CN116628457A - Harmful gas detection method and device in operation of magnetic resonance equipment - Google Patents
Harmful gas detection method and device in operation of magnetic resonance equipment Download PDFInfo
- Publication number
- CN116628457A CN116628457A CN202310923517.7A CN202310923517A CN116628457A CN 116628457 A CN116628457 A CN 116628457A CN 202310923517 A CN202310923517 A CN 202310923517A CN 116628457 A CN116628457 A CN 116628457A
- Authority
- CN
- China
- Prior art keywords
- data
- magnetic resonance
- sequence characteristics
- time sequence
- cooling system
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims description 13
- 238000001816 cooling Methods 0.000 claims abstract description 55
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012423 maintenance Methods 0.000 claims abstract description 16
- 230000004927 fusion Effects 0.000 claims abstract description 5
- 239000007789 gas Substances 0.000 claims description 82
- 239000001307 helium Substances 0.000 claims description 15
- 229910052734 helium Inorganic materials 0.000 claims description 15
- SWQJXJOGLNCZEY-UHFFFAOYSA-N helium atom Chemical compound [He] SWQJXJOGLNCZEY-UHFFFAOYSA-N 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000004821 distillation Methods 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 claims 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 16
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 14
- 239000007788 liquid Substances 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 229910052757 nitrogen Inorganic materials 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000010791 quenching Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 125000002015 acyclic group Chemical group 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N9/00—Investigating density or specific gravity of materials; Analysing materials by determining density or specific gravity
- G01N9/36—Analysing materials by measuring the density or specific gravity, e.g. determining quantity of moisture
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
- G01R33/28—Details of apparatus provided for in groups G01R33/44 - G01R33/64
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Combustion & Propulsion (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention relates to a method and a device for detecting harmful gas in the operation of magnetic resonance equipment, wherein the method comprises the following steps: acquiring data of a cooling system of target magnetic resonance equipment, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; constructing a training data set through a local rough model; respectively extracting time sequence characteristics and non-time sequence characteristics based on a transducer network; based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment, fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model; training the double-flow multi-mode model to obtain a trained double-flow multi-mode model; and predicting the harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment. According to the invention, feature fusion is carried out through the transducer and the double-flow multi-mode model, so that the timeliness and accuracy of the model for predicting harmful gas in the operation of the magnetic resonance equipment are improved.
Description
Technical Field
The invention belongs to the technical field of gas detection and machine learning, and particularly relates to a method and a device for detecting harmful gas in the operation of magnetic resonance equipment.
Background
Magnetic resonance imaging (MRI, magnetic Resonance Imaging) equipment is one of the medical equipment currently in hospitals which is advanced, expensive and in large demand, and plays an important role in clinical disease diagnosis and treatment detection. The MRI apparatus is constantly depleted of liquid nitrogen during installation and normal use. In order to maintain the liquid level of the liquid nitrogen container stable, prevent quench (below 35% of full liquid nitrogen level) and keep the magnet operating properly, it is necessary to perfuse the liquid periodically. Liquid nitrogen is high in cost, if leakage occurs, the examination of a patient is delayed, the use cost of a hospital is greatly increased, and people can be choked to death by high-concentration nitrogen. The research shows that the nitrogen leakage (accounting for 23.3%) is the first of the adverse event types of 30 MRI devices in China, so that the research on the nitrogen leakage of the MRI room has important significance for enterprises, medical institutions and supervision departments to control the risk of the MRI devices and ensure the efficient operation of the devices. The record of the magnet refrigerating system data and the magnet running state data is checked and recorded in daily cultivation, and the abnormality is found to inform manufacturers of maintenance in time so as to prevent the risk of magnet quench and even magnet rejection caused by liquid helium exhaustion.
Failure of the cooling system of magnetic resonance imaging can lead to image distortion and increased noise in the magnetic resonance imaging, the principle being as follows:
1. superconducting magnets require extremely low temperature (several kelvin) liquid helium to maintain superconducting states, creating a powerful static magnetic field. If the cooling system fails, the liquid helium can evaporate, the superconducting magnet can return to a normal state, and the static magnetic field can drop or disappear sharply, resulting in image distortion or failure to image.
2. Both electromagnetic wave illumination and signal detection require a stable and uniform magnetic field that can produce imaging noise or contrast ambiguity if the magnetic field is abnormal. Cooling failure directly causes magnetic field anomalies, so noise and distortion can increase.
3. Clinical mri cryogenically components such as coldheads may be as low as-269 ℃ and ambient temperature may be at 20 ℃. This requires a large temperature differential to be created and maintained. If cooling fails, the low temperature components will rapidly warm up and such severe temperature changes will also interfere with imaging.
Disclosure of Invention
In order to improve the timeliness and accuracy of the harmful gas prediction of the magnetic resonance equipment, in a first aspect of the invention, there is provided a harmful gas detection method in the operation of the magnetic resonance equipment, comprising: acquiring data of a cooling system of target magnetic resonance equipment, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model; based on a transducer network, respectively extracting time sequence characteristics and non-time sequence characteristics of data of a cooling system, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment, fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model; based on the fused time sequence characteristics and non-time sequence characteristics, training the double-flow multi-mode model through a training data set until the error tends to be stable and is lower than a threshold value, and obtaining a trained double-flow multi-mode model; and predicting the harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment.
In some embodiments of the invention, the constructing, by means of the local asperity model, a training data set based on the data of the cooling system, the magnetic resonance image, the operational data of the gradient coils and the gas density data between corresponding scans comprises: constructing a first training data set based on the data of the cooling system and the gas density data between corresponding scans by means of a local roughness model; and aligning the first training data set with the data of the cooling system, the magnetic resonance image and the operation data of the gradient coil through a label of preset gas leakage to obtain a second training data set.
In some embodiments of the invention, the transforming network based extraction of timing and non-timing features of the cooling system data, the gradient coil operational data, the magnetic resonance image and the gas density data between corresponding scans, respectively, comprises: extracting non-sequential features of the data of the cooling system and the gas density data through a first transducer network; and extracting the operation data of the gradient coil and the time sequence characteristics of the magnetic resonance image through a second transducer network.
Further, the non-sequential features of extracting data of the cooling system and gas density data via the first transducer network include: respectively extracting data of a cooling system and gas density data by using a convolutional neural network to obtain a first feature vector and a second feature vector; and fusing the first characteristic vector and the second characteristic vector by using a transducer network through the attention head and preset weighting parameters to obtain the non-sequence characteristics of the data of the cooling system and the gas density data.
In some embodiments of the present invention, the fusing the non-timing feature and the timing feature through the dual-flow multi-modal model based on the knowledge graph of the operation and maintenance data of the magnetic resonance device includes: acquiring non-time sequence characteristics of the magnetic resonance equipment based on a knowledge graph of the operation and maintenance data of the magnetic resonance equipment; and fusing the non-time sequence characteristics and the time sequence characteristics through a self-distillation method and a double-flow multi-mode model.
In the foregoing embodiment, the predicting the harmful gas in the target magnetic resonance device using the dual-flow multi-mode model and the real-time data of the target magnetic resonance device includes: predicting the real-time air density between devices where the target magnetic resonance device is located by using the double-flow multi-mode model and the real-time data of the target magnetic resonance device; and predicting the spatial distribution of helium of the target magnetic resonance equipment based on an indoor physical model among the equipment where the target magnetic resonance equipment is located and the real-time air density.
In a second aspect of the present invention, there is provided a harmful gas detection system in operation of a magnetic resonance apparatus, comprising: the acquisition module is used for acquiring data of a cooling system of the target magnetic resonance equipment, operation data of the gradient coil, magnetic resonance images and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model; the extraction module is used for respectively extracting time sequence characteristics and non-time sequence characteristics of the data of the cooling system, the operation data of the gradient coil, the magnetic resonance image and the gas density data between corresponding scans based on the transducer network; the fusion module is used for fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model based on a knowledge graph of the operation and maintenance data of the magnetic resonance equipment; the training module is used for training the double-flow multi-mode model through a training data set based on the fused time sequence characteristics and non-time sequence characteristics until the error tends to be stable and is lower than a threshold value, so as to obtain a trained double-flow multi-mode model; and the prediction module is used for predicting the harmful gas of the target magnetic resonance equipment by utilizing the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment.
In a third aspect of the present invention, there is provided an electronic apparatus comprising: one or more processors; and a storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method for detecting a harmful gas in operation of a magnetic resonance apparatus provided in the first aspect of the present invention.
In a fourth aspect of the present invention, there is provided a computer readable medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the method for detecting a harmful gas in operation of a magnetic resonance apparatus provided in the first aspect of the present invention.
The beneficial effects of the invention are as follows:
the invention relates to a method and a device for detecting harmful gas in the operation of magnetic resonance equipment, wherein the method comprises the following steps: acquiring data of a cooling system of target magnetic resonance equipment, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; constructing a training data set through a local rough model; respectively extracting time sequence characteristics and non-time sequence characteristics based on a transducer network; based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment, fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model; training the double-flow multi-mode model through a training data set to obtain a trained double-flow multi-mode model; and predicting the harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment. According to the invention, the sequential characteristics and the non-sequential characteristics are respectively extracted through the transducer, so that the coupling between the characteristics is improved, the characteristic fusion is performed through the double-flow multi-mode model, and the prediction result is output, so that compared with a general prediction model, the timeliness and the accuracy of the harmful gas prediction in the operation of the magnetic resonance equipment are improved.
Drawings
FIG. 1 is a schematic flow diagram of a method for detecting harmful gases during operation of a magnetic resonance apparatus in accordance with some embodiments of the present invention;
figure 2 is a schematic diagram of the basic principle of a method for detecting harmful gases in operation of a magnetic resonance apparatus in some embodiments of the present invention;
FIG. 3 is a schematic diagram of a method for detecting harmful gases during operation of a magnetic resonance apparatus according to some embodiments of the present invention;
FIG. 4 is a schematic diagram of an in-room physical model of a corresponding MRI room of a magnetic resonance apparatus in some embodiments of the present invention;
FIG. 5 is a schematic diagram of a harmful gas detection unit in operation of a magnetic resonance apparatus in accordance with some embodiments of the present invention;
fig. 6 is a schematic structural diagram of an electronic device in some embodiments of the invention.
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
Referring to fig. 1 and 2, in a first aspect of the present invention, there is provided a method of detecting harmful gases in operation of a magnetic resonance apparatus, comprising: s100, acquiring data of a cooling system of target magnetic resonance equipment, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model; s200, respectively extracting time sequence characteristics and non-time sequence characteristics of data of a cooling system, operation data of a gradient coil, magnetic resonance images and gas density data between corresponding scans based on a transducer network; s300, fusing the non-time sequence features and the time sequence features through a double-flow multi-mode model based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment; s400, training the double-flow multi-mode model through a training data set based on the fused time sequence characteristics and non-time sequence characteristics until the error tends to be stable and is lower than a threshold value, so as to obtain a trained double-flow multi-mode model; s500, predicting harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and real-time data of the target magnetic resonance equipment.
In step S100 of some embodiments of the present invention, the constructing, by means of the local roughness model, a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coils and the gas density data between corresponding scans comprises: s101, constructing a first training data set based on data of the cooling system and gas density data between corresponding scans through a local rough model;
specifically, corresponding one or more of data of a cooling system, magnetic resonance images, operation data of a gradient coil and gas density data between corresponding scans are taken as two sets, and a first data set containing mixed attribute and limited tag data is constructed based on different granularity parameters. The defects that the samples are difficult to collect and the samples are few when helium leaks are overcome through the local rough model.
S102, aligning the first training data set with data of the cooling system, the magnetic resonance image and operation data of the gradient coil through a preset gas leakage label to obtain a second training data set.
The cooling data includes temperature, pressure, density, flow rate concentration and humidity, liquid level and other data of liquid or gas involved in a compressor, a helium press, an engine, a cooling pipeline and a heat exchanger in the cooling system; for example, the relationship analysis between the cold head and the helium machine is mainly to perform heat exchange by using high purity helium distributed in a closed circulation pipeline, if the pressure of helium circulated by the cold head and the helium press is insufficient and the operation state cannot be supported, the refrigeration efficiency of the cold head is reduced, and the loss amount of the helium in the magnetic liquid is increased, so that the pressure value of the helium must be observed and recorded in detail, and the high purity helium should be replenished immediately once the pressure value reaches 16-17 bar. Without losing generality, MRI equipment is divided according to different areas, and is mainly divided into an operation room, a magnet room, an equipment room and an outdoor unit 4; the coils are arranged between the magnets, and the main data related to the magnets or gradient coils include the magnetic field strength, the magnetic field gradient and the temperature and humidity between the magnets.
Magnetic resonance imaging involves obtaining sequential images of different cross-sections of different objects (typically the human body) through a series of scans. In addition, the characteristics of images such as contrast, resolution, signal-to-noise ratio, field of view, scanning order, scanning slice thickness, scanning slice spacing, scanning time and artifacts related to the sequence images can be used as references of magnetic resonance image data or as image attributes (labels or features) to participate in acquisition or construction of a dataset.
Referring to fig. 3, in step S200 of some embodiments of the present invention, the extracting, based on the transducer network, time-series features and non-time-series features of data of the cooling system, operation data of the gradient coil, gas density data between the magnetic resonance image and the corresponding scan, respectively, includes: s201, extracting non-sequence characteristics of data and gas density data of a cooling system through a first transducer network; s202, extracting operation data of the gradient coil and time sequence characteristics of the magnetic resonance image through a second transducer network.
Further, the non-sequential features of extracting data of the cooling system and gas density data via the first transducer network include: s2011, respectively extracting data of a cooling system and gas density data by using a convolutional neural network to obtain a first feature vector and a second feature vector; s2012, fusing the first characteristic vector and the second characteristic vector by using a transducer network through the attention head and preset weighting parameters to obtain the non-sequence characteristics of the data and the gas density data of the cooling system.
Specifically, the data of each modality is connected to the information of the other modalities (e.g., magnetic resonance image as one of the modality data, data of the connection cooling system, operation data of the gradient coils, and gas density data) through one cross-modality attention layer. This can be achieved by linking the hidden states of all modalities together and feeding this large sequence into a self-attention layer. The outputs of all modalities are then fused together. By direct connection or using more complex methods such as product or weighted average, etc. The fused results may then be passed through one or more fully connected layers for final tasks such as classification, regression, etc.
In step S300 of some embodiments of the present invention, the fusing the non-timing feature and the timing feature through the dual-flow multi-mode model based on the knowledge graph of the operation and maintenance data of the magnetic resonance device includes: s301, obtaining non-time sequence characteristics of magnetic resonance equipment based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment; s302, fusing the non-time sequence characteristics and the time sequence characteristics through a self-distillation method and a double-flow multi-mode model. The knowledge graph of the magnetic resonance apparatus operation and maintenance data can be constructed with reference to the operation and maintenance data, fault data and related validated troubleshooted directed acyclic graphs. As the intrinsic representation of the knowledge graph is added and integrated into the model, the robustness and accuracy of the model are improved.
Referring to fig. 4, fig. 4 shows a two-dimensional MRI room internal physical model. As seen in the upper left portion of FIG. 4, 6000mmX3900mm, 1000mm from the bottom of the MRI room, a 20mm diameter leak orifice with a door on the opposite side from the leak orifice, and a distance of 2200mm from the MRI room floor, the door closing during MRI operation. As seen in the upper right part of fig. 4, there is a waveguide window of 300mm width 1000mm from the gate, i.e. 4.7m from the edge where the leakage opening is located. While the lower part of fig. 4 shows the situation where there are two leakage holes as described above.
In step S500 in the foregoing embodiment, predicting the harmful gas in the target magnetic resonance device using the dual-current multi-mode model and the real-time data of the target magnetic resonance device includes:
s501, predicting the real-time air density between devices where the target magnetic resonance device is located by using the double-flow multi-mode model and the real-time data of the target magnetic resonance device;
s502, predicting the spatial distribution of helium of the target magnetic resonance equipment based on an indoor physical model among the equipment where the target magnetic resonance equipment is located and the real-time air density.
In particular, the spatial distribution may also refer to a hydrodynamic model of the gas (e.g., a diffusion differential equation), etc., to improve the accuracy of the predictions.
Example 2
Referring to fig. 5, in a second aspect of the present invention, there is provided a harmful gas detection system 1 in operation of a magnetic resonance apparatus, comprising: an acquisition module 11 for acquiring data of a cooling system of the target magnetic resonance apparatus, operation data of the gradient coils, magnetic resonance images, and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model; an extraction module 12 for extracting timing features and non-timing features of the data of the cooling system, the operation data of the gradient coils, the magnetic resonance image, and the gas density data between corresponding scans, respectively, based on the transducer network; the fusion module 13 is used for fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model based on a knowledge graph of the operation and maintenance data of the magnetic resonance equipment; a training module 14, configured to train the dual-flow multi-mode model through a training data set based on the fused time sequence feature and non-time sequence feature until an error thereof tends to be stable and lower than a threshold value, to obtain a trained dual-flow multi-mode model; and the prediction module 15 is used for predicting the harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment.
Further, the extracting module 12 includes: a first extraction unit for extracting non-sequential features of the data of the cooling system and the gas density data through a first transducer network;
and the second extraction unit is used for extracting the operation data of the gradient coil and the time sequence characteristics of the magnetic resonance image through a second transducer network.
Example 3
Referring to fig. 6, a third aspect of the present invention provides an electronic device, including: one or more processors; and storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method of the present invention in the first aspect.
The electronic device 500 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with programs stored in a Read Only Memory (ROM) 502 or loaded from a storage 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
The following devices may be connected to the I/O interface 505 in general: input devices 506 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 507 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 508 including, for example, a hard disk; and communication means 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 6 shows an electronic device 500 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead. Each block shown in fig. 6 may represent one device or a plurality of devices as needed.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or from the storage means 508, or from the ROM 502. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 501. It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In an embodiment of the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. Whereas in embodiments of the present disclosure, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device. The computer readable medium carries one or more computer programs which, when executed by the electronic device, cause the electronic device to:
computer program code for carrying out operations of embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++, python and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (10)
1. A method for detecting harmful gases in operation of a magnetic resonance apparatus, comprising:
acquiring data of a cooling system of target magnetic resonance equipment, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model;
based on a transducer network, respectively extracting time sequence characteristics and non-time sequence characteristics of data of a cooling system, operation data of a gradient coil, a magnetic resonance image and gas density data between corresponding scans;
based on a knowledge graph of operation and maintenance data of the magnetic resonance equipment, fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model;
based on the fused time sequence characteristics and non-time sequence characteristics, training the double-flow multi-mode model through a training data set until the error tends to be stable and is lower than a threshold value, and obtaining a trained double-flow multi-mode model;
and predicting the harmful gas of the target magnetic resonance equipment by using the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment.
2. The method of claim 1, wherein constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operational data of the gradient coils, and the gas density data between corresponding scans by the local asperity model comprises: constructing a first training data set based on the data of the cooling system and the gas density data between corresponding scans by means of a local roughness model;
and aligning the first training data set with the data of the cooling system, the magnetic resonance image and the operation data of the gradient coil through a label of preset gas leakage to obtain a second training data set.
3. The method of claim 1, wherein extracting timing and non-timing characteristics of the cooling system data, the gradient coil operational data, the magnetic resonance image and the gas density data between corresponding scans, respectively, based on the transducer network comprises:
extracting non-sequential features of the data of the cooling system and the gas density data through a first transducer network; and extracting the operation data of the gradient coil and the time sequence characteristics of the magnetic resonance image through a second transducer network.
4. A method of detecting a harmful gas during operation of a magnetic resonance apparatus according to claim 3, wherein the non-sequential characterization of the data of the cooling system and the gas density data extracted by the first transducer network comprises:
respectively extracting data of a cooling system and gas density data by using a convolutional neural network to obtain a first feature vector and a second feature vector;
and fusing the first characteristic vector and the second characteristic vector by using a transducer network through the attention head and preset weighting parameters to obtain the non-sequence characteristics of the data of the cooling system and the gas density data.
5. The method for detecting harmful gas during operation of magnetic resonance equipment according to claim 1, wherein the fusing the non-timing features and the timing features through a dual-flow multi-modal model based on a knowledge graph of operational data of the magnetic resonance equipment comprises:
acquiring non-time sequence characteristics of the magnetic resonance equipment based on a knowledge graph of the operation and maintenance data of the magnetic resonance equipment;
and fusing the non-time sequence characteristics and the time sequence characteristics through a self-distillation method and a double-flow multi-mode model.
6. The method of claim 1 to 5, wherein predicting the target magnetic resonance device for harmful gas using the dual-flow multi-modal model and real-time data of the target magnetic resonance device comprises:
predicting the real-time air density between devices where the target magnetic resonance device is located by using the double-flow multi-mode model and the real-time data of the target magnetic resonance device;
and predicting the spatial distribution of helium of the target magnetic resonance equipment based on an indoor physical model among the equipment where the target magnetic resonance equipment is located and the real-time air density.
7. A harmful gas detection system in operation of a magnetic resonance apparatus, comprising:
the acquisition module is used for acquiring data of a cooling system of the target magnetic resonance equipment, operation data of the gradient coil, magnetic resonance images and gas density data between corresponding scans; constructing a training data set based on the data of the cooling system, the magnetic resonance image, the operation data of the gradient coil and the gas density data between corresponding scans through a local rough model;
the extraction module is used for respectively extracting time sequence characteristics and non-time sequence characteristics of the data of the cooling system, the operation data of the gradient coil, the magnetic resonance image and the gas density data between corresponding scans based on the transducer network;
the fusion module is used for fusing the non-time sequence characteristics and the time sequence characteristics through a double-flow multi-mode model based on a knowledge graph of the operation and maintenance data of the magnetic resonance equipment;
the training module is used for training the double-flow multi-mode model through a training data set based on the fused time sequence characteristics and non-time sequence characteristics until the error tends to be stable and is lower than a threshold value, so as to obtain a trained double-flow multi-mode model;
and the prediction module is used for predicting the harmful gas of the target magnetic resonance equipment by utilizing the double-flow multi-mode model and the real-time data of the target magnetic resonance equipment.
8. The harmful gas detection system of claim 7, wherein the extraction module comprises:
a first extraction unit for extracting non-sequential features of the data of the cooling system and the gas density data through a first transducer network;
and the second extraction unit is used for extracting the operation data of the gradient coil and the time sequence characteristics of the magnetic resonance image through a second transducer network.
9. An electronic device, comprising: one or more processors; storage means for storing one or more programs which when executed by the one or more processors cause the one or more processors to implement the method of harmful gas detection in operation of a magnetic resonance apparatus as claimed in any one of claims 1 to 6.
10. A computer readable medium having stored thereon a computer program, wherein the computer program when executed by a processor implements a method of harmful gas detection in operation of a magnetic resonance apparatus as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310923517.7A CN116628457B (en) | 2023-07-26 | 2023-07-26 | Harmful gas detection method and device in operation of magnetic resonance equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310923517.7A CN116628457B (en) | 2023-07-26 | 2023-07-26 | Harmful gas detection method and device in operation of magnetic resonance equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116628457A true CN116628457A (en) | 2023-08-22 |
| CN116628457B CN116628457B (en) | 2023-09-29 |
Family
ID=87610341
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310923517.7A Active CN116628457B (en) | 2023-07-26 | 2023-07-26 | Harmful gas detection method and device in operation of magnetic resonance equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116628457B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117490908A (en) * | 2023-12-31 | 2024-02-02 | 武汉华康世纪医疗股份有限公司 | Negative pressure detection method and system for negative pressure ward |
| CN117496133A (en) * | 2024-01-03 | 2024-02-02 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102004038619A1 (en) * | 2004-08-09 | 2006-02-23 | Siemens Ag | Medical device e.g. magnetic resonance device, operating method, involves calculating required electrical power for selected sequence of operations that is executed by device, and utilizing power to determine operability of operations |
| US20070290685A1 (en) * | 2006-06-12 | 2007-12-20 | Qiang He | Temperature control method for a permanent magnet arrangement of a magnetic resonance system |
| US20080231269A1 (en) * | 2007-03-19 | 2008-09-25 | Masashi Ookawa | Magnetic resonance imaging apparatus, magnetic-resonance imaging maintenance apparatus, magnetic-resonance imaging maintenance system, and magnetic-resonance apparatus inspecting method |
| EP2423699A1 (en) * | 2010-08-30 | 2012-02-29 | Koninklijke Philips Electronics N.V. | Magnetic resonance imaging system, computer system, and computer program product for sending control messages to an anesthesia system |
| CN103091345A (en) * | 2013-01-15 | 2013-05-08 | 湖南省电力公司检修公司 | Nuclear-magnetic-resonance-technology-based method for detecting transformer oil ageing state parameters |
| CN104587493A (en) * | 2015-02-05 | 2015-05-06 | 东南大学 | Detection reagents for fast and real-time dynamic monitoring and multimode imaging in early stage of cardio-cerebrovascular related diseases |
| WO2015101556A1 (en) * | 2014-01-03 | 2015-07-09 | Koninklijke Philips N.V. | Calculation of the probability of gradient coil amplifier failure using environment data |
| CN105891754A (en) * | 2016-03-03 | 2016-08-24 | 哈尔滨医科大学 | Multisource frequency spectrum spectrometer control system for multi-nuclear magnetic resonance |
| CN108187247A (en) * | 2017-12-28 | 2018-06-22 | 中国科学院深圳先进技术研究院 | Guided by magnetic resonance focuses on focusing target spot adjustment system, the method and device of ultrasound |
| US20200278408A1 (en) * | 2019-03-01 | 2020-09-03 | The Regents Of The University Of California | Systems, Methods and Media for Automatically Segmenting and Diagnosing Prostate Lesions Using Multi-Parametric Magnetic Resonance Imaging Data |
| US20200320704A1 (en) * | 2018-07-24 | 2020-10-08 | Shenzhen Institutes Of Advanced Technology | Method and device of processing plaques in magnetic resonance imaging of vessel wall, and computer device |
| WO2020215985A1 (en) * | 2019-04-22 | 2020-10-29 | 腾讯科技（深圳）有限公司 | Medical image segmentation method and device, electronic device and storage medium |
| US20200364500A1 (en) * | 2019-05-13 | 2020-11-19 | Shanghai Neusoft Medical Technology Co., Ltd. | Training image enhancement model and enhancing image |
| WO2021119875A1 (en) * | 2019-12-15 | 2021-06-24 | 中国科学院深圳先进技术研究院 | Fast magnetic resonance imaging method and apparatus based on neural architecture search |
| DE102020207363A1 (en) * | 2020-06-15 | 2021-12-16 | Siemens Healthcare Gmbh | Prediction of a possible failure of a module for use in a magnetic resonance device |
| WO2022006917A1 (en) * | 2019-07-05 | 2022-01-13 | 深圳市安测健康信息技术有限公司 | Artificial intelligence-based lung magnetic resonance image recognition apparatus and method |
| CA3159995A1 (en) * | 2021-08-17 | 2022-09-27 | Beijing Friendship Hospital, Capital Medical University | Cerebral perfusion state classification apparatus and method, device, and storage medium |
| US20220381857A1 (en) * | 2021-05-26 | 2022-12-01 | Canon Medical Systems Corporation | Magnetic resonance imaging system, magnetic resonance imaging apparatus, cooling control device, and cooling control method |
-
2023
- 2023-07-26 CN CN202310923517.7A patent/CN116628457B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102004038619A1 (en) * | 2004-08-09 | 2006-02-23 | Siemens Ag | Medical device e.g. magnetic resonance device, operating method, involves calculating required electrical power for selected sequence of operations that is executed by device, and utilizing power to determine operability of operations |
| US20070290685A1 (en) * | 2006-06-12 | 2007-12-20 | Qiang He | Temperature control method for a permanent magnet arrangement of a magnetic resonance system |
| US20080231269A1 (en) * | 2007-03-19 | 2008-09-25 | Masashi Ookawa | Magnetic resonance imaging apparatus, magnetic-resonance imaging maintenance apparatus, magnetic-resonance imaging maintenance system, and magnetic-resonance apparatus inspecting method |
| EP2423699A1 (en) * | 2010-08-30 | 2012-02-29 | Koninklijke Philips Electronics N.V. | Magnetic resonance imaging system, computer system, and computer program product for sending control messages to an anesthesia system |
| CN103091345A (en) * | 2013-01-15 | 2013-05-08 | 湖南省电力公司检修公司 | Nuclear-magnetic-resonance-technology-based method for detecting transformer oil ageing state parameters |
| WO2015101556A1 (en) * | 2014-01-03 | 2015-07-09 | Koninklijke Philips N.V. | Calculation of the probability of gradient coil amplifier failure using environment data |
| CN105874345A (en) * | 2014-01-03 | 2016-08-17 | 皇家飞利浦有限公司 | Calculation of the probability of gradient coil amplifier failure using environment data |
| CN104587493A (en) * | 2015-02-05 | 2015-05-06 | 东南大学 | Detection reagents for fast and real-time dynamic monitoring and multimode imaging in early stage of cardio-cerebrovascular related diseases |
| CN105891754A (en) * | 2016-03-03 | 2016-08-24 | 哈尔滨医科大学 | Multisource frequency spectrum spectrometer control system for multi-nuclear magnetic resonance |
| CN108187247A (en) * | 2017-12-28 | 2018-06-22 | 中国科学院深圳先进技术研究院 | Guided by magnetic resonance focuses on focusing target spot adjustment system, the method and device of ultrasound |
| US20200320704A1 (en) * | 2018-07-24 | 2020-10-08 | Shenzhen Institutes Of Advanced Technology | Method and device of processing plaques in magnetic resonance imaging of vessel wall, and computer device |
| US20200278408A1 (en) * | 2019-03-01 | 2020-09-03 | The Regents Of The University Of California | Systems, Methods and Media for Automatically Segmenting and Diagnosing Prostate Lesions Using Multi-Parametric Magnetic Resonance Imaging Data |
| WO2020215985A1 (en) * | 2019-04-22 | 2020-10-29 | 腾讯科技（深圳）有限公司 | Medical image segmentation method and device, electronic device and storage medium |
| US20200364500A1 (en) * | 2019-05-13 | 2020-11-19 | Shanghai Neusoft Medical Technology Co., Ltd. | Training image enhancement model and enhancing image |
| WO2022006917A1 (en) * | 2019-07-05 | 2022-01-13 | 深圳市安测健康信息技术有限公司 | Artificial intelligence-based lung magnetic resonance image recognition apparatus and method |
| WO2021119875A1 (en) * | 2019-12-15 | 2021-06-24 | 中国科学院深圳先进技术研究院 | Fast magnetic resonance imaging method and apparatus based on neural architecture search |
| DE102020207363A1 (en) * | 2020-06-15 | 2021-12-16 | Siemens Healthcare Gmbh | Prediction of a possible failure of a module for use in a magnetic resonance device |
| US20220381857A1 (en) * | 2021-05-26 | 2022-12-01 | Canon Medical Systems Corporation | Magnetic resonance imaging system, magnetic resonance imaging apparatus, cooling control device, and cooling control method |
| CA3159995A1 (en) * | 2021-08-17 | 2022-09-27 | Beijing Friendship Hospital, Capital Medical University | Cerebral perfusion state classification apparatus and method, device, and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 张斌等: "结合磁共振成像和脑机接口的新型在体生物电子鼻的研究", 《中国生物医学工程学报》, vol. 37, no. 01 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117490908A (en) * | 2023-12-31 | 2024-02-02 | 武汉华康世纪医疗股份有限公司 | Negative pressure detection method and system for negative pressure ward |
| CN117490908B (en) * | 2023-12-31 | 2024-04-09 | 武汉华康世纪医疗股份有限公司 | Negative pressure detection method and system for negative pressure ward |
| CN117496133A (en) * | 2024-01-03 | 2024-02-02 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
| CN117496133B (en) * | 2024-01-03 | 2024-03-22 | 山东工商学院 | Closed bus R-CNN temperature fault monitoring method based on multi-mode data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116628457B (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116628457B (en) | Harmful gas detection method and device in operation of magnetic resonance equipment | |
| Amland et al. | Clinical decision support for early recognition of sepsis | |
| Amland et al. | Quick Sequential [Sepsis-Related] Organ Failure Assessment (qSOFA) and St. John Sepsis Surveillance Agent to detect patients at risk of sepsis: an observational cohort study | |
| Birkhoff et al. | A review on the current applications of artificial intelligence in the operating room | |
| CN105593680B (en) | Cooperation Electronic Nose management in personal device | |
| Yang et al. | Healthcare intelligence: turning data into knowledge | |
| Rahmani et al. | Early prediction of central line associated bloodstream infection using machine learning | |
| US11227684B2 (en) | Systems and methods for processing electronic images for health monitoring and forecasting | |
| Luo et al. | Self context and shape prior for sensorless freehand 3D ultrasound reconstruction | |
| Radke et al. | Deep learning-based post-processing of real-time MRI to assess and quantify dynamic wrist movement in health and disease | |
| Chu et al. | MRI-based radiomics analysis for intraoperative risk assessment in gravid patients at high risk with placenta accreta spectrum | |
| Caroli et al. | Abdominal imaging in ADPKD: beyond total kidney volume | |
| Xu et al. | Automatic detection of pulmonary embolism in computed tomography pulmonary angiography using Scaled‐YOLOv4 | |
| Zhang et al. | MRI-based radiomics models to discriminate Hepatocellular Carcinoma and Non-Hepatocellular Carcinoma in LR-M According to LI-RADS Version 2018 | |
| van Zyl et al. | A pilot study of a palliative care service embedded in a hepatology clinic at a large public hospital | |
| Grazzini et al. | Local Recurrences in Rectal Cancer: MRI vs. CT | |
| Brown et al. | A multifactorial severity score for left congenital diaphragmatic hernia in a high-risk population using fetal magnetic resonance imaging | |
| Do et al. | Creating Computer Vision Models for Respiratory Status Detection | |
| Jiang et al. | Application and Feasibility Study of Integrated Nursing Information Construction in Nephrology Nursing | |
| US11170889B2 (en) | Smooth image scrolling | |
| Podgórska et al. | Intravoxel incoherent motion MRI in evaluating inflammatory activity in ulcerative colitis: a pilot study | |
| Baskaran et al. | Using facial landmark detection on thermal images as a novel prognostic tool for emergency departments | |
| Zhang et al. | 7T MRI for intracranial vessel wall lesions and its associated neurological disorders: a systematic review | |
| Rockenschaub et al. | From Single-Hospital to Multi-Centre Applications: Enhancing the Generalisability of Deep Learning Models for Adverse Event Prediction in the ICU | |
| Jiang et al. | Timeline registration for electronic health records |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |