Intelligent ultrasonic workstation system with remote consultation and voice control image acquisition functions

Abstract

The invention discloses an intelligent ultrasonic workstation system with remote consultation and voice control image acquisition, and relates to the technical field of medical equipment. The ultrasonic imaging module is used for controlling an ultrasonic probe to transmit and receive ultrasonic waves and generating ultrasonic images and video streams, and the compatibility of large imaging depth and high resolution is realized by dynamically adjusting a transmission time interval and a depth section weight coefficient through a multi-frequency pulse fusion technology; a voice interaction module; and the voice interaction module is used for receiving a voice instruction of a user, converting the voice instruction into a system control command and text information, and providing voice feedback for the user. According to the invention, the examination efficiency is improved, and the diagnosis information omission caused by operation distraction is reduced.

Description

Technical Field

[0001] This invention relates to the field of medical equipment technology, and in particular to an intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition. Background Technology

[0002] Ultrasound examination is an indispensable non-invasive diagnostic tool in modern medicine and is widely used in various clinical departments. Simplifying ultrasound procedures is something that companies in the medical industry have been working on.

[0003] However, existing medical workstation systems have several limitations. Operationally, doctors need to use foot switches or manual buttons to control image acquisition. When operating the instrument with both hands, the foot switch restricts operational freedom, easily leading to acquisition delays, missed key images, or inaccurate labeling. Verbal descriptions of patient conditions cannot be recorded in real time; after examination, medical reports must be manually entered using a keyboard, which is time-consuming and prone to missing details. In terms of data management, image data and text reports are stored separately, lacking structured association. HIS and PACS systems are relatively independent; after image acquisition and report writing, manual uploading and entry are required, resulting in data delays and hindering the formation of a real-time, efficient clinical information flow. Furthermore, historical cases do not form a searchable and reusable knowledge system, making it difficult for doctors to find similar cases and generate standardized reports during research. In addition, most are local and closed architectures, unable to support multi-center, cross-institutional remote consultations and teaching, restricting the downward flow and sharing of high-quality medical resources, especially affecting primary healthcare institutions' access to real-time expert guidance and instruction. Therefore, there is room for improvement. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition. Its advantages lie in freeing doctors' hands and feet through voice interaction, providing real-time diagnostic assistance through artificial intelligence, and enabling real-time consultation and teaching through a remote collaboration platform, thereby comprehensively improving the efficiency, accuracy, and accessibility of ultrasound examinations.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: An intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition, including an ultrasound imaging module; The ultrasonic imaging module is used to control the ultrasonic probe to emit and receive ultrasonic waves, generate ultrasonic images and video streams, and use multi-frequency pulse fusion technology to achieve compatibility between large imaging depth and high resolution by dynamically adjusting the emission time interval and depth segment weight coefficient. Voice interaction module; the voice interaction module is used to receive the user's voice commands and convert them into system control commands and text information, and provide voice feedback to the user; An AI-assisted diagnostic module; the AI-assisted diagnostic module is connected to the ultrasound imaging module and is used to analyze the generated ultrasound images and video streams in real time or near real time; the module has a built-in trained deep learning model that can identify specific anatomical structures and lesion areas, and provide preliminary qualitative or quantitative analysis results, including lesion identification, benign or malignant risk warning and automatic measurement; Remote teaching module; The remote teaching module is used to establish a connection with the Internet platform to realize real-time audio and video communication with remote experts or students, ultrasound screen sharing, and real-time interactive annotation function based on the shared screen, so that remote experts can watch the examination process and provide real-time guidance. The central processing and control module is connected to all the above modules and is used to coordinate the work of each module, execute voice commands, integrate AI analysis results and remote interactive information, and display them uniformly on the user interface.

[0006] The present invention is further configured such that the voice interaction module includes a voice recognition unit and a voice synthesis unit. The voice recognition unit is used to receive the doctor's voice instructions and convert them into executable system control commands, including image freeze, storage, measurement, start recording, end recording, and diagnostic descriptions in text form. After enhanced training with medical corpus and professional instruction set, it can accurately recognize medical professional terms and operation instructions.

[0007] The present invention is further configured such that the speech synthesis unit is used to provide speech feedback to doctors without needing to be woken up, and the speech is converted into Chinese in real time for filling out professional reports.

[0008] The present invention is further configured such that the analysis results of the artificial intelligence-assisted diagnostic module are overlaid on the real-time ultrasound image in the form of visual labels, including bounding boxes, contour lines and heat maps, or pop-up prompts.

[0009] The present invention is further configured such that the artificial intelligence-assisted diagnosis module uses forward propagation of a convolutional neural network to output feature maps, and the calculation formula is as follows: , where x represents the input ultrasound image pixel matrix, W represents the convolution kernel weight parameters, b represents the bias term, * represents the convolution operation, f represents the activation function, and y represents the feature map output.

[0010] The present invention is further configured such that the deep learning model is trained using a cross-entropy loss function, and the training formula is as follows: Where N represents the number of training samples and C represents the number of classification categories. This represents the true label of the i-th sample. This represents the probability value predicted by the model.

[0011] The present invention is further configured such that the remote teaching module supports multiple consultation modes, including one-on-one guidance, one-to-many live teaching, and multi-expert consultation modes.

[0012] The invention is further configured such that the remote teaching module supports the establishment of a secure and encrypted communication link and allows remote users to perform real-time drawing annotations on their shared ultrasound screen.

[0013] An examination method for an intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition includes the following steps: Step 1: The doctor controls the acquisition, freezing, and storage of ultrasound images via voice commands; Step 2: The system analyzes ultrasound images in real time through the artificial intelligence module and provides auxiliary diagnostic information to the doctor in real time; Step 3: During the examination, the doctor inputs diagnostic descriptive statements via voice, which the system converts into text and records. Step 4: When needed, the doctor activates the remote collaboration function to connect with remote experts, share the real-time ultrasound screen, and receive remote guidance.

[0014] The beneficial effects of this invention are as follows: 1. Freeing doctors' hands and improving operational efficiency: Doctors can complete all key operations via voice without interrupting the scan, making the operation process smoother, significantly improving examination efficiency, and reducing the omission of diagnostic information due to distraction.

[0015] 2. Reduce over-reliance on doctors' experience: The built-in AI-assisted diagnosis module acts like a real-time online "advanced assistant," providing doctors with key imaging clues, effectively reducing missed diagnoses and misdiagnoses, and improving the consistency and accuracy of diagnoses.

[0016] 3. Breaking the limitations of time and space to realize the downward flow of high-quality medical resources: Through the remote collaboration module, primary care physicians can immediately obtain "cloud consultation" guidance from experts in higher-level hospitals when encountering difficult cases. At the same time, it also provides a convenient platform for standardized training and surgical demonstrations, fundamentally improving the overall level of primary care medical services. 4. Optimize workflows and ensure medical safety: Voice input of diagnostic descriptions combined with AI prompts can make diagnostic reports more standardized and structured, reducing the time spent on later note-taking and lowering the medical risks caused by typos or memory errors. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition proposed in this invention. Figure 2This is a schematic diagram of the intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition proposed in this invention. Figure 3 This is a schematic diagram of the intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition proposed in this invention. Detailed Implementation

[0018] The technical solution of this patent will be further described in detail below with reference to specific embodiments.

[0019] The embodiments of this patent are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this patent, and should not be construed as limiting this patent.

[0020] Reference Figure 1-3 The intelligent ultrasound workstation system, which features remote consultation and voice-controlled image acquisition, includes an ultrasound imaging module; The ultrasound imaging module is used to control the ultrasound probe to emit and receive ultrasound waves, generate ultrasound images and video streams, and use multi-frequency pulse fusion technology to achieve compatibility between large imaging depth and high resolution by dynamically adjusting the emission time interval and depth segment weight coefficient. Voice interaction module; The voice interaction module is used to receive the user's voice commands and convert them into system control commands and text information, and provide voice feedback to the user; Artificial intelligence-assisted diagnosis module; The artificial intelligence-assisted diagnosis module is connected to the ultrasound imaging module to analyze the generated ultrasound images and video streams in real time or near real time; This module has a built-in trained deep learning model that can identify specific anatomical structures and lesion areas, and provide preliminary qualitative or quantitative analysis results, including lesion identification, benign or malignant risk indication, and automatic measurement; Remote teaching module; The remote teaching module is used to establish a connection with the Internet platform to realize real-time audio and video communication with remote experts or trainees, ultrasound screen sharing, and real-time interactive annotation based on the shared screen, so that remote experts can watch the examination process and provide real-time guidance. Central processing and control module; The central processing and control module is connected to all the above modules to coordinate the work of each module, execute voice commands, integrate AI analysis results and remote interactive information, and display them uniformly on the user interface.

[0021] In this embodiment, the voice interaction module includes a voice recognition unit and a voice synthesis unit. The voice recognition unit receives voice commands from the doctor and converts them into executable system control commands, including image freeze, storage, measurement, start recording, end recording, and diagnostic descriptions in text form. After enhanced training with medical corpus and professional instruction sets, it can accurately recognize medical professional terms and operation instructions. The voice synthesis unit provides voice feedback to the doctor without requiring wake-up, and the voice is converted into Chinese in real time for filling out professional reports.

[0022] Furthermore, the analysis results of the AI-assisted diagnostic module are overlaid on the real-time ultrasound image in the form of visual labels, including bounding boxes, contour lines, and heatmaps, or pop-up prompts; the AI-assisted diagnostic module uses forward propagation of a convolutional neural network to output feature maps, and the calculation formula is as follows: Where x represents the input ultrasound image pixel matrix, W represents the convolution kernel weight parameters, b represents the bias term, * represents the convolution operation, f represents the activation function, and y represents the feature map output; the deep learning model uses the cross-entropy loss function for model training, and the training formula is: Where N represents the number of training samples and C represents the number of classification categories. This represents the true label of the i-th sample. This represents the probability value predicted by the model.

[0023] It is worth mentioning that the remote teaching module supports multiple consultation modes, including one-on-one guidance, one-to-many live teaching, and multi-expert consultation. The remote teaching module supports the establishment of secure and encrypted communication links and allows remote users to draw and annotate in real time on their shared ultrasound screen.

[0024] The examination method of an intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition includes the following steps: Step 1: The doctor controls the acquisition, freezing, and storage of ultrasound images via voice commands; Step 2: The system analyzes ultrasound images in real time through the artificial intelligence module and provides auxiliary diagnostic information to the doctor in real time; Step 3: During the examination, the doctor inputs diagnostic descriptive statements via voice, which the system converts into text and records. Step 4: When needed, the doctor activates the remote collaboration function to connect with remote experts, share the real-time ultrasound screen, and receive remote guidance.

[0025] First, the piezoelectric crystal in the ultrasound probe generates high-frequency vibrations (typically 2-18MHz) under electrical signal excitation, emitting ultrasonic pulses.12 When the sound waves propagate through human tissue, they are reflected when they encounter interfaces with different acoustic impedances (such as organ capsules and blood vessel walls). The reflected echoes are received by the same probe and converted into electrical signals. After amplification and filtering, the received echo signals are used by a computer to calculate the depth based on the echo time (1μs delay ≈ 0.77mm depth).12 The amplitude information is converted into a two-dimensional grayscale image through digital-to-analog conversion. Strongly reflective interfaces (such as bones) are displayed in white, while weakly reflective areas (such as liquids) are displayed in dark areas. This ultrasound imaging has the following three image display types: B-mode display: Multiple information lines are combined to form a tomographic image by electronically or mechanically scanning the sound beam. M-mode display: The probe position is fixed, and the depth-time curve in a single direction is recorded for cardiac motion analysis. Three-dimensional imaging: A two-dimensional high-density probe array (such as an 80×80 array element) is used for sound beam scanning to reconstruct a stereoscopic image in real time. After all the algorithms in the ultrasound image acquisition module have been processed, a complete RGB pixel image is formed and transmitted to the central processing and control module.

[0026] After the central processing and control module processes the ultrasound images using algorithms, they are displayed on the display module. Doctors can then use the voice interaction module to issue commands such as "freeze image," "save current frame," "start recording," or "description: a hypoechoic nodule of approximately 2cm is visible, with indistinct borders." Upon receiving these commands, the central processing module controls the ultrasound imaging module to execute the corresponding actions. Simultaneously, the AI-assisted diagnostic module continuously analyzes the real-time images. If a suspicious lesion is detected, it is highlighted on the screen with a message such as "Suspected nodule, measurement recommended." For assistance, doctors can activate the remote collaboration module via voice to connect with remote experts. The experts can view the real-time ultrasound images on their terminals and provide guidance through voice and annotation tools.

[0027] The backend system continuously builds an intelligent case knowledge base, automatically classifying, clustering, and labeling case information. Doctors can use keywords to search for historical cases during research and teaching. Furthermore, the data is strictly encrypted and subject to access control, meeting medical information security requirements. In terms of hospital information integration, the system can communicate bidirectionally with HIS, PACS and electronic medical record systems to achieve real-time information sharing. It can also connect to third-party AI image analysis modules and has recording and playback functions to establish a case teaching database, comprehensively improving medical work efficiency and treatment quality.

[0028] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition, characterized in that, Includes an ultrasound imaging module; The ultrasonic imaging module is used to control the ultrasonic probe to emit and receive ultrasonic waves, generate ultrasonic images and video streams, and use multi-frequency pulse fusion technology to achieve compatibility between large imaging depth and high resolution by dynamically adjusting the emission time interval and depth segment weight coefficient. Voice interaction module; the voice interaction module is used to receive the user's voice commands and convert them into system control commands and text information, and provide voice feedback to the user; An AI-assisted diagnostic module; the AI-assisted diagnostic module is connected to the ultrasound imaging module and is used to analyze the generated ultrasound images and video streams in real time or near real time; the module has a built-in trained deep learning model that can identify specific anatomical structures and lesion areas, and provide preliminary qualitative or quantitative analysis results, including lesion identification, benign or malignant risk warning and automatic measurement; Remote teaching module; The remote teaching module is used to establish a connection with the Internet platform to realize real-time audio and video communication with remote experts or students, ultrasound screen sharing, and real-time interactive annotation function based on the shared screen, so that remote experts can watch the examination process and provide real-time guidance. The central processing and control module is connected to all the above modules and is used to coordinate the work of each module, execute voice commands, integrate AI analysis results and remote interactive information, and display them uniformly on the user interface.

2. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition as described in claim 1, characterized in that, The voice interaction module includes a voice recognition unit and a voice synthesis unit. The voice recognition unit receives voice commands from doctors and converts them into executable system control commands, including image freeze, storage, measurement, start recording, end recording, and diagnostic descriptions in text form. After enhanced training with medical corpus and professional instruction sets, it can accurately recognize medical professional terms and operation instructions.

3. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 2, characterized in that, The speech synthesis unit is used to provide voice feedback to doctors without needing to be woken up, and the speech is converted into Chinese in real time for filling out professional reports.

4. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 1, characterized in that, The analysis results of the AI-assisted diagnostic module are displayed on the real-time ultrasound image in the form of visual labels, including bounding boxes, outlines, and heatmaps, or pop-up prompts.

5. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 4, characterized in that, The AI-assisted diagnostic module uses forward propagation of a convolutional neural network to output feature maps, and the calculation formula is as follows: , where x represents the input ultrasound image pixel matrix, W represents the convolution kernel weight parameters, b represents the bias term, * represents the convolution operation, f represents the activation function, and y represents the feature map output.

6. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 5, characterized in that, The deep learning model is trained using the cross-entropy loss function, and the training formula is as follows: Where N represents the number of training samples and C represents the number of classification categories. This represents the true label of the i-th sample. This represents the probability value predicted by the model.

7. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 1, characterized in that, The remote teaching module supports multiple consultation modes, including one-on-one guidance, one-to-many live teaching, and multi-expert consultation.

8. The intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition according to claim 7, characterized in that, The remote teaching module supports the establishment of a secure and encrypted communication link and allows remote users to perform real-time drawing and annotation on their shared ultrasound screen.

9. An examination method for an intelligent ultrasound workstation system with remote consultation and voice-controlled image acquisition as described in any one of claims 1-8, characterized in that, Includes the following steps: Step 1: The doctor controls the acquisition, freezing, and storage of ultrasound images via voice commands; Step 2: The system analyzes ultrasound images in real time through the artificial intelligence module and provides auxiliary diagnostic information to the doctor in real time; Step 3: During the examination, the doctor inputs diagnostic descriptive statements via voice, which the system converts into text and records. Step 4: When needed, the doctor activates the remote collaboration function to connect with remote experts, share the real-time ultrasound screen, and receive remote guidance.

Images