Multi-mode eyeball protrusion measuring equipment

Abstract

The invention relates to multi-mode eyeball protrusion measuring equipment, which belongs to the technical field of medical instruments and comprises a host, a power supply module is arranged at the bottom of the host, a signal processing and data storage module is arranged in the middle of the host, and the host is connected with a measuring probe assembly through a special interface. The measuring probe assembly is connected with the host through a bendable mechanical arm, the front end is provided with an ultrasonic-optical integrated measuring probe, and the rear end is provided with a circuit board. And the control unit is connected with the host through a data line and comprises a touch operation panel and a display screen. The bendable mechanical arm is composed of metal joints, and a micro motor and an angle sensor are arranged in the bendable mechanical arm. The ultrasonic-optical integrated measuring probe is integrated with an ultrasonic transducer, a projector and a camera. Through multi-technology fusion and precise structure design, high-precision eyeball protrusion measurement is achieved, the device effectively adapts to complex eye conditions, operation is easy and convenient, data management is convenient and fast, and powerful support is provided for clinical diagnosis of the ophthalmology department.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a multimodal eyeball exophthalmos degree measuring device. Background Art

[0002] In ophthalmological clinical practice, accurate measurement of exophthalmos plays an indispensable role in the diagnosis, assessment, and treatment planning of a variety of diseases. For example, changes in exophthalmos in patients with thyroid-related eye diseases are important indicators for assessing disease progression and treatment effectiveness. Measuring exophthalmos in patients with orbital tumors helps determine the tumor's location, size, and degree of pressure on surrounding tissues. However, existing exophthalmos measurement equipment has numerous limitations, making it difficult to meet the demands of complex clinical environments.

[0003] Traditional contact measurement devices, such as the Hertel exophthalmosometer, determine eye proptosis by measuring the distance from the lateral orbital rim to the corneal vertex. This method not only requires the patient to maintain a specific posture, increasing discomfort, but also involves direct contact with the skin around the eye during measurement, which is susceptible to factors such as skin condition and operator technique, leading to significant measurement errors. When the patient's eye is inflamed, swollen, or wearing an eye occlusion, determining the contact point becomes difficult, making measurement accuracy even more challenging.

[0004] Currently, the non-contact measurement devices commonly used in clinical practice primarily include ultrasonic measurement devices, CT scanning technology, and MRI imaging technology. Ultrasonic measurement devices acquire depth information of the eye by emitting and receiving ultrasonic waves. However, their resolution is relatively low and requires high operator skill, so they may not always achieve the desired high accuracy. CT scanning technology can provide high-resolution three-dimensional images, but in complex ocular conditions, such as those associated with severe ocular structural deformities or lesions that result in poor tissue contrast, the interpretation and accurate measurement of CT and MRI images may be affected, making judgment difficult. Furthermore, they are expensive and require patients to receive a certain radiation dose. Furthermore, CT scanning equipment is bulky, making it unsuitable for use in primary care settings or emergency situations. MRI imaging technology offers high soft tissue resolution, but the imaging time is long and the equipment cost is high, making it unsuitable for primary care settings or emergency situations.

[0005] Furthermore, most existing measurement devices are single-function, relying on a single measurement principle and lacking the flexibility to address complex clinical situations. Faced with diverse patient conditions, it is impossible to switch measurement methods to obtain accurate data based on actual conditions. Furthermore, these devices are often not compact enough in design and complex to operate, requiring specialized personnel to operate them. This hinders rapid and efficient measurement in primary healthcare settings or in emergency situations. Therefore, developing a novel multimodal exophthalmos measurement device to overcome the shortcomings of existing technologies has important clinical significance and practical application value. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to propose a multimodal exophthalmos degree measurement device, which is provided with an ultrasonic transducer made of high-precision piezoelectric ceramics and capable of emitting and receiving ultrasonic waves to obtain the approximate position information of the eyeball, an optical measurement component composed of a micro-structured light projector, a high-resolution CMOS camera and related signal processing circuits and capable of accurately obtaining the surface contour information of the eyeball, an integrated signal transmission and processing circuit for fusing and processing ultrasonic and optical measurement data, a mechanical structure adjustment component that can flexibly adjust the position and angle of the measuring probe, a data fusion and processing algorithm that can deeply analyze data and accurately calculate the exophthalmos degree, and an integrated device structure including a host, a measuring probe assembly and a control unit, so as to solve the problems in the prior art of exophthalmos degree measurement equipment that cannot adapt to complex clinical environments, has low measurement accuracy, a complex structure and inconvenient operation, different measurement principle modules are independent and lack coordination, and the data processing capability is weak and it is difficult to provide accurate results.

[0007] The present invention is achieved through the following technical solutions:

[0008] A multimodal eyeball exophthalmos measurement device includes a host, a measuring probe assembly and a control unit. The host is connected to the measuring probe assembly via a flexible mechanical arm. An ultrasonic-optical integrated measuring probe is fixedly provided at the front end of the measuring probe assembly. A circuit board for preliminary processing of measurement data and control of the measuring probe assembly is provided at the rear end of the measuring probe assembly. The control unit is connected to the host via a data cable.

[0009] Furthermore, a power module is fixedly installed at the bottom of the host, a multi-layer printed circuit board integrating a signal processing module and a data storage module is fixedly installed in the middle of the host, a power switch and a status indicator light are arranged horizontally side by side on the front of the host, and a USB interface, a power charging interface and a dedicated interface for connecting to the measurement probe assembly are arranged on the back of the host.

[0010] Furthermore, the flexible robotic arm is composed of multiple metal joints hinged end to end in sequence, and a micro motor and an angle sensor are provided in the metal joint. The output shaft of the micro motor is connected to the rotating part of the metal joint, and the angle sensor is connected to the host through an internal circuit. The metal joints are connected by high-precision bearings.

[0011] Furthermore, the control unit includes a touch operation panel and a high-resolution liquid crystal display screen. The touch operation panel is located above the front of the control unit, and the high-resolution liquid crystal display screen is fixedly arranged below the front of the control unit.

[0012] Furthermore, the ultrasonic-optical integrated measurement probe includes an ultrasonic transducer, a projector and a camera. The ultrasonic transducer is fixed to the front end of the measurement probe assembly through a front mounting seat. Three micro-optical component mounting slots distributed at 120° are evenly arranged around the ultrasonic transducer. The micro-optical component mounting slots are used to accurately place the projector and camera.

[0013] Furthermore, the ultrasonic transducer is tightly connected to the positioning groove of the front end mounting seat in an interference fit manner.

[0014] Furthermore, the projector adopts a micro structured light projector, which is adapted to the micro optical component mounting slot through a metal bracket. One end of the metal bracket is fixed to the micro structured light projector by screws, and the other end is fixed to the optical component mounting frame by screws.

[0015] Furthermore, the camera uses a CMOS image sensor, and the CMOS image sensor is installed in the micro-optical component installation groove through an elastic rubber pad.

[0016] Further, 9. The method for using the device is as follows: Step S1: The device starts and performs self-test, supplies power through the power module at the bottom of the host, turns on the power switch on the front of the host, the status indicator light turns green and lights up, and the host performs self-test on key components such as the power module and the multi-layer printed circuit board;

[0017] Step S2: Mode selection, selecting a normal or special measurement mode on the touch operation panel of the control unit;

[0018] Step S3: Probe positioning: the control unit sends instructions, the host controls the rotation of the micro-motors in the metal joints of the flexible robotic arm, and the angle sensor feeds back angle data in real time until the integrated ultrasound-optical measurement probe is aligned with the patient's eyeball;

[0019] Step S4: Ultrasonic measurement, including the following sub-steps:

[0020] Step S41: signal transmission, the host sends a command to the circuit board through the dedicated interface to drive the ultrasonic transducer to transmit ultrasonic waves;

[0021] Step S42: echo reception, the ultrasonic transducer receives the echo reflected from the surface of the eyeball, converts it into an electrical signal and transmits it to the circuit board;

[0022] Step S43: Signal preprocessing, the circuit board performs bandpass filtering and wavelet threshold denoising on the electrical signal;

[0023] Step S44: feature extraction, extracting the echo envelope through Hilbert transform and locating the corneal reflection peak;

[0024] Step S45: Distance calculation: Calculate the distance from the corneal vertex Ec to the probe according to the formula X3=v×t / 2, where v is the speed of sound and t is the echo time;

[0025] Step S5: Optical measurement, including the following sub-steps:

[0026] Step S51: pattern projection, the host controls the micro structured light projector to project a structured light pattern onto the surface of the eyeball;

[0027] Step S52: image acquisition, the CMOS image sensor captures the deformation pattern;

[0028] Step S53: Pre-processing, the circuit board performs non-uniformity correction, contrast enhancement, and distortion correction on the image;

[0029] Step S54: Feature extraction, the host calculates the three-dimensional contour data of the eyeball surface through edge detection and feature matching algorithms;

[0030] Step S6: Data fusion, the host transforms the ultrasonically measured position data and the optically measured contour data into a unified coordinate system, performs fusion processing using a deep learning model, and outputs a protrusion value Δ;

[0031] Step S7: result display and storage, the host transmits the Δ value to the high-resolution LCD screen of the control unit, and stores the measurement data in a structured manner in the data storage module;

[0032] Step S8: Mode switching. If eyelid swelling or occlusion of the artificial eye is detected, the system automatically switches to the occlusion mode. If orbital deformity is detected, the system switches to the deformity mode.

[0033] The beneficial effects of the present invention are:

[0034] The present invention facilitates all-round and precise measurement of the eyeball through the coordinated work of a flexible robotic arm and an integrated ultrasonic-optical measurement probe, overcomes the interference of complex eye conditions, improves the accuracy and reliability of measurement, and provides more accurate data support for clinical diagnosis.

[0035] The coordinated combination of multiple sensors such as angle sensors, ultrasonic transducers, projectors and cameras facilitates the comprehensive collection of multi-dimensional data. The host integrates and analyzes these data to accurately measure the degree of eye protrusion, assess the condition of the eyes, and then customize personalized diagnostic plans for patients, significantly enhancing the targetedness and effectiveness of the diagnosis.

[0036] The host's intelligent regulation of the parameters of each measurement component and the real-time feedback function of the control unit facilitate dynamic optimization of measurement parameters based on the real-time measurement conditions. The touch operation panel and display screen promptly display measurement data and operation prompts, allowing operators to complete measurement work efficiently and improve measurement quality and efficiency.

[0037] The data storage and transmission module's data storage management, export and upload functions, combined with the connection structure of external storage devices and remote servers, facilitates medical staff to remotely obtain patient measurement data and eye conditions, provide professional diagnosis, and ensure the continuity and comprehensiveness of diagnosis, so that patients can receive appropriate medical support in different scenarios. At the same time, the device can receive data such as updated algorithms to maintain the optimal state of performance and data accuracy, thereby improving the overall level of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is the overall structure diagram of the assembly;

[0039] Figure 2 This is the front view of the overall structure;

[0040] Figure 3 It is the side view of the overall structure;

[0041] Figure 4 This is the rear view of the whole assembly;

[0042] Figure 5 This is the internal structure diagram of the host;

[0043] Figure 6 This is a structural diagram of the bendable robotic arm;

[0044] Figure 7 This is an enlarged view of the ultrasonic-optical integrated measurement probe.

[0045] Description of reference numerals:

[0046] 1. Main unit; 2. Measuring probe assembly; 3. Control unit; 11. Power module; 12. Multilayer printed circuit board; 13. Power switch; 14. Status indicator light; 15. USB port; 16. Power charging port; 17. Special port; 21. Flexible robotic arm; 211. Metal joint; 212. Micro motor; 213. Angle sensor; 214. High-precision bearing; 22. Ultrasonic-optical integrated measuring probe; 221. Ultrasonic transducer; 222. Front-end mounting base; 223. Projector; 224. Metal bracket; 225. Camera; 226. Elastic rubber pad; 227. Optical component mounting bracket; 23. Circuit board; 31. Touch operation panel; 32. High-resolution LCD display; 33. Data cable. DETAILED DESCRIPTION

[0047] like Figure 1-7 As shown, an embodiment of the present invention provides: a multimodal eyeball exophthalmos degree measuring device, including a host 1, a measuring probe assembly 2 and a control unit 3.

[0048] Host 1 is the data processing and control core of the entire device and plays a vital role in the entire system. A power module 11 is fixed to its bottom, which uses a high-performance rechargeable lithium battery pack. This lithium battery pack has high energy density and can store large amounts of electricity in a limited space, providing long-lasting power support for the device. It also has long battery life and can meet the needs of continuous operation for a long time. Power module 11 is connected to the internal power supply circuit via power cables. These power cables are made of low-resistance, high-conductivity materials, which can effectively reduce losses during power transmission and ensure stable power transmission to all components of the device.

[0049] A multilayer printed circuit board (PCB) 12, integrating the signal processing module and data storage module, is fixed to the center of the main unit. PCB 12 utilizes specialized wiring methods, such as microstrip and stripline, to enhance signal transmission speed and stability. It is also equipped with high-speed data transmission interfaces, enabling rapid processing and transmission of large amounts of data. This PCB is responsible for in-depth processing and long-term storage of measurement data, applying complex algorithms for data analysis, filtering, and noise reduction, providing accurate and reliable data support for subsequent diagnosis and research.

[0050] A power switch 13 and a status indicator 14 are positioned horizontally side by side on the front of the device. The power switch 13 is used to turn the device on and off. Its tactile and user-friendly design ensures easy control of the device's power. The status indicator 14 displays the device's operating status in real time, displaying solid green during power on, flashing yellow during standby mode, solid blue during measurement, and flashing white during data transmission. This allows the operator to quickly understand the device's status and make appropriate adjustments.

[0051] The back of the main unit features a USB port 15, a power charging port 16, and a dedicated port 17 for connecting to the measurement probe assembly. The USB port 15 facilitates data export and interaction with external devices. It supports high-speed data transfer protocols, enabling rapid transfer of stored data to external storage devices or computers. The power charging port 16 is used to charge the power module 11 and supports fast charging, fully recharging the battery in a short period of time. The dedicated port 17 connects to the measurement probe assembly 2 via a dedicated data cable for high-speed, stable data transmission and control command exchange. The shielded dedicated data cable effectively prevents external electromagnetic interference, ensuring accurate and stable data transmission.

[0052] The measurement probe assembly 2 is connected to the host computer 1 via a flexible robotic arm 21. The flexible robotic arm 21 consists of multiple metal joints 211 articulated end-to-end, forming a flexible motion structure. Each metal joint 211 houses a micromotor 212 and an angle sensor 213. The micromotor 212 is a high-precision, low-noise stepper motor, with its output shaft connected to the rotating component of the metal joint 211. The stepper motor offers precise control accuracy and can rotate according to a preset number of steps, thereby precisely controlling the joint's rotation angle and providing stable and reliable power for the movement of the robotic arm 21. The angle sensor 213 is connected to the host computer 1 via internal wiring, providing real-time feedback on the joint's rotation angle. This internal wiring utilizes shielded cables with strong anti-interference capabilities, ensuring accurate transmission of angle information to the host computer 1. The host computer 1 precisely adjusts the motion of the robotic arm 21 based on this feedback information. Using a control algorithm (the robotic arm's operation algorithm is a conventional technique and is publicly available), the control algorithm achieves precise multi-degree-of-freedom motion of the robotic arm 21 in space.

[0053] The metal joints 211 are connected by high-precision bearings 214, which feature low friction and high rotational accuracy. This low friction reduces energy loss during joint rotation, improving the efficiency of the robotic arm 21. The high rotational accuracy ensures flexible and stable rotation of the robotic arm 21, minimizing motion errors and enabling the measurement probe to accurately align with the target.

[0054] The front end of the measurement probe assembly 2 is fixedly mounted with an integrated ultrasonic-optical measurement probe 22. This integrated ultrasonic-optical measurement probe 22 includes an ultrasonic transducer 221, a projector 223, and a camera 225. The ultrasonic transducer 221 is fixed to the front end of the measurement probe assembly 2 via a front mounting base 222. The ultrasonic transducer 221 engages tightly with the positioning groove of the front mounting base 222 through an interference fit. This interference fit ensures the stable position of the ultrasonic transducer 221 during operation, reducing displacement caused by vibration or external forces, and ensuring the accuracy of ultrasonic measurements. Three micro-optical component mounting slots are evenly spaced 120° around the ultrasonic transducer 221. These slots precisely position the projector 223 and camera 225. This layout ensures optimal optical measurement angles and range.

[0055] Projector 223 utilizes a micro-structured light projector, which is adapted to the micro-optical assembly mounting slot via a metal bracket 224. Metal bracket 224 is constructed from a high-strength, lightweight aluminum alloy, ensuring structural stability while reducing overall weight for ease of operation. One end of metal bracket 224 is secured to micro-structured light projector 223 via screws, while the other end is secured to optical assembly mounting bracket 227 via screws. Fine-tuning screws allow the angle of projector 223 to be adjusted to meet diverse measurement requirements. This adjustable design allows projector 223 to precisely project a specific structured light pattern based on the patient's eye position and measurement requirements, providing accurate data for subsequent optical measurements.

[0056] Camera 225 uses a CMOS image sensor and is mounted within the micro-optical assembly mounting slot via an elastic rubber pad 226. Made of medical-grade silicone, the rubber pad 226 offers excellent elasticity and shock absorption. This pad not only ensures stable mounting of camera 225 but also effectively reduces the impact of external vibration on image acquisition, preventing image blur or distortion caused by vibration. Once mounted via the rubber pad 226, camera 225 maintains a stable position to capture images of the deformed structured light pattern projected onto the eye's surface, providing high-quality data for subsequent image processing and analysis.

[0057] The back end of the measurement probe assembly 2 is equipped with a circuit board 23 for preliminary processing of measurement data and control of the measurement probe assembly 2. Circuit board 23 integrates a variety of electronic components and chips, and performs functions such as signal amplification, filtering, and digital processing. It performs preliminary processing of the ultrasonic signals received by the ultrasonic transducer 221 and the image signals captured by the camera 225, removing noise and interference and improving signal quality and reliability. Furthermore, circuit board 23 precisely controls the operating state of the measurement probe assembly 2 based on control commands sent by the host computer 1, such as controlling the transmission frequency of the ultrasonic transducer 221, the projection mode of the projector 223, and the image acquisition parameters of the camera 225.

[0058] Signal preprocessing technical details

[0059] The board's preprocessing of the raw signal includes the following steps:

[0060] Ultrasonic signal processing chain: Bandpass filtering: An 8th-order Chebyshev type I filter (passband 5-15MHz) is used to suppress low-frequency motion noise and high-frequency switching interference.

[0061] Time domain denoising: wavelet threshold denoising is applied (Daubechies 6 wavelet basis, threshold $0.2\sigma$, $\sigma$ is the noise variance).

[0062] Peak detection: Extract the echo envelope through Hilbert transform and locate the corneal reflection peak.

[0063] Optical signal processing chain:

[0064] Non-uniformity correction: Compensates for CMOS pixel response differences based on pre-stored black / white calibration maps.

[0065] Dynamic range extension: The CLAHE algorithm (grid size 32×32, contrast limiter 2.0) is applied to enhance the structured light contrast.

[0066] Distortion correction: The Brown-Conrady model is used to correct lens distortion (radial coefficients $k_1=-0.21, k_2=0.03$).

[0067] Control unit 3 includes a touchscreen operation panel 31 and a high-resolution LCD display 32, connected to host computer 1 via a data cable 33. Touchscreen operation panel 31, located on the upper front of control unit 3, features various function buttons, such as power on / off, measurement mode selection, parameter setting, and data query. These buttons utilize sensitive touch-sensing technology, allowing operators to enter commands with a simple touch. The ergonomic design of touchscreen operation panel 31 ensures convenient and efficient operation, improving operator efficiency.

[0068] A high-resolution LCD display 32 is fixed to the lower front of the control unit 3, featuring high clarity, high contrast, and a wide viewing angle. It clearly and in real time displays measurement data, device status, and operational prompts. For example, during a measurement, the display 32 displays information such as the exophthalmos measurement results, the intensity and waveform of the ultrasonic signal, and the results of optical image processing. It also displays the device's operating status, such as measurement progress and battery charge. As the operator performs an operation, the display 32 provides prompts to guide them through the process.

[0069] Data cable 33 utilizes a high-speed, stable data transmission cable with excellent anti-interference performance. It enables two-way data exchange and device control between the control unit 3 and the host computer 1. Commands entered by the operator via the touch panel 31 are quickly and accurately transmitted to the host computer 1 via data cable 33. After processing the commands, the host computer 1 transmits the processing results and related data via data cable 33 to the high-resolution LCD screen 32 of the control unit 3 for display.

[0070] When implementing this embodiment, first ensure that the device is started normally, select the appropriate measurement mode according to the patient's specific situation, and accurately align the measurement probe assembly with the patient's eyeball to prepare for subsequent measurement work.

[0071] The operator presses the power switch 13 on the front of the host 1, turning the device on. The status indicator 14 indicates the power-on status (steady green). The host 1 then performs a self-test, using internal testing procedures to check key components such as the power module 11 and the multilayer printed circuit board 12 (which integrates the signal processing module and data storage module) to ensure proper functioning of all parts of the device. If any abnormality is detected during the self-test, the status indicator 14 will display the corresponding fault code (e.g., flashing red), and the high-resolution LCD 32 of the control unit 3 will display detailed fault information, allowing the operator to quickly troubleshoot and resolve the problem.

[0072] After the self-test is complete, the operator selects the appropriate measurement mode on the touchscreen operation panel 31 of the control unit 3. The measurement mode can be set based on the patient's eye condition and measurement requirements. For example, the regular measurement mode is suitable for most patients with normal eye conditions; a special measurement mode can be selected for patients with eye obstructions or swelling. Different measurement modes correspond to different measurement parameters and algorithms to ensure the accuracy of the measurement results.

[0073] Special measurement mode configuration table

[0074] Mode Name Applicable clinical scenarios Key parameter adjustment Algorithm Switching Occlusion Mode Eyelid swelling / prosthetic eye occlusion Ultrasonic power increased to +30% Optical sampling rate adjusted to 15fps Enable penetrating ultrasound positioning algorithm Deformation Mode Severe deformation of the orbital structure Turn off autofocus and manually set the depth of field to ±5mm Activate 3D point cloud non-rigid registration algorithm Dynamic Mode Pathological nystagmus Optical exposure time <1ms Ultrasonic repetition frequency 1kHz Using motion compensation tracking algorithm

[0075] The operator then sends commands via the touchscreen operation panel 31 to control the operation of the flexible robotic arm 21. Upon receiving the commands, the host computer 1 sends control signals to the micromotors 212 within each metal joint 211 of the flexible robotic arm 21. The micromotors 212 rotate the metal joints 211 in response to the signals. The angle sensors 213 monitor the rotation angles of the joints in real time and feed this information back to the host computer 1 via internal circuitry. The host computer 1 analyzes and processes this feedback, comparing it with the preset target angle. If any deviation is detected, the host computer 1 adjusts the control signals to the micromotors 212 until the measurement probe assembly 2 is properly positioned and the integrated ultrasonic-optical measurement probe 22 is accurately aligned with the patient's eye.

[0076] Robotic arm closed-loop control implementation process:

[0077] Initialization: Load the patient's pupillary distance data and calculate the target pose $T_{target}$.

[0078] Kinematics solution: Call the inverse kinematics algorithm to solve $\theta_{target} = IK(T_{target})$.

[0079] Real-time control loop:

[0080] (python)

[0081] while norm(θ_target-θ_actual) > 0.1deg: # convergence threshold

[0082] e = θ_target - θ_actual

[0083] u = K_p * e + K_d * diff(e) # Generate control quantity

[0084] send_motor_command(u)

[0085] update(θ_actual) # Update angle feedback

[0086] Safety monitoring: If there is no convergence within 10 seconds or the collision sensor is triggered, the movement will be stopped and an alarm will be issued.

[0087] During the adjustment process, the operator can observe the position and angle information of the measuring probe assembly 2 in real time through the high-resolution liquid crystal display 32 of the control unit 3 to ensure the accuracy of the adjustment.

[0088] During the ultrasonic measurement phase, host computer 1 sends ultrasonic measurement instructions to circuit board 23 at the rear end of measurement probe assembly 2 via dedicated interface 17 and a data cable. Upon receiving the instructions, circuit board 23 controls the ultrasonic transmission and reception circuits, driving ultrasonic transducer 221 to emit high-frequency ultrasonic waves. Ultrasonic transducer 221 is made of high-frequency (5-10 MHz), high-resolution piezoelectric ceramic material, efficiently converting electrical signals into ultrasonic signals for transmission. Ultrasonic waves propagate toward the eye in pulses. During propagation, some of the sound waves are reflected by the eye's surface, and the reflected echo is received by ultrasonic transducer 221.

[0089] The ultrasonic transducer 221 converts the received ultrasonic reflection signal into an electrical signal, which is then transmitted via metal wires to the ultrasonic transmitting and receiving circuits. The metal wires are made of high-temperature and corrosion-resistant materials to ensure stable signal transmission. The ultrasonic transmitting and receiving circuits perform preliminary processing on the raw electrical signal. First, they amplify the weak signal to a manageable level. Then, they filter the signal to remove noise and interference, improving signal quality. Finally, they digitize the analog signal to convert it into a digital signal. The processed ultrasonic measurement data is transmitted to the circuit board 23, which performs simple analysis and organization before transmitting it to the host computer 1 via a data cable.

[0090] The multilayer printed circuit board 12, which integrates the signal processing module and data storage module in the host computer 1, further processes and analyzes the ultrasonic measurement data. Using information such as the propagation time and intensity of the ultrasonic echo, combined with a preset algorithm, it calculates the approximate position of the eyeball. This information is stored in the data storage module and displayed on the high-resolution LCD 32 of the control unit 3 for the operator's reference.

[0091] During the optical measurement phase, after the ultrasonic measurement completes preliminary positioning, host 1 controls projector 223. Projector 223 is a micro-structured light projector with a high-precision convex lens. Through rigorous calculation and calibration, it can project specific structured light patterns, such as Gray code or sinusoidal fringe patterns. Host 1 sends control signals to projector 223, adjusting projection parameters such as pattern type, brightness, and angle to ensure that the structured light pattern is accurately projected onto the surface of the eye.

[0092] Due to the varying shape and position of the eye's surface, the structured light pattern projected onto it will be distorted. Camera 225 utilizes a CMOS image sensor, offering high resolution and rapid image acquisition capabilities. Camera 225 captures images of the eye with the distorted structured light pattern from a specific angle. Its lens is a convex lens with a focal length that can be electrically adjusted within a range of 8-12mm based on measurement requirements to ensure clear and accurate images.

[0093] Camera 225 transmits the captured image signals via a flexible flat cable to circuit board 23 at the rear end of measurement probe assembly 2. The flexible flat cable offers excellent flexibility and signal transmission performance, ensuring fast and stable transmission of image signals. Circuit board 23 performs pre-processing on the image signals, including denoising to remove noise interference and improve image clarity; contrast enhancement to enhance image detail by adjusting brightness and color distribution; and image correction to ensure image accuracy and authenticity by performing geometric and distortion correction.

[0094] The preprocessed image signal is transmitted to host 1. The signal processing module in host 1 performs in-depth analysis and processing on the image signal, extracting the contour information of the eyeball surface using image processing algorithms. For example, an edge detection algorithm identifies edge features of the eyeball surface in the image, and a feature matching algorithm compares the deformed structured light pattern with the original pattern to calculate the three-dimensional shape and contour information of the eyeball surface. The processed eyeball surface contour information is stored in the data storage module and correlated with the approximate eyeball position information obtained from previous ultrasound measurements, providing comprehensive data support for subsequent data fusion and calculation. This allows a projector to project a structured light pattern onto the eyeball surface, a camera to capture the deformed pattern, and image signal processing to extract the eyeball surface contour information. This information, combined with the position information obtained from ultrasound measurements, provides more detailed data for accurate calculation of eyeball exophthalmos.

[0095] Data fusion and calculation stage

[0096] Objective: To accurately calculate the degree of eyeball proptosis by fusing the approximate eyeball position information obtained by ultrasound measurement and the eyeball surface contour information obtained by optical measurement.

[0097] Specific operation: After receiving the two sets of data from ultrasonic and optical measurements, the multilayer printed circuit board 12, which integrates the signal processing module and data storage module of the host computer 1, first performs data alignment. Because the coordinate systems of ultrasonic and optical measurements may differ, coordinate conversion and matching algorithms are used to unify the two sets of data into the same coordinate system to ensure data consistency and accuracy.

[0098] Next, a deep learning-based neural network model was used for data fusion. This model utilizes a combined convolutional neural network (CNN) and recurrent neural network (RNN) architecture. The CNN extracts image features from the optical measurement data. Through multiple layers of convolution and pooling, it gradually extracts local and global image features, such as the curvature and texture of the eye's surface. The RNN processes the time series information of the ultrasound measurement data and, combined with the image features extracted by the CNN, explores the complex relationships between the different measurement data. The model is trained using a large amount of training data to learn the mapping between normal and abnormal eye exophthalmos and ultrasound and optical measurement data.

[0099] Neural network implementation details:

[0100] Model Architecture:

[0101] Input layer: ultrasonic depth map (64×64) + optical point cloud (256×256)

[0102] Feature extraction: dual-branch CNN (ultrasound branch: 3-layer Conv1D; optical branch: ResNet-18)

[0103] Fusion layer: Feature concatenation and input into LSTM (128 units) to capture temporal associations

[0104] Output layer: Fully connected layer regresses protrusion value (unit: mm)

[0105] Training data:

[0106] Dataset: 500 clinical data (including normal / pathological samples), divided into training / validation / test sets at 8:1:1

[0107] Enhancement strategy: random translation (±3mm), rotation (±5°), adding Gaussian noise (SNR=30dB)

[0108] Verification results:

[0109] Test set MAE = 0.18mm (better than Hertel measuring instrument’s 0.5mm)

[0110] Bland-Altman analysis showed that the 95% limits of agreement were [-0.23 mm, 0.27 mm]

[0111] Deployment optimization: The model is quantized to INT8 precision, and the inference latency is less than 50ms (accelerated using TensorRT).

[0112] During the calculation process, an optimization algorithm is used to iteratively refine the results, further improving measurement accuracy. A combination of particle swarm optimization (PSO) and genetic algorithms (GA) is employed. The PSO algorithm is used to quickly search for the approximate range of the global optimal solution. The GA algorithm, building on the PSO algorithm's search, further optimizes the solution through genetic operations such as selection, crossover, and mutation, improving both accuracy and quality. After multiple iterations of optimization, accurate exophthalmos measurement results are ultimately obtained.

[0113] Result display and storage stage

[0114] Purpose: To display the calculated exophthalmos results to the operator in a timely manner to facilitate diagnosis and analysis, and to store the measurement data and results to provide data support for subsequent case tracking and research.

[0115] Specific operation: Host 1 transmits the calculated exophthalmos data via a data cable to control unit 3's high-resolution LCD screen 32 for real-time display. Display screen 32 displays the exophthalmos measurement results in intuitive numerical, graphical, or graphical form. It also displays relevant measurement parameters and analytical information, such as the measurement error range and comparison with the normal range, enabling the operator to quickly and accurately understand the measurement results.

[0116] Simultaneously, the data storage module of host 1 organizes and stores all relevant measurement data, including ultrasonic and optical measurement data, processed image information, exophthalmos calculation results, measurement time, and patient information. The data storage module uses high-capacity storage media, such as high-speed solid-state drives, to ensure storage of large amounts of measurement data. This structured data storage facilitates subsequent query, statistics, and analysis.

[0117] In addition, the host 1 can also export the measurement data to an external storage device through the USB interface 15, or upload the data to a remote server through the network interface to achieve data sharing and backup, which facilitates remote diagnosis and case management for medical staff.

[0118] The integrated ultrasonic-optical measurement probe 22 combines ultrasonic and optical measurement technologies, along with data fusion and processing algorithms, to achieve high-precision measurement of eyeball exophthalmos. The ultrasonic transducer 221 can penetrate potential obstructions and swollen tissue to obtain approximate eye position information. Its high frequency (5-10 MHz) and high resolution ensure the accuracy of this positional information. The optical measurement system, consisting of a projector 223 and camera 225, projects and captures structured light patterns to precisely capture the contours of the eyeball's surface.

[0119] In terms of data processing, a deep learning-based neural network model and optimization algorithm (a combination of particle swarm optimization (PSO) and genetic algorithm (GA)) can deeply explore the complex relationships between ultrasonic and optical measurement data, iteratively optimize the measurement results, and effectively reduce measurement errors. Compared with traditional exophthalmos measurement devices, this invention can reduce measurement errors by over 50%, providing more accurate data support for clinical diagnosis, helping doctors make more accurate diagnosis and treatment decisions.

[0120] The flexible robotic arm 21 and the multi-angle adjustable integrated ultrasound-optical measurement probe 22 enable the device to adapt to the eye position and condition of different patients. In actual clinical applications, patients' eyes may present various complex conditions, such as obstructions (such as eyelashes, swollen eyelids, etc.) or inflammation and swelling. Ultrasound measurement technology can penetrate these obstructions and swollen tissue to obtain the approximate position of the eyeball, providing preliminary positioning for optical measurement. Optical measurement, in turn, can further accurately obtain the contour of the eyeball's surface based on ultrasound positioning.

[0121] Through data fusion and processing algorithms, the device can accurately distinguish between swollen tissue and protrusion changes of the eyeball itself. Even in complex clinical environments, it can provide doctors with reliable eyeball exophthalmos measurement data, thereby providing effective diagnostic basis for patients with different conditions.

[0122] The control unit 3's touchscreen operation panel 31 and high-resolution LCD display 32 greatly simplify the device's operation. Operators can easily start the device, select measurement modes, and adjust parameters simply by touching the function buttons on the operation panel 31. The intuitive and user-friendly interface, with clear icons and text prompts, allows even those without a technical background to operate the device proficiently after simple training.

[0123] During the measurement process, display screen 32 displays real-time measurement data, device status, and operational prompts, guiding the operator through each step. The flexible operation of flexible robotic arm 21 also enables the measurement probe assembly 2 to quickly and accurately align with the patient's eye, improving measurement efficiency and reducing the operator's workload.

[0124] The data storage module of host computer 1 has a large storage capacity, capable of storing large amounts of patient measurement data and medical records. This data is stored in a structured manner, making it easy for medical staff to query, compile statistics, and analyze it. Medical staff can enter query criteria such as patient name, measurement time, and condition type through the operation panel 31 of control unit 3 to quickly locate and retrieve the required measurement data and medical records.

[0125] The device also supports data export and upload. Through the USB port 15, medical staff can export measurement data to external storage devices, such as USB flash drives and portable hard drives, for offline analysis and backup. Furthermore, the device can upload data to a remote server via a network interface, enabling remote data sharing and management. This allows doctors to access patient measurement data and medical records anytime, anywhere, facilitating patient condition tracking and diagnosis, and improving the efficiency and quality of medical services.

[0126] This device was designed and manufactured using high-quality materials and precision manufacturing processes, ensuring its stability and reliability. The aluminum alloy housing of the main unit 1 offers excellent heat dissipation and protection, effectively shielding the internal electronic components from environmental influences. The metal joints 211 and high-precision bearings 214 in the flexible robotic arm 21 are constructed from high-strength, wear-resistant metal materials, ensuring the arm will not loosen or wear out over extended periods of use, and guaranteeing the positional accuracy and stability of the measurement probe assembly 2.

[0127] The integrated signal transmission and processing circuitry utilizes multi-layer printed circuit board technology and a shielded design, effectively reducing signal interference and ensuring accurate data transmission and processing. Furthermore, the device's software system has undergone rigorous testing and optimization, ensuring excellent stability and compatibility, adapting to diverse working environments and operational requirements. This ensures the device will operate without failure for extended periods of time, providing reliable support for clinical measurements.

[0128] In summary, the multimodal eyeball exophthalmos measurement device of the present invention has high-precision measurement, strong adaptability, simple operation, convenient data management and stability.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multimodal eyeball exophthalmos measurement device, characterized by: The system comprises a host, a measuring probe assembly and a control unit. The host is connected to the measuring probe assembly via a flexible mechanical arm. An ultrasonic-optical integrated measuring probe is fixedly provided at the front end of the measuring probe assembly. A circuit board for preliminary processing of measurement data and control of the measuring probe assembly is provided at the rear end of the measuring probe assembly. The control unit is connected to the host via a data cable.

2. A multimodal eyeball exophthalmos measurement device according to claim 1, characterized in that: A power module is fixedly installed at the bottom of the host, a multi-layer printed circuit board integrating a signal processing module and a data storage module is fixedly installed in the middle of the host, a power switch and a status indicator light are arranged horizontally side by side on the front of the host, and a USB interface, a power charging interface and a dedicated interface for connecting to the measurement probe assembly are arranged on the back of the host.

3. A multimodal exophthalmos measurement device according to claim 2, characterized in that: The flexible robotic arm is composed of multiple metal joints hinged end to end in sequence. A micro motor and an angle sensor are installed in the metal joints. The output shaft of the micro motor is connected to the rotating part of the metal joint. The angle sensor is connected to the host through an internal circuit. The metal joints are connected by high-precision bearings.

4. The multimodal exophthalmos measurement device according to claim 2, characterized in that: The control unit includes a touch operation panel and a high-resolution liquid crystal display screen. The touch operation panel is located above the front of the control unit, and the high-resolution liquid crystal display screen is fixedly arranged below the front of the control unit.

5. The multimodal exophthalmos measurement device according to claim 1, characterized in that: The ultrasonic-optical integrated measurement probe includes an ultrasonic transducer, a projector, and a camera. The ultrasonic transducer is fixed to the front end of the measurement probe assembly via a front mounting base. Three micro-optical component mounting slots are evenly arranged around the ultrasonic transducer at a 120° angle. The micro-optical component mounting slots are used to precisely position the projector and camera.

6. The multimodal exophthalmos measurement device according to claim 5, characterized in that: The ultrasonic transducer is tightly connected to the positioning groove of the front end mounting seat in an interference fit manner.

7. A multimodal eyeball exophthalmos measurement device according to claim 5, characterized in that :The projector adopts a micro structured light projector, which is adapted to the micro optical component mounting slot through a metal bracket. One end of the metal bracket is fixed to the micro structured light projector by screws, and the other end is fixed to the optical component mounting frame by screws.

8. A multimodal eyeball exophthalmos measurement device according to claim 5, characterized in that The camera adopts a CMOS image sensor, which is installed in the micro-optical component installation groove through an elastic rubber pad.

9. A multimodal exophthalmos measurement device according to any one of claims 1 to 8, characterized in that: The device is used as follows: Step S1: The device starts and performs self-tests. Power is supplied through the power module at the bottom of the host. The power switch on the front of the host is turned on. The status indicator light is solid green. The host performs self-tests on key components such as the power module and the multi-layer printed circuit board. Step S2: Mode selection, selecting a normal or special measurement mode on the touch operation panel of the control unit; Step S3: Probe positioning: the control unit sends instructions, the host controls the rotation of the micro-motors in the metal joints of the flexible robotic arm, and the angle sensor feeds back angle data in real time until the integrated ultrasound-optical measurement probe is aligned with the patient's eyeball; Step S4: Ultrasonic measurement, including the following sub-steps: Step S41: signal transmission, the host sends a command to the circuit board through the dedicated interface to drive the ultrasonic transducer to transmit ultrasonic waves; Step S42: echo reception, the ultrasonic transducer receives the echo reflected from the surface of the eyeball, converts it into an electrical signal and transmits it to the circuit board; Step S43: Signal preprocessing, the circuit board performs bandpass filtering and wavelet threshold denoising on the electrical signal; Step S44: feature extraction, extracting the echo envelope through Hilbert transform and locating the corneal reflection peak; Step S45: Distance calculation: Calculate the distance from the corneal vertex Ec to the probe according to the formula X3=v×t / 2, where v is the speed of sound and t is the echo time; Step S5: Optical measurement, including the following sub-steps: Step S51: pattern projection, the host controls the micro structured light projector to project a structured light pattern onto the surface of the eyeball; Step S52: image acquisition, the CMOS image sensor captures the deformation pattern; Step S53: Pre-processing, the circuit board performs non-uniformity correction, contrast enhancement, and distortion correction on the image; Step S54: Feature extraction, the host calculates the three-dimensional contour data of the eyeball surface through edge detection and feature matching algorithms; Step S6: Data fusion, the host transforms the ultrasonically measured position data and the optically measured contour data into a unified coordinate system, performs fusion processing using a deep learning model, and outputs a protrusion value Δ; Step S7: result display and storage, the host transmits the Δ value to the high-resolution LCD screen of the control unit, and stores the measurement data in a structured manner in the data storage module; Step S8: Mode switching. If eyelid swelling or occlusion of the artificial eye is detected, the system automatically switches to the occlusion mode. If orbital deformity is detected, the system switches to the deformity mode.

Images