Dynamic monitoring device for pituitary tumor hormone level

Abstract

The invention relates to the field of pituitary tumor monitoring, and discloses a pituitary tumor hormone level dynamic monitoring device, which comprises a multi-modal data acquisition and integration module, a detection analysis module, a data processing and analysis module, a clinical decision support module and a remote interaction and management module, the multi-modal data acquisition and integration module is electrically connected with the laboratory data interface, the multi-modal data acquisition and integration module is in communication connection with the minimally invasive implantable sensor and the dynamic monitoring sensor, and the detection and analysis module comprises a hypersensitive detection module and a hormone dynamic analysis module. And the data processing and analysis module comprises an edge calculation module and a cloud big data analysis module. The hormone level is accurately monitored through multi-modal acquisition and super-sensitive detection, data is efficiently processed through edge and cloud computing, decision is optimized through the clinical decision module, space-time limitation is broken through the remote interaction module, and quality improvement and efficiency improvement of medical services are assisted.

Description

Technical Field

[0001] The present invention relates to the field of pituitary tumor monitoring, and in particular to a device for dynamically monitoring hormone levels in pituitary tumors. Background Art

[0002] Pituitary tumors are common intracranial tumors that originate from residual cells of the anterior and posterior pituitary glands and the craniopharyngeal epithelium. As a crucial endocrine organ in the human body, the pituitary gland secretes a variety of hormones that play a key role in regulating physiological processes such as growth, development, metabolism, and reproduction. The presence of pituitary tumors can severely disrupt the normal function of the pituitary gland, leading to hormonal imbalances and, in turn, a series of complex and severe clinical symptoms.

[0003] During the treatment of pituitary tumors, accurate monitoring of the dynamic changes in hormone levels is crucial for evaluating treatment efficacy, adjusting treatment plans, and determining prognosis. However, traditional hormone level monitoring methods have many limitations.

[0004] Traditional laboratory testing relies primarily on periodic blood sampling for hormone testing. This method is subject to significant time lag, as test results take time to arrive and fail to reflect real-time changes in a patient's hormone levels. Furthermore, blood samples only reflect hormone levels at a specific moment in time, making it difficult to capture subtle fluctuations and rhythmic changes in hormone secretion. For example, pituitary hormone secretion is pulsatile, and hormone levels can vary significantly at different time points. Traditional single-sample blood tests struggle to accurately reflect these dynamic changes.

[0005] Furthermore, traditional testing methods lack comprehensive monitoring of the hormone secretion environment. Changes in hormone levels in the tissue fluid surrounding pituitary tumors may more directly reflect the tumor's biological behavior and response to treatment, but traditional testing methods are unable to capture this information. Furthermore, the patient's physiological state, such as sleep, exercise, and mood, can affect hormone secretion, and traditional testing methods struggle to comprehensively account for the dynamic impact of these factors on hormone levels.

[0006] Furthermore, traditional monitoring methods lack effective data integration and analysis capabilities. Different types of hormone data and patient clinical information are often scattered across disparate systems and records, making unified management and analysis difficult. Doctors spend considerable time and effort organizing and interpreting this data, and it can be difficult to extract valuable insights from this vast amount of data, hindering the scientific nature and accuracy of clinical decision-making. Summary of the Invention

[0007] In response to the shortcomings of the existing technology, the present invention provides a device for dynamic monitoring of pituitary tumor hormone levels, which solves the problems of traditional pituitary tumor hormone monitoring methods such as time lag, inability to fully capture hormone fluctuations and secretion environment information, weak data integration and analysis capabilities, and lack of personalization in clinical decision-making, thereby facilitating precise treatment.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a pituitary tumor hormone level dynamic monitoring device, including a multimodal data acquisition and integration module, a detection and analysis module, a data processing and analysis module, a clinical decision support module and a remote interaction and management module, the multimodal data acquisition and integration module establishes an electrical connection with the laboratory data interface, the multimodal data acquisition and integration module establishes a communication connection with the minimally invasive implantable sensor and the dynamic monitoring sensor, the detection and analysis module includes an ultra-sensitive detection module and a hormone dynamic analysis module, the data processing and analysis module includes an edge computing module and a cloud-based big data analysis module, the clinical decision support module includes a treatment response prediction unit, an early warning threshold adjustment unit and a guideline dynamic adaptation unit, and the remote interaction and management module includes a doctor-side management module and a three-dimensional visualization interaction module.

[0009] Preferably, the laboratory data interface automatically connects to the hospital LIS system to obtain hormone test results in real time. The minimally invasive implantable sensor is used to sense and collect hormone information in the surrounding tissue fluid or blood in real time. The operation method is to implant the microsensor manufactured by nanotechnology into the area near the pituitary tumor and the blood circulation system in the patient's body through a minimally invasive method. The dynamic monitoring sensor is used to continuously monitor the patient's physiological indicators and is worn on the patient using a wearable device.

[0010] Preferably, the ultra-sensitive detection module includes a quantum dot immunofluorescence detection unit and a micro-nano chip mass spectrometry unit. The quantum dot immunofluorescence detection unit uses quantum dot labeling technology, combined with the immunofluorescence detection principle, to perform hormone quantitative analysis on the collected samples. The micro-nano chip mass spectrometry unit combines micro-nano chip technology with mass spectrometry analysis to complete sample preprocessing and separation on the micro-nano chip, and then performs accurate qualitative and quantitative analysis through a mass spectrometer.

[0011] Preferably, the hormone dynamic analysis module includes a time series modeling unit and a hormone correlation map creation unit. The time series modeling unit constructs a hormone secretion rhythm model based on the ARIMA or LSTM algorithm to identify abnormal fluctuation patterns. The hormone correlation map creation unit analyzes the interaction relationship between multiple hormones through a graph neural network.

[0012] Preferably, the edge computing module includes a data receiving unit and a data classification storage unit, the data receiving unit is used to receive various types of data transmitted, the data classification storage unit includes a distributed relational database, distributed non-relational data and a PI database, the distributed relational database is used to store structured data, the distributed non-relational data is used to store unstructured data, and the PI database is used to store time series data. The cloud-based big data analysis module includes a data screening unit, a data conversion unit and a mapping fusion unit, the data screening unit is used to screen various types of patient data and remove duplicate and useless data, the data conversion unit is used to convert data in different formats into a unified format, and the mapping fusion unit is used to fuse the screened and converted data.

[0013] Preferably, the treatment response prediction unit predicts the effects of drug and surgical intervention based on historical data, the warning threshold adjustment unit is used to dynamically adjust the abnormal value alarm range according to the patient's baseline level, and the guideline dynamic adaptation unit is used to embed the NCCN / ESE clinical guideline knowledge base to provide classification treatment recommendations.

[0014] Preferably, the doctor-side management module includes a remote monitoring unit, a treatment plan adjustment unit and a consultation unit. The remote monitoring unit is used for doctors to remotely monitor changes in hormone levels of patients. The treatment plan adjustment unit is used to make real-time adjustments to treatment plans based on changes in hormone levels. The consultation unit supports video calls to facilitate online consultations between doctors from different departments.

[0015] Preferably, the three-dimensional visualization interaction module includes a hormone heat map generation unit, a virtual pituitary model generation unit and a dynamic trend deduction generation unit. The hormone heat map generation unit is used to display the correlation between hormone concentration changes and pituitary tumor volume in the time and space dimensions. The virtual pituitary model generation unit reconstructs the relationship between the lesion and the surrounding structure in 3D and marks the hormone secretion hotspot area. The dynamic trend deduction generation unit simulates the long-term prognosis of different treatment options through digital twin technology.

[0016] The present invention provides a device for dynamically monitoring hormone levels in pituitary tumors. It has the following beneficial effects:

[0017] 1. The present invention uses a multimodal data acquisition method, combined with an ultra-sensitive detection module and a hormone dynamic analysis module, to comprehensively and accurately obtain and analyze hormone level information, promptly discover abnormal fluctuations in hormone levels and the interaction between multiple hormones, and provide a reliable basis for subsequent treatment.

[0018] 2. The edge computing module and the cloud-based big data analysis module of the present invention work together to achieve efficient reception, classified storage, screening, conversion and integration of massive data, thereby improving data processing efficiency and analysis accuracy.

[0019] 3. The clinical decision support module of the present invention is based on historical data and clinical guideline knowledge base, which can predict treatment effects, dynamically adjust warning thresholds and provide classification treatment recommendations, helping doctors make more scientific and reasonable clinical decisions.

[0020] 4. The remote interaction and management module of the present invention allows doctors to remotely monitor patient conditions, adjust treatment plans in real time, and conduct online consultations, breaking down time and space constraints and improving the efficiency and quality of medical services. Furthermore, the 3D visualization interaction module provides doctors with intuitive and comprehensive information display, helping them gain a deeper understanding of the patient's condition and formulate treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a system architecture diagram of the present invention;

[0022] Figure 2 This is a system structure diagram of the multimodal data acquisition and integration module in the present invention;

[0023] Figure 3 This is a system structure diagram of the detection and analysis module in the present invention;

[0024] Figure 4 This is a system structure diagram of the data processing and analysis module in the present invention;

[0025] Figure 5 This is a system structure diagram of the clinical decision support module in the present invention;

[0026] Figure 6 This is a system structure diagram of the remote interaction and management module in the present invention. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0028] Example:

[0029] Please see the attached Figure 1 -Attached Figure 6 , an embodiment of the present invention provides a device for dynamically monitoring hormone levels of pituitary tumors, such as Figure 1As shown, it includes a multimodal data acquisition and integration module, a detection and analysis module, a data processing and analysis module, a clinical decision support module and a remote interaction and management module. The device forms a complete and efficient pituitary tumor hormone level monitoring system through the collaborative work between the modules, aiming to provide comprehensive and accurate information support for the diagnosis and treatment of pituitary tumors. The multimodal data acquisition and integration module establishes an electrical connection with the laboratory data interface. This electrical connection method ensures that the data can be transmitted stably and quickly, providing a reliable basis for subsequent analysis and processing. The multimodal data acquisition and integration module establishes a communication connection with the minimally invasive implantable sensor and the dynamic monitoring sensor. With the help of advanced communication technology, it realizes the real-time and accurate collection of data from different types of sensors, thereby obtaining multi-dimensional information about the hormone levels of pituitary tumors. The detection and analysis module The module includes an ultra-sensitive detection module and a hormone dynamic analysis module. These two sub-modules cooperate with each other to conduct in-depth analysis of the collected data to explore the potential patterns behind the changes in hormone levels. The data processing and analysis module includes an edge computing module and a cloud-based big data analysis module. They respectively undertake data processing tasks at different levels, and through division of labor and cooperation, the efficiency and accuracy of data processing are improved. The clinical decision support module includes a treatment response prediction unit, an early warning threshold adjustment unit, and a guideline dynamic adaptation unit. These units provide clinicians with scientific and reasonable decision-making basis, which helps to formulate more personalized treatment plans. The remote interaction and management module includes a doctor-side management module and a three-dimensional visualization interaction module, which provides doctors with convenient remote management means and intuitive information display methods, making it convenient for doctors to understand the patient's condition at any time.

[0030] In this embodiment, the laboratory data interface automatically connects to the hospital's LIS system to obtain hormone test results in real time. The hospital's LIS system is the core system for hospital laboratory information management and stores a large amount of patient test data. By automatically connecting to the system, the latest hormone test results can be obtained in a timely manner, ensuring the timeliness and accuracy of the data. The minimally invasive implantable sensor is used to sense and collect hormone information in the surrounding tissue fluid or blood in real time. The operation method is to implant the microsensor manufactured by nanotechnology into the area near the pituitary tumor and the blood circulation system of the patient's body through a minimally invasive method. The microsensor manufactured by nanotechnology has the characteristics of small size and high sensitivity. It can accurately sense hormone information in the surrounding tissue fluid or blood without causing major trauma to the patient, providing the possibility of real-time monitoring of hormone levels. The dynamic monitoring sensor is used to continuously monitor the patient's physiological indicators. It is worn on the patient using a wearable device. The wearable device has the advantages of being easy to carry and real-time monitoring. The patient can wear it normally in daily life to continuously record changes in physiological indicators and provide doctors with more comprehensive disease information.

[0031] Furthermore, the ultra-sensitive detection module includes a quantum dot immunofluorescence detection unit and a micro-nano chip mass spectrometry unit. The quantum dot immunofluorescence detection unit uses quantum dot labeling technology, combined with the immunofluorescence detection principle, to perform quantitative hormone analysis on the collected samples. Quantum dot labeling technology has the characteristics of high sensitivity and high stability, and can accurately label hormone molecules in the sample. Combined with the immunofluorescence detection principle, it can achieve quantitative analysis of hormones by detecting the intensity of the fluorescence signal, providing accurate data support for clinical diagnosis. The micro-nano chip mass spectrometry unit combines micro-nano chip technology with mass spectrometry analysis, completes sample preprocessing and separation on the micro-nano chip, and then performs accurate qualitative and quantitative analysis through the mass spectrometer. Micro-nano chip technology can realize rapid preprocessing and separation of samples on a tiny chip, thereby improving analysis efficiency. The mass spectrometer has the characteristics of high resolution and high accuracy, and can perform accurate qualitative and quantitative analysis on the separated samples, providing a more reliable means for detecting hormone levels.

[0032] Furthermore, the hormone dynamic analysis module includes a time series modeling unit and a hormone correlation map creation unit. The time series modeling unit constructs a hormone secretion rhythm model based on the ARIMA or LSTM algorithm to identify abnormal fluctuation patterns. ARIMA (autoregressive integrated moving average model) and LSTM (long short-term memory network) algorithms are two commonly used time series analysis algorithms. They can predict future hormone secretion based on historical data and identify abnormal fluctuation patterns. By modeling and analyzing the hormone secretion rhythm, doctors can better understand the changes in the patient's condition and adjust the treatment plan in time. The hormone correlation map creation unit analyzes the interaction relationship between multiple hormones through graph neural networks. Graph neural networks are a type of neural network specially used to process graph structured data. They can effectively analyze the complex interaction relationship between multiple hormones. By creating hormone correlation maps, doctors can have a more comprehensive understanding of the synergistic effects and regulatory mechanisms between hormones, providing a basis for formulating personalized treatment plans.

[0033] Specifically, the hormone dynamic analysis process includes a pulse secretion model and a state-space model. The pulse secretion model is used for hormone rhythm analysis.

[0034]

[0035] Among them, H0 is the basic secretion rate, λ is the hormone metabolism rate, A k , t k is the amplitude and time of the kth pulse, and β is the pulse attenuation coefficient. Its main application is to quantify the diurnal pulse secretion pattern of GH (growth hormone).

[0036] State space model for multi-hormone interaction analysis, reference:

[0037] Equation of state:

[0038] α t+1 =Tα t +Rη t (η t ~N(0,Q))

[0039] Observation equation:

[0040] y t =Zα t +ε t (ε t ~N(0,R)

[0041] Among them, α t is the hidden state, T is the state transfer matrix, and Z is the observation matrix. Its main application is to analyze the feedback regulation relationship between ACTH and cortisol.

[0042] Furthermore, the edge computing module includes a data receiving unit and a data classification storage unit. The data receiving unit is used to receive various types of data transmitted. In this process, the data receiving unit needs to have the characteristics of high bandwidth and low latency to ensure that it can quickly and accurately receive data from various sensors and data sources. The data classification storage unit includes a distributed relational database, a distributed non-relational data and a PI database. The distributed relational database is used to store structured data, the distributed non-relational data is used to store unstructured data, and the PI database is used to store time series data. Different types of databases are suitable for storing different types of data. Through reasonable classification storage, the data management efficiency and query speed can be improved. Cloud big data classification The analysis module includes a data screening unit, a data conversion unit and a mapping fusion unit. The data screening unit is used to screen various types of patient data and remove duplicate and useless data. In massive data, there may be a large amount of duplicate and useless information. The data screening unit can remove these interference information and improve the quality of data by setting reasonable screening rules. The data conversion unit is used to convert data in different formats into a unified format. Since the data formats of different data sources may be different, the data conversion unit can convert these data in a unified format for subsequent analysis and processing. The mapping fusion unit is used to fuse the screened and converted data. By fusing data from different sources, more valuable information can be mined to provide more comprehensive support for clinical decision-making.

[0043] Furthermore, the treatment response prediction unit predicts the effects of drug and surgical interventions based on historical data. Historical data contains rich information. The treatment response prediction unit establishes a prediction model by analyzing and mining a large amount of historical data, and can accurately predict the effects of drug and surgical interventions. This helps doctors evaluate the feasibility and effectiveness of treatment plans before treatment and choose the treatment method that best suits the patient. The early warning threshold adjustment unit is used to dynamically adjust the abnormal value alarm range according to the patient's baseline level. Each patient has a different baseline level, and the normal fluctuation range of hormone levels is also different. The early warning threshold adjustment unit can dynamically adjust the abnormal value alarm range according to the patient's individual situation to avoid false alarms or missed alarms and improve the accuracy of the early warning. The guideline dynamic adaptation unit is used to embed the NCCN / ESE clinical guideline knowledge base and provide classification treatment recommendations. The NCCN (National Comprehensive Cancer Network) and ESE (European Society of Endocrinology) clinical guidelines are internationally authoritative clinical guidelines. The guideline dynamic adaptation unit embeds these guideline knowledge bases into the system to provide personalized classification treatment recommendations based on the patient's specific situation, providing an important reference basis for clinicians.

[0044] Specifically, the warning threshold adjustment unit uses dynamic threshold calculation and reinforcement learning reward function, refer to:

[0045]

[0046] Among them, ω1, ω2, and ω3 are weight coefficients determined through medical cost analysis, and FP / TP / FN are the number of false positives, true positives, and false negatives. Their main application is to adjust the alarm threshold for abnormally elevated ACTH.

[0047] Accordingly, the treatment effect is evaluated using the causal forest individual treatment effect, refer to:

[0048]

[0049] in, is the prediction result of the b-th tree for individual i receiving treatment, It is a predictor of outcome without treatment, and its main application is to predict the individualized effect of surgery on TSH (thyroid-stimulating hormone) levels.

[0050] Furthermore, the doctor-side management module includes a remote monitoring unit, a treatment plan adjustment unit and a consultation unit. The remote monitoring unit is used by doctors to remotely monitor changes in patients' hormone levels. With the help of modern communication technology, doctors can understand the changes in patients' hormone levels in real time through the remote monitoring unit at any time and any place, and promptly discover abnormal fluctuations in the condition. The treatment plan adjustment unit is used to make real-time treatment plan adjustments based on changes in hormone levels. When the patient's hormone level changes, the treatment plan adjustment unit can provide doctors with suggestions for adjusting the treatment plan based on preset rules and algorithms, helping doctors to adjust treatment strategies in a timely manner and improve treatment effects. The consultation unit supports video calls to facilitate online consultations between doctors from different departments. When faced with complex pituitary tumor cases, doctors from multiple departments may need to participate in diagnosis and treatment. The consultation unit breaks the limitations of time and space by supporting video calls, facilitating online communication and discussion between doctors from different departments to jointly formulate the best treatment plan.

[0051] Furthermore, the three-dimensional visualization interaction module includes a hormone heat map generation unit, a virtual pituitary model generation unit and a dynamic trend deduction generation unit. The hormone heat map generation unit is used to display the correlation between hormone concentration changes and pituitary tumor volume in the time and space dimensions. The hormone heat map can display the changes in hormone concentration in different time and space in the form of intuitive color distribution, as well as the correlation with pituitary tumor volume. By observing the hormone heat map, doctors can more clearly understand the relationship between hormone level changes and pituitary tumor development, providing an important reference for the formulation of treatment plans. The virtual pituitary model generation unit reconstructs the relationship between the lesion and the surrounding structure in 3D, and marks the hormone secretion hotspot area. The 3D reconstruction technology can Pituitary tumors and surrounding structures are presented in the form of three-dimensional models. Doctors can observe the morphology and location of lesions from different angles. At the same time, marking hormone secretion hotspots can help doctors understand the source and distribution of hormone secretion more accurately, and provide more precise guidance for surgical treatment. The dynamic trend deduction generation unit simulates the long-term prognosis of different treatment options through digital twin technology. Digital twin technology is a technology that combines physical entities with virtual models. The dynamic trend deduction generation unit simulates the development trend of pituitary tumors and the long-term prognosis of patients under different treatment options by establishing a digital twin model of the patient. Doctors can choose the treatment plan that best suits the patient based on the simulation results, thereby improving the success rate of treatment.

[0052] Specifically, the hormone heat map generation unit uses spatiotemporal graph convolution to generate hormone heat maps, reference:

[0053]

[0054] in, is the spatiotemporal adjacency matrix, is the degree matrix, are trainable weights, and their main application is to model the spatiotemporal propagation pattern of GH secretion.

[0055] Working principle: First, the multimodal data acquisition and integration module automatically connects to the hospital LIS system through the laboratory data interface to obtain hormone test results in real time; the minimally invasive implantable sensor is manufactured using nanotechnology and is minimally invasively implanted in the patient's body near the pituitary tumor area and blood circulation system to sense and collect hormone information in the surrounding tissue fluid or blood in real time; the dynamic monitoring sensor is worn on the patient using a wearable device to continuously monitor the patient's physiological indicators. The collected data is aggregated to the multimodal data acquisition and integration module, and then the ultra-sensitive detection module in the detection and analysis module uses quantum dot immunofluorescence detection unit and micro-nano chip mass spectrometry The combined unit processes the collected samples, and the quantum dot immunofluorescence detection unit uses quantum dot labeling technology combined with immunofluorescence detection principles to quantitatively analyze hormones; the micro-nano chip mass spectrometry unit combines micro-nano chip technology with mass spectrometry analysis to complete sample preprocessing, separation, and accurate qualitative and quantitative analysis. The time series modeling unit of the hormone dynamic analysis module builds a hormone secretion rhythm model based on ARIMA or LSTM algorithm to identify abnormal fluctuation patterns; the hormone correlation map creation unit analyzes the interaction relationship between multiple hormones through graph neural networks, and then the data receiving unit of the edge computing module receives various types of transmitted data, and the data is classified and stored. The storage unit stores structured data in a distributed relational database, unstructured data in a distributed non-relational database, and time series data in a PI database. The data screening unit of the cloud-based big data analysis module screens patient data and removes duplicate and useless data. The data conversion unit converts data of different formats into a unified format. The mapping and fusion unit integrates the screened and converted data. The treatment response prediction unit of the clinical decision support module predicts the effects of drug and surgical interventions based on historical data. The warning threshold adjustment unit dynamically adjusts the abnormal value alarm range based on the patient's baseline level. The guideline dynamic adaptation unit embeds the NCCN / ESE clinical guideline knowledge base to provide categorized treatment recommendations. Finally, the remote monitoring unit of the physician management module allows physicians to remotely monitor changes in patient hormone levels. The treatment plan adjustment unit adjusts treatment plans in real time based on changes in hormone levels. The consultation unit supports video calls, facilitating online consultations between physicians from different departments. The 3D visualization interaction module uses the hormone heat map generation unit to display the correlation between hormone concentration changes and pituitary tumor volume in the spatiotemporal dimension. The virtual pituitary model generation unit reconstructs the relationship between the lesion and surrounding structures in 3D and labels hormone secretion hotspots. The dynamic trend deduction generation unit uses digital twin technology to simulate the long-term prognosis of different treatment plans.

[0056] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A device for dynamic monitoring of pituitary tumor hormone levels, comprising a multimodal data acquisition and integration module, a detection and analysis module, a data processing and analysis module, a clinical decision support module, and a remote interaction and management module, characterized in that: The multimodal data acquisition and integration module establishes an electrical connection with the laboratory data interface, and the multimodal data acquisition and integration module establishes a communication connection with the minimally invasive implantable sensor and the dynamic monitoring sensor. The detection and analysis module includes an ultra-sensitive detection module and a hormone dynamic analysis module. The data processing and analysis module includes an edge computing module and a cloud-based big data analysis module. The clinical decision support module includes a treatment response prediction unit, an early warning threshold adjustment unit and a guideline dynamic adaptation unit. The remote interaction and management module includes a doctor-side management module and a three-dimensional visualization interaction module.

2. A device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The laboratory data interface automatically connects to the hospital's LIS system to obtain hormone test results in real time. The minimally invasive implantable sensor is used to sense and collect hormone information in the surrounding tissue fluid or blood in real time. The operation method is to implant the microsensor manufactured by nanotechnology into the area near the pituitary tumor and the blood circulation system in the patient's body through a minimally invasive method. The dynamic monitoring sensor is used to continuously monitor the patient's physiological indicators and is worn on the patient using a wearable device.

3. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The ultra-sensitive detection module includes a quantum dot immunofluorescence detection unit and a micro-nano chip mass spectrometry unit. The quantum dot immunofluorescence detection unit uses quantum dot labeling technology, combined with the principle of immunofluorescence detection, to perform hormone quantitative analysis on the collected samples. The micro-nano chip mass spectrometry unit combines micro-nano chip technology with mass spectrometry analysis to complete sample pretreatment and separation on the micro-nano chip, and then performs accurate qualitative and quantitative analysis through a mass spectrometer.

4. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The hormone dynamic analysis module includes a time series modeling unit and a hormone correlation map creation unit. The time series modeling unit constructs a hormone secretion rhythm model based on the ARIMA or LSTM algorithm to identify abnormal fluctuation patterns. The hormone correlation map creation unit analyzes the interaction relationship between multiple hormones through a graph neural network.

5. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The edge computing module includes a data receiving unit and a data classification storage unit. The data receiving unit is used to receive various types of data transmitted. The data classification storage unit includes a distributed relational database, distributed non-relational data and a PI database. The distributed relational database is used to store structured data, the distributed non-relational data is used to store unstructured data, and the PI database is used to store time series data. The cloud-based big data analysis module includes a data screening unit, a data conversion unit and a mapping fusion unit. The data screening unit is used to screen various types of patient data and remove duplicate and useless data. The data conversion unit is used to convert data in different formats into a unified format. The mapping fusion unit is used to fuse the screened and converted data.

6. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The treatment response prediction unit predicts the effects of drug and surgical intervention based on historical data, the early warning threshold adjustment unit is used to dynamically adjust the abnormal value alarm range according to the patient's baseline level, and the guideline dynamic adaptation unit is used to embed the NCCN / ESE clinical guideline knowledge base to provide classification treatment recommendations.

7. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The doctor-side management module includes a remote monitoring unit, a treatment plan adjustment unit and a consultation unit. The remote monitoring unit is used for doctors to remotely monitor changes in patients' hormone levels. The treatment plan adjustment unit is used to make real-time adjustments to treatment plans based on changes in hormone levels. The consultation unit supports video calls to facilitate online consultations between doctors from different departments.

8. The device for dynamic monitoring of pituitary tumor hormone levels according to claim 1, characterized in that: The three-dimensional visualization interaction module includes a hormone heat map generation unit, a virtual pituitary model generation unit and a dynamic trend deduction generation unit. The hormone heat map generation unit is used to display the correlation between hormone concentration changes and pituitary tumor volume in the time and space dimensions. The virtual pituitary model generation unit reconstructs the relationship between the lesion and the surrounding structure in 3D and marks the hormone secretion hotspot area. The dynamic trend deduction generation unit simulates the long-term prognosis of different treatment options through digital twin technology.

Images