CN113222988A - Infusion tube liquid level identification method based on deep learning and image overdivision - Google Patents
Infusion tube liquid level identification method based on deep learning and image overdivision Download PDFInfo
- Publication number
- CN113222988A CN113222988A CN202110633471.6A CN202110633471A CN113222988A CN 113222988 A CN113222988 A CN 113222988A CN 202110633471 A CN202110633471 A CN 202110633471A CN 113222988 A CN113222988 A CN 113222988A
- Authority
- CN
- China
- Prior art keywords
- image
- liquid level
- infusion tube
- deep learning
- method based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001802 infusion Methods 0.000 title claims abstract description 70
- 239000007788 liquid Substances 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000013135 deep learning Methods 0.000 title claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000001514 detection method Methods 0.000 claims description 20
- 238000002372 labelling Methods 0.000 claims description 9
- 238000002360 preparation method Methods 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 238000012549 training Methods 0.000 description 8
- 239000003814 drug Substances 0.000 description 6
- 230000000694 effects Effects 0.000 description 3
- 238000012806 monitoring device Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000001990 intravenous administration Methods 0.000 description 2
- 230000000474 nursing effect Effects 0.000 description 2
- 208000028399 Critical Illness Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000002637 fluid replacement therapy Methods 0.000 description 1
- 230000002706 hydrostatic effect Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30021—Catheter; Guide wire
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an infusion tube liquid level identification method based on deep learning and image overdivision, which belongs to the technical field of intelligent image processing, and is used for judging whether the liquid level of each infusion tube is lower than a critical value or not by monitoring the liquid level conditions of one or more infusion tubes and each infusion tube in a video in real time so as to provide an alarm signal and ensure the infusion safety of patients. The method can identify a plurality of infusion tubes simultaneously, the real-time performance meets the actual requirement, the related medical accidents can be effectively reduced or even eliminated, and the method has high application value.
Description
Technical Field
The invention relates to the technical field of intelligent image processing, in particular to a transfusion tube liquid level identification method based on deep learning and image overdivision.
Background
Intravenous infusion is a common medical procedure. It is a medicine supply mode for inputting large dose of injection into body by utilizing the principle of atmospheric pressure and hydrostatic pressure. In the actual medical environment, the problem that the liquid level height of the infusion tube cannot be checked in time due to the fact that patients are numerous, the infusion site is large and the like always troubles nursing staff and patients. Due to the limitation of manpower and material resources, limited intelligent detection equipment and medical care personnel need to be preferentially supplied to critically ill patients for use; the infusion hall of the mild patients can only be managed by a plurality of medical care personnel. In most cases, the patient only can pay attention to the amount of the residual liquid medicine in the infusion tube. Therefore, patients often sleep due to fatigue, and the disease is not discovered until 1 or 2 hours after the liquid medicine is dripped. In order to avoid the occurrence of similar events and to enable patients to have better medical experience, an inexpensive system capable of tracking and detecting the liquid level condition of the infusion tube of each patient in real time is needed.
At present, there are many automatic monitoring systems, for example, chinese patent publication No. CN112604074A discloses a multifunctional automatic transfusion device, which can automatically perform the actions of inserting needle, pulling needle and replacing transfusion bottle. Meanwhile, the automatic infusion adjusting assembly can monitor the infusion speed in real time and control the infusion speed to be always in the range required by the medical advice. Chinese patent publication No. CN112121258A discloses an infusion monitoring device and an infusion monitoring method, which are used when intravenous infusion is performed, and monitor the liquid level of an infusion bottle, the flow rate of a liquid medicine, and whether the liquid medicine drips out of a drip cup by detecting a camera. The invention discloses an intelligent monitoring system and method applied to clinical infusion nursing, which integrates remote alarm, automatic clamping, infusion medicine and patient information display, greatly reduces unnecessary manual infusion detection burden of patients and accompanying personnel, and improves the fluid replacement efficiency of medical personnel. However, these patents require special devices or special sensors, which greatly limit their applications.
The development of computer image and video technology at present provides possibility for automatically tracking the transfusion condition, and the liquid level condition of each transfusion tube is detected in real time by adopting a video image processing and deep learning method. The detection system shoots and records the infusion hall through the camera and uploads the shot and recorded infusion hall to the PC end, and the target object can be screened out by the video processing program of the PC end: the infusion tubes detect the liquid level of each infusion tube in real time, and send out alarm signals when the liquid level of the infusion tubes is lower than a critical value so as to ensure the infusion safety of patients.
Disclosure of Invention
The invention aims to provide an infusion tube liquid level identification method based on deep learning and image overdivision, which judges whether the liquid level of each infusion tube is lower than a critical value or not by monitoring the liquid level conditions of one or more infusion tubes and each infusion tube in a video in real time so as to provide an alarm signal and ensure the infusion safety of patients. In order to achieve the above object, the infusion tube liquid level identification method based on depth learning and image super-resolution provided by the invention specifically comprises the following steps as shown in fig. 1:
1) intercepting image data from a camera;
2) carrying out infusion tube detection by using YOLOv 5;
3) judging whether the infusion tube is detected or not, and if not, going to step 9);
4) cutting the image according to the boundary information and reserving the original image;
5) amplifying the cut image by using a DRN-master;
6) carrying out liquid level detection on the amplified image by using YOLOv 5;
7) judging whether the liquid level is detected or not, and if not, going to step 9);
8) labeling on the original image according to the information of the bounding box, as shown in fig. 5;
9) and outputting the image in real time.
According to the technical scheme, high-definition shooting and recording equipment is used for recording, the angle of the camera can be rotated or fixed, after a relatively stable video image is obtained, the video information is preprocessed, then the trained model is used for classification and judgment, and alarm information is given according to the judgment result. The method can identify a plurality of infusion tubes simultaneously, the real-time performance meets the actual requirement, the related medical accidents can be effectively reduced or even eliminated, and the method has high application value.
Preferably, in step 1), image data is intercepted from a camera; the sample set preparation method specifically comprises two steps:
step 1-1) intercepting image frames in a real-time video; the video recording was performed when 50%, 35%, 20% and less than 5% of the liquid remained in the infusion tube, and the video recording frame rate was 30fps and the pixel was 720p, respectively. During recording, the factors such as shooting angle, distance and the like are changed, and then the picture is intercepted from the video through software free studio. An example of a sample of a video set is shown in fig. 2.
Step 1-2) sample labeling; and manually labeling the shot image by using a labelImg sample labeling tool, and programming to convert the xml file into a txt file required by YOLO.
Preferably, in step 2), the infusion tube is detected by using YOLOv 5; in practical application, one or more cameras are often adopted to monitor the condition of the whole infusion space, the definition of some monitoring devices is often low, the difficulty of detecting the infusion tube from a video is high, the YOLOv5m.pt of YOLOv5 is adopted as the weight of a model, the model gives consideration to the size of a weight file, the difficulty of training, the detection accuracy and the detection speed, and the performance is balanced.
Preferably, in step 4), the image is cut according to the boundary information, and the original image is reserved; and cutting the detected infusion tube image according to the boundary frame for subsequent image processing, liquid level detection and the like.
Preferably, in step 5), the cropped image is enlarged by a DRN-master; in order to improve the identification capability of the system on small targets such as liquid level and the like, the DRN-master image super-resolution algorithm is adopted in the embodiment, the image of the infusion tube cut in the step 4) is amplified, and characteristic information such as boundary, liquid level and the like is highlighted.
The DRN-master designs a network structure of dual regression, which not only provides additional training supervision, but also allows the algorithm to be trained by using non-one-to-one corresponding high and low resolution images. The network structure of the DRN-master is shown in fig. 3, and the effect of increasing the picture definition by 4 times using the DRN-master is shown in fig. 4 in comparison.
Preferably, in step 6), the amplified image is subjected to liquid level detection by using YOLOv 5; the level detection in this example uses YOLOv5s.pt from YOLOv5 as model weight. The depth of the network of yolov5s and the width of the characteristic diagram are small, so the training difficulty is low, the detection speed is high, and the accuracy can meet the requirement.
Compared with the prior art, the invention has the beneficial effects that:
the invention relates to an infusion tube liquid level identification method based on deep learning and image overdivision, which judges whether the liquid level of each infusion tube is lower than a critical value or not by monitoring the liquid level conditions of one or more infusion tubes and each infusion tube in a video in real time so as to provide an alarm signal and ensure the infusion safety of patients. The method can identify a plurality of infusion tubes simultaneously, the real-time performance meets the actual requirement, the related medical accidents can be effectively reduced or even eliminated, and the method has high application value.
Drawings
FIG. 1 is a flow chart of an infusion tube liquid level identification method based on depth learning and image hyper-resolution in an embodiment of the present invention;
FIG. 2 is an example of a sample video set according to an embodiment of the present invention;
FIG. 3 is a network structure of a DRN-master in an embodiment of the present invention;
FIG. 4 is a comparison graph of the DRN-master super resolution processing effect (magnified 4 times, right is the input image) in the embodiment of the present invention, wherein (a) is the input image and (b) is the processed image;
FIG. 5 is an image labeled according to bounding box information in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the following embodiments and accompanying drawings.
Examples
Referring to fig. 1 to 5, a method for identifying a liquid level of an infusion tube based on depth learning and image hyper-resolution according to the present embodiment is shown in fig. 1, and includes the following steps:
and step S100, intercepting image data from a camera. And judging whether the sample is made into a sample set or not, and if not, going to the step S200.
Step S110, intercepting image frames in a real-time video; the video recording was performed when 50%, 35%, 20% and less than 5% of the liquid remained in the infusion tube, and the video recording frame rate was 30fps and the pixel was 720p, respectively. During recording, the factors such as shooting angle, distance and the like are changed, and then the picture is intercepted from the video through software free studio. An example of a sample of a video set is shown in fig. 2.
Step S120, marking a sample; manually labeling the shot image by using a labelImg sample labeling tool, programming to convert the xml file into a txt file required by YOLO, and performing the following steps according to the sequence of 4: a ratio of 1 generates a training set and a test set. Format of each line within txt:
<object-class><x_center><y_center><width><height>。
in step S200, the infusion tube is detected by YOLOv 5. In practical application, one or more cameras are often adopted to monitor the condition of the whole infusion space, the definition of some monitoring devices is often low, the difficulty in detecting the infusion tube directly from a video is high, YOLOv5m.pt of YOLOv5 is adopted as the model weight, the model gives consideration to the size of a weight file, the difficulty in training, the detection accuracy and the detection speed, and the performance is balanced. In this example, under yolov5s.pt weight, the maximum set batch size (batch-size) is 21, the input picture size is 352 × 352, and the total training step number is 800.
Step S300, judging whether the infusion tube is detected, if not, going to step S900.
And step S400, cutting the image according to the boundary information and reserving the original image.
And step S500, amplifying the cut image by using a DRN-master. In order to improve the identification capability of the system on small targets such as liquid level and the like, the DRN-master image super-resolution algorithm is adopted in the embodiment, the image of the infusion tube cut in the step 4) is amplified, and characteristic information such as boundary, liquid level and the like is highlighted.
The DRN-master designs a network structure of dual regression, which not only provides additional training supervision, but also allows the algorithm to be trained by using non-one-to-one corresponding high and low resolution images. The network structure of the DRN-master is shown in fig. 3, and the effect of increasing the picture definition by 4 times using the DRN-master is shown in fig. 4 in comparison.
In step S600, the level of the enlarged image is detected by YOLOv 5. The level detection in this example uses YOLOv5s.pt from YOLOv5 as model weight. The depth of the network of yolov5s and the width of the characteristic diagram are small, so the training difficulty is low, the detection speed is high, and the accuracy can meet the requirement.
Step S700, judging whether the liquid level is detected, if not, going to step S900.
In step S800, the original image is marked according to the bounding box information, as shown in fig. 5.
And step S900, outputting the image in real time.
Claims (6)
1. An infusion tube liquid level identification method based on deep learning and image overdivision is characterized by comprising the following steps:
1) intercepting image data from a camera;
2) carrying out infusion tube detection by using YOLOv 5;
3) judging whether the infusion tube is detected or not, and if not, going to step 9);
4) cutting the image according to the boundary information and reserving the original image;
5) amplifying the cut image by using a DRN-master;
6) carrying out liquid level detection on the amplified image by using YOLOv 5;
7) judging whether the liquid level is detected or not, and if not, going to step 9);
8) marking on the original image according to the information of the boundary frame;
9) and outputting the image in real time.
2. The infusion tube liquid level identification method based on the deep learning and the image super-resolution as claimed in claim 1, wherein in the step 1), image data is intercepted from a camera; the sample set preparation method specifically comprises two steps:
step 1-1) intercepting image frames in a real-time video; during recording, the factors such as shooting angle, distance and the like are changed, and then the picture is intercepted from the video through software free studio.
Step 1-2) sample labeling; and manually labeling the shot image by using a labelImg sample labeling tool, and programming to convert the xml file into a txt file required by YOLO.
3. The infusion tube liquid level identification method based on the deep learning and the image super-resolution as claimed in claim 1, wherein in the step 2), the infusion tube is detected by YOLOv 5; this example uses YOLOv5m.pt from YOLOv5 as the model weight.
4. The infusion tube liquid level identification method based on the deep learning and the image super-resolution as claimed in claim 1, wherein in the step 4), the image is cut according to the boundary information, and the original image is reserved; and cutting the detected infusion tube image according to the boundary frame for subsequent image processing, liquid level detection and the like.
5. The infusion tube liquid level recognition method based on the deep learning and the image super-resolution as claimed in claim 1, wherein in the step 5), the cut image is amplified by a DRN-master; in order to improve the identification capability of the system on small targets such as liquid level and the like, the DRN-master image super-resolution algorithm is adopted in the embodiment, the image of the infusion tube cut in the step 4) is amplified, and characteristic information such as boundary, liquid level and the like is highlighted.
6. The infusion tube liquid level identification method based on the deep learning and the image super-resolution as claimed in claim 1, wherein in the step 6), the amplified image is subjected to liquid level detection by using YOLOv 5; the level detection in this example uses YOLOv5s.pt from YOLOv5 as model weight.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110633471.6A CN113222988A (en) | 2021-06-07 | 2021-06-07 | Infusion tube liquid level identification method based on deep learning and image overdivision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110633471.6A CN113222988A (en) | 2021-06-07 | 2021-06-07 | Infusion tube liquid level identification method based on deep learning and image overdivision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113222988A true CN113222988A (en) | 2021-08-06 |
Family
ID=77083071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110633471.6A Pending CN113222988A (en) | 2021-06-07 | 2021-06-07 | Infusion tube liquid level identification method based on deep learning and image overdivision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113222988A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115019456A (en) * | 2022-05-31 | 2022-09-06 | 浙江大学 | Ward calling system and calling method thereof |
| CN117392135A (en) * | 2023-12-12 | 2024-01-12 | 深圳市普朗医疗科技发展有限公司 | Injection dose analysis method and system based on image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635666A (en) * | 2018-11-16 | 2019-04-16 | 南京航空航天大学 | A kind of image object rapid detection method based on deep learning |
| CN110188807A (en) * | 2019-05-21 | 2019-08-30 | 重庆大学 | Tunnel pedestrian target detection method based on cascade super-resolution network and improvement Faster R-CNN |
-
2021
- 2021-06-07 CN CN202110633471.6A patent/CN113222988A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635666A (en) * | 2018-11-16 | 2019-04-16 | 南京航空航天大学 | A kind of image object rapid detection method based on deep learning |
| CN110188807A (en) * | 2019-05-21 | 2019-08-30 | 重庆大学 | Tunnel pedestrian target detection method based on cascade super-resolution network and improvement Faster R-CNN |
Non-Patent Citations (2)
| Title |
|---|
| YONG GUO 等: "Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution", 《2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION （CVPR)》 * |
| ZEYONG HUANG 等: "Infusion port level detection for intravenous infusion based on Yolo v3 neural network", 《MATHEMATICAL BIOSCIENCES AND ENGINEERING》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115019456A (en) * | 2022-05-31 | 2022-09-06 | 浙江大学 | Ward calling system and calling method thereof |
| CN117392135A (en) * | 2023-12-12 | 2024-01-12 | 深圳市普朗医疗科技发展有限公司 | Injection dose analysis method and system based on image |
| CN117392135B (en) * | 2023-12-12 | 2024-03-01 | 深圳市普朗医疗科技发展有限公司 | Injection dose analysis method and system based on image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113222988A (en) | Infusion tube liquid level identification method based on deep learning and image overdivision | |
| US11145130B2 (en) | Method for automatically capturing data from non-networked production equipment | |
| CN108830824A (en) | Transfusion liquid facial vision detects alarm system and its detection method | |
| WO2011094639A3 (en) | Methods and systems for analyzing, prioritizing, visualizing, and reporting medical images | |
| US7853063B2 (en) | Method for producing a cropped medical image | |
| KR20090006295A (en) | Apparatus, system and method for examination of cell slide and readable medium thereof | |
| CN113891072B (en) | Video monitoring and anomaly analysis system and method based on hundred million-level pixel data | |
| US20070140538A1 (en) | Method for processing unenhanced medical images | |
| JP2017069963A (en) | Data collection system of display panel and operation panel, device, method, program and recording medium | |
| CN101546376A (en) | Human body biological information acquisition system, human face photo acquisition and quality inspection system and method | |
| US11900549B2 (en) | Method for automatically capturing data from non-networked production equipment | |
| CN113610077A (en) | System method and equipment for monitoring and analyzing dissolution behavior by using artificial intelligence image recognition technology | |
| CN113485555B (en) | Medical image film reading method, electronic equipment and storage medium | |
| CN109223178B (en) | Cavity mirror intelligent edge computing system with target positioning function | |
| CN113537064A (en) | Weak pig automatic detection marking method and system | |
| CN100384222C (en) | Method and device for adjusting camera position | |
| KR102423255B1 (en) | System and method for integrated management synchronizing sensor information and video information based on the internet of things | |
| CN112862752A (en) | Image processing display method, system electronic equipment and storage medium | |
| CN116912900A (en) | Method and device for detecting falling of breathing mask of patient | |
| CN115514918B (en) | Remote video method, cloud platform, communication mobile platform and storage medium | |
| CN110599471A (en) | Rain gauge horizontal monitoring system based on image processing and detection method thereof | |
| US20230011929A1 (en) | Interface between heart pump controller database and hospital | |
| CN117524448A (en) | Medical image display mode adjusting method and device | |
| CN105286897A (en) | Medical image diagnosis system | |
| CN113901934A (en) | Intelligent visual detection method, system and device for large infusion package product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210806 |