead to over-fatigue, and ensuring the sustainability of human resources and stable clinical quality.

[0081] 3. Outputs and Delivery: Generate executable intelligent schedules. After optimization and calculation, the module finally outputs a detailed and visualized daily schedule. This schedule not only includes the appointment time slot for each patient down to the minute, but also clearly indicates the corresponding treatment equipment and the type of mold required, providing clear guidance for the treatment team's daily execution.

[0082] The resource collaborative scheduling module, data acquisition module, and intelligent prediction module form an interconnected intelligent closed loop, specifically: The "raw materials" for the resource collaborative scheduling module depend entirely on the data acquisition module. In particular, the critical constraint of mold type directly originates from the static characteristic data obtained by the data acquisition module. Without accurate and complete initial data, collaborative scheduling is impossible.

[0083] The accuracy of the resource-coordinated scheduling module directly depends on the accuracy of the intelligent prediction module's predictions. The dynamic, personalized treatment durations provided by the intelligent prediction module are the absolute foundation for the scheduling module to calculate the daily total workload, assess equipment capacity limits, and perform precise time slot matching. If the duration predictions are inaccurate, no matter how advanced the scheduling algorithm is, the generated plan will collapse during execution. Therefore, the intelligent prediction module is the prerequisite and guarantee for the reliable operation of the resource-coordinated scheduling module.

[0084] The patient service module, connected to the resource coordination and scheduling module, is used to push the patient's estimated queue number, treatment progress, and real-time status of treatment equipment to the patient's terminal in real time. This module serves as the window for interaction between the system and the patient, typically implemented through a mobile app, WeChat mini-program, or in-hospital large screen.

[0085] Estimated queue number: Displays the patient's current position in the queue.

[0086] Treatment progress: Displays the current patient number being treated, and the estimated waiting time.

[0087] Real-time device status: Clearly displays whether the device is "operating normally," "service suspended (fault / maintenance)," or "delayed." This is the key difference between this embodiment and existing general queuing systems, allowing patients, especially those living far away, to avoid "making a wasted trip."

[0088] The patient service module also includes a reminder function: it can assess the distance and time to the hospital based on the patient's location and provide timely reminders (please ask the patient's opinion when requesting location).

[0089] The working process of the patient service module includes: 1. Once the resource collaboration scheduling module generates the final daily schedule, the patient service module will automatically receive this "master schedule." It will immediately initialize the estimated queue number and planned treatment time for each patient in the queue, and push the patient's personal appointment information to the client for the first time via mobile app, mini-program, or other terminals, informing the patient of their basic arrangements for the day.

[0090] 2. On the day of treatment, the patient service module enters real-time operation, and its data is closely linked to the actual situation at the hospital treatment site, including: Data-driven: The patient service module obtains data in real time from the treatment room terminal and equipment monitoring system, including the patient ID currently being treated and the equipment operating status (such as start, end, fault, maintenance).

[0091] Dynamic Updates: Based on this real-time data, the patient service module dynamically performs the following operations: updates queue status, automatically calculates and updates the estimated queue number and expected waiting time for all preceding patients; synchronizes treatment progress, clearly displaying the "under treatment" patient number on public displays and individual patient terminals, making the waiting process visible; broadcasts device status: once a device is detected to be in a "fault," "planned maintenance," or "treatment delay" state, the module immediately pushes a notification to all affected patients. It should be noted that a notification for the current time period indicates a treatment pause due to machine malfunction, while for the next time period, a notification indicates a postponement of the overall treatment time based on the engineer's assessment of the repair duration. The final content sent requires therapist review before distribution. Broadcasting device status effectively prevents patients from making wasted trips and manages their expectations, reducing negative emotions.

[0092] 3. Intelligent reminders and interaction: The patient service module has proactive service capabilities. Based on the scheduling plan, it will automatically send a check-in reminder to the patient when preset trigger conditions (such as 30 minutes before the patient's appointment time) are met, guiding the patient to arrange their arrival time accordingly.

[0093] The intelligent appointment and queuing system described above is a data-driven, intelligent decision-making, and closed-loop optimization integrated solution. Its four core modules—data acquisition, intelligent prediction, resource collaborative scheduling, and patient services—act like the "sensors," "brain," "commander," and "spokesperson" of an organism, working collaboratively to completely transform the traditional operating model of radiotherapy centers. The system's working process is as follows: First, the data acquisition module performs comprehensive data perception and aggregation: the system first automatically obtains medical data through the data acquisition module. The system comprehensively acquires static characteristic data (such as treatment plan, model type, and self-care ability) and dynamic treatment data (such as the actual time taken for each step) from multiple sources, including the hospital information system, radiotherapy planning system, and treatment room terminals. This lays a solid and reliable data foundation for the entire system.

[0094] Then, the intelligent prediction module performs personalized and adaptive intelligent predictions: for new patients, it uses a model trained on historical big data for initial prediction; for returning patients, it activates a dynamic correction mechanism, continuously revising the subsequent predicted duration based on the patient's actual performance in the previous two treatments. This enables the system to provide increasingly accurate personalized duration predictions for each patient, achieving an intelligent leap from "group experience" to "individual patterns."

[0095] Secondly, the resource collaboration scheduling module performs multi-objective resource optimization and scheduling: Acting as the command center, this module receives "patient model type" and "accurate estimated duration" from the first two modules, with the core objective of maximizing daily treatment volume, while intelligently balancing three major constraints: model collaboration, equipment load, and manpower allocation. Through advanced optimization algorithms, it ultimately generates an optimal daily scheduling plan that improves efficiency while ensuring treatment quality, and provides proactive warnings of overload risks.

[0096] Finally, the patient service module enables transparent information exchange and patient services: It translates optimized scheduling plans into an excellent patient experience, providing patients with real-time and transparent updates on their queue number, estimated wait time, treatment progress, and the real-time status of crucial equipment via mobile apps and mini-programs. Simultaneously, the module proactively sends check-in reminders. This allows patients to manage their time effectively, avoiding aimless waiting and wasted trips.

[0097] Through the precise coordination of the four major modules of the system described above, the present invention achieves the following beneficial effects compared to existing queuing systems: This invention significantly improves the treatment throughput and utilization efficiency of linear accelerators through precise prediction and resource coordination, while reducing the workload of therapists and the risk of equipment failure. This enables a shift in hospital operations from "extensive management" to "refined and intelligent scheduling."

[0098] This invention empowers patients with full access to information and a sense of control through information transparency and proactive services, greatly improving patient satisfaction.

[0099] This invention constructs an intelligent closed loop of "perception-decision-execution-feedback," enabling the system to continuously learn from actual treatment. Learn and dynamically optimize subsequent arrangements to form a virtuous cycle that becomes smarter and more efficient the more it is used.

[0100] Example 2 like Figure 2 As shown, this embodiment provides an intelligent appointment and queuing method, applied to the system of Embodiment 1, including the following steps: S101: Data Acquisition Steps: Obtain the patient's static characteristic data and historical dynamic treatment data. The system continuously acquires static patient data from systems such as HIS and TPS, and collects dynamic data after each treatment from the treatment room terminal.

[0101] The specific process of step S101 includes: The system automatically captures patients' static characteristic data, including radiotherapy prescriptions, treatment plans, imaging scan types, and self-care ability ratings, through standard medical data interfaces (such as HL7 and DICOM RT) or hospital APIs.

[0102] Dynamic treatment data is automatically recorded through terminal devices or IoT sensors in the treatment room, including timestamps for key steps such as patient preparation, positioning, scanning, registration, beam delivery, and departure.

[0103] The data acquisition module cleans and integrates the raw data to form structured patient treatment records, providing data support for subsequent modules.

[0104] S102: Intelligent Prediction Step: Based on the static feature data and the last two dynamic treatment data, dynamically calculate and output the estimated duration of the patient's next treatment. When it is necessary to schedule or adjust the time for the patient, the intelligent prediction module is invoked to calculate the estimated treatment duration. The specific implementation process includes: For patients receiving their first treatment, the system outputs an initial estimated duration based on their static feature data using a pre-trained machine learning regression model.

[0105] For patients who are not receiving their first treatment, the system employs a dynamic correction mechanism: extracting the total time of their first two actual treatments, calculating the average value, and using it as the estimated duration of the next treatment, thus achieving rolling optimization that makes the treatment more accurate with each treatment.

[0106] The estimated results are updated to the system database in real time for use by the scheduling module.

[0107] S103: Resource Collaboration Scheduling Steps: Based on the estimated duration and template type of all scheduled patients, scheduling is optimized with the goals of resource collaboration and load balancing, generating a scheduling plan and early warning. For example... Figure 3 As shown, it includes the following steps: S1031: Get a list of all patients awaiting scheduling, along with their estimated duration and template type.

[0108] S1032: The schedule is dynamically optimized with the goal of rationally arranging the maximum number of treatments while ensuring clinical quality.

[0109] S1033: Add “mold type collaboration” as a hard or soft constraint to the model, and prioritize arranging patients with the same mold type to be adjacent.

[0110] S1034: Calculate the total scheduled load and compare it with the equipment upper limit.

[0111] S1035: If overloaded, generate an alert and prompt the administrator to make adjustments; if normal, generate the final scheduling plan.

[0112] The specific implementation process of step S103 includes: The system uses optimization algorithms (such as genetic algorithms and greedy algorithms) to generate the optimal daily scheduling plan under multi-dimensional constraints such as mold coordination, equipment load, and therapist working status.

[0113] Mold type collaboration achieves efficient scheduling of physical resources by calculating "mold switching cost" and minimizing total cost.

[0114] The system monitors equipment load in real time and automatically generates early warning information when it approaches or exceeds a preset threshold, prompting management personnel to intervene.

[0115] S104: Information Synchronization and Queue Calling Steps: Based on the scheduling plan, synchronize queuing information and device status to the patient's terminal, and issue a queue call reminder at an appropriate time. Convert the scheduling plan into patient appointment information and push it through the patient service module. The system automatically triggers queue calling based on real-time treatment progress (e.g., displayed on the waiting area screen and via voice call), and notifies patients in the next appointment slot to prepare. The specific implementation process includes: The patient service module pushes the scheduling plan to the patient's terminal (such as APP, mini program) in real time, displaying information such as queue number, estimated waiting time, and device status.

[0116] The system dynamically updates the queue status based on the real-time progress feedback from the treatment room terminal and triggers a call reminder at the appropriate time.

[0117] If the equipment malfunctions or is delayed, the system will automatically send a notification to the affected patients to prevent them from waiting blindly.

[0118] To further demonstrate the significant advantages and beneficial effects of the queuing method of the present invention compared to existing queuing methods, this example illustrates a real-world application scenario: A radiotherapy center has one linear accelerator, operating for 8 hours on weekdays. There are currently 5 patients waiting to be booked: Patient 1: Mold A, estimated duration 30 minutes.

[0119] Patient 2: Mold B, estimated duration 25 minutes.

[0120] Patient 3: Mold A, estimated duration 28 minutes.

[0121] Patient 4: Mold B, estimated duration 26 minutes.

[0122] Patient 5: Mold A, estimated duration 32 minutes.

[0123] Traditional scheduling (based solely on time or first-come, first-served): might be 1(A)-2(B)-3(A)-4(B)-5(A). This means the mold base needs to be replaced 4 times.

[0124] The scheduling method of this invention: The system uses a resource coordination algorithm to schedule the system into a sequence of 1(A)-3(A)-5(A)-2(B)-4(B). This means the mold base only needs to be replaced once, after treating patient 5 and before treating patient 2, significantly improving efficiency.

[0125] The method described above in this invention forms an intelligent closed loop of "data → prediction → scheduling → service → feedback," with each step interdependent and interconnected to achieve efficient system operation. Specifically, this is manifested in: Improve equipment utilization and treatment throughput: Reduce downtime and waiting time through accurate forecasting and resource coordination, and maximize the number of patients treated per day.

[0126] Reduce non-therapeutic workload: The combination of different mold types significantly reduces the number of times physical resources need to be switched, thus alleviating the workload of therapists.

[0127] Enhancing system transparency and patient experience: Real-time information push and call number reminders allow patients to keep track of their treatment progress, reducing anxiety and blind waiting.

[0128] Build an intelligent early warning and decision support mechanism: The system can proactively identify equipment overload risks and assist managers in intervention and scheduling optimization.

[0129] This will enable an intelligent transformation from "experience-based scheduling" to "data-driven scheduling," creating a virtuous cycle that becomes more accurate and efficient with each run.

[0130] The specific implementation methods of the above steps are the same as the specific implementation methods of the above module functions, and will not be repeated here.

[0131] Example 3 The present invention also provides an electronic device, including: a processor, a transmitting device, an input device, an output device, and a memory. The processor may be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory may be implemented using a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), and is used to store computer program code. The computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any of the above possible implementation methods.

[0132] Example 4 The present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform a method as described in any of the above possible implementations.

[0133] In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0134] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. An intelligent appointment and queuing system, characterized in that, include: The data acquisition module is used to acquire the patient's static characteristic data and dynamic treatment data; the static characteristic data includes the patient's radiotherapy prescription, treatment plan, type of body positioning device, type of imaging scan, and self-care ability rating; the dynamic treatment data includes the actual time taken for each stage of the patient's historical treatment. The intelligent prediction module, connected to the data acquisition module, is used to make an initial prediction based on the static feature data, and integrate the dynamic treatment data of the patient's first two beam exit treatments. It dynamically outputs and continuously corrects the predicted duration of a single treatment for the patient through a preset algorithm model. The resource coordination and scheduling module, connected to the intelligent prediction module, is used to generate an optimized daily scheduling plan with the goal of maximizing the daily treatment volume and with multi-dimensional constraints such as mold type coordination, equipment load limit, and therapist working status. It also generates early warning information for situations that approach or exceed the load limit. The patient service module, connected to the resource collaborative scheduling module, is used to push the patient's estimated queue number, treatment progress, and real-time status of treatment equipment to the patient's terminal in real time.

2. The system according to claim 1, characterized in that, The data acquisition module interfaces with the hospital information system, radiotherapy planning system, and treatment room terminal via HL7 or DICOM RT interface to achieve automatic acquisition and integration of static and dynamic data.

3. The system according to claim 1, characterized in that, The intelligent prediction module uses a machine learning model trained on historical data to make an initial prediction during the patient's first treatment, and then makes dynamic rolling corrections based on the actual treatment time of the previous two treatments from the second treatment onwards.

4. The system according to claim 1, characterized in that, The resource coordination scheduling module employs an optimization algorithm to minimize mold switching costs during the scheduling process, and uses mold type coordination as one of the key constraints for scheduling.

5. The system according to claim 1, characterized in that, The patient service module pushes queuing information, equipment status, and check-in reminders to patients via mobile APP, WeChat mini-program, or in-hospital display screen.

6. An intelligent appointment and queuing method, applied to the system as described in any one of claims 1 to 5, characterized in that, Includes the following steps: Data acquisition steps: Obtain the patient's static characteristic data and historical dynamic treatment data; Intelligent prediction step: Based on the static feature data and the last two dynamic treatment data, dynamically calculate and output the predicted duration of the patient's next treatment; Resource-coordinated scheduling steps: Based on the estimated duration and template type of all scheduled patients, the scheduling is optimized with the goal of resource coordination and load balancing, generating a scheduling plan and early warning; Information synchronization and queuing process: According to the scheduling plan, synchronize queuing information and device status to the patient's terminal, and remind the patient to call their number at the appropriate time.

7. The method according to claim 6, characterized in that, In the intelligent prediction step, for patients who are not receiving their first treatment, the average of the previous two actual treatment durations is used as the predicted duration of the next treatment.

8. The method according to claim 6, characterized in that, The resource collaborative scheduling steps include: Obtain the estimated duration and mold type for all patients awaiting scheduling; The scheduling model is initialized with the goal of maximizing the number of patients treated. Using mold type coordination as a constraint, patients with the same type of mold are given priority in being placed next to each other; Calculate the total load and compare it with the equipment limit to generate an early warning or final scheduling plan.

9. An electronic device comprising a processor, a memory, and a computer program stored in the memory, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 6 to 8.

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 as described in any one of claims 6 to 8.

Images