CN108492874A - The critical value diagnosis and therapy system of intelligence - Google Patents
The critical value diagnosis and therapy system of intelligence Download PDFInfo
- Publication number
- CN108492874A CN108492874A CN201810404112.1A CN201810404112A CN108492874A CN 108492874 A CN108492874 A CN 108492874A CN 201810404112 A CN201810404112 A CN 201810404112A CN 108492874 A CN108492874 A CN 108492874A
- Authority
- CN
- China
- Prior art keywords
- critical value
- library
- positive
- data
- negative
- 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
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Pathology (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention provides a kind of critical value diagnosis and therapy systems of intelligence, including critical value screening module and critical value transfer module, wherein：Critical value screening module establishes critical Value Data library, and the critical Value Data library is trained and is verified, and obtains critical value-based algorithm；Critical Value Data library is divided into critical value positive data library and the negative database of critical value by critical value screening module, and critical value positive data library is randomly divided into positive training library and positive verification library by a certain percentage, the negative database of critical value is randomly divided into negative training library and negative verification library by a certain percentage, critical value-based algorithm is trained using positive training library and negative training library, critical value-based algorithm is verified using positive verification library and negative verification library；It checks whether the data of radiology image and radiology report have critical illness by the critical value-based algorithm, if having critical illness, and starts warning device after being verified, and clinician is sent to by the critical value transfer module.
Description
Technical field
The present invention relates to field of artificial intelligence, more particularly to a kind of critical value diagnosis and therapy system of intelligence.
Background technology
Country, which defends, requires emergency treatment report time to be accurate to point in planning commission tertiary hospitals evaluation detailed rules and regulations, Chinese doctors'associations patient
Security target is to strengthen clinical critical value reporting system, and certification joint committee of International Medical health organ requires to establish effective
Critical value reporting system.It can be that the quality time is won in medical act that critical value, which effectively finds and transmits, in medical image, reduce
The death rate and disability rate.The screening of critical value relies on doctor's judgement in medical image at present, and the transmission of critical value is by artificial
Phone and record.
The diagnoses and treatment as early as possible of critical disease can obviously reduce the death rate, improve prognosis.Pulmonary embolism, aorta clamp
The critical disease such as layer, pressure pneumothorax can rapid contribution cardio-pulmonary function and lethality is high, dept. of radiology check in these three medicals diagnosis on disease
In play a key effect.Pulmonary embolism is one of clinical common acute disease, and incidence may be up to 70/,100,000, nearly 25% it is acute
Pulmonary embolism can die suddenly, and the pulmonary embolism death rate of untreated is up to 60%, if can be diagnosed in time and correct
Treatment case fatality rate can drop to 2-8%.Dissection of aorta annual morbidity about 2.6-3.5/10 ten thousand, the death rate reaches in 24 hours
25%, the death rate is up to 50% in one week；Early diagnosis, which is intervened, can obviously reduce the death rate.The incidence of spontaneous pneumothorax is in 2.5-
18/100000, pneumothorax or pressure pneumothorax are formed in original underlying lung diseases such as pulmonary emphysema, general symptom is heavier, influences the heart
Lung function is apparent, endangers larger, the death rate about 4.7%.These three critical sick diverse clinical manifestations, lack specificity, radiation
It is the most important evidence diagnosed that section, which checks, and diagnosis as early as possible can be that treatment gain time.
Critical disease delays the diagnosis and treatment as physician-patient relationship tense major reason.The first sentence in 2013 of Beijing municipal court conclude 1044
In part case, emergency department accounts for 8.3%, occupies the 4th.China of Beijing material evidence evaluating center was from January, 2010 to 2014 10
Accept Medical Dispute Appraisal 500 moon altogether, orthopaedics case section office occurred frequently are the first, wherein because pulmonary artery occur in fracture or fracture surgery
Embolism and death large percentage.It is counted in being summarized according to one of New England Journal of Medicine：Emergency department and essence in United States Hospital
Refreshing ward is the place that occurs most frequently of violence wound doctor, and Nurses of Emergency Department is abused in the interviewed prior year 100%, 82.1% by
Personal attack, about 1/4 emergency department doctor become the target of workplace violence in last year.Report delay has been the U.S.
The main reason for dept. of radiology is by lawsuit, single case are judged to claimable amount up to 1,900,000 U.S. dollars.To the critical value in dept. of radiology's picture and text
Detection will reduce medical-risk.
In order to protect patient safety and improve quality of medical care, the clinical procedure guide of domestic and international associated mechanisms, which has, clearly to be radiated
The requirement of the critical value reporting system of section.Certification joint committee of International Medical health organ determines danger in patient safety target
The report validity being suddenly worth.Certification joint committee of synthesis International Medical health organ of Univ Massachusetts Medical, the U.S. are put
The specification for penetrating association, Massachusetts malpractice prevention alliance, has formulated the critical value reporting system of dept. of radiology, has been divided into three-level：(1) red
Alarm, it is necessary to be reported in 1 hour, including airway obstruction, new pneumothorax, the unexpected free gas in abdominal cavity for occurring or being obviously in progress
Pipeline position is supported in body, aneurysm rupture, artery dissection, acute spinal compressing, intestinal obstruction, pulmonary embolism, medical treatment detection
Abnormal, ectopic pregnancy；(2) Amber Alert need to be reported in 8 hours, including doubtful child abuse, unexpected lump, new life
Three degree of hydrocephalus of youngster, doubtful active tuberculosis；(3) warning yellow needs to report in 24 hours, including preoperative planning finds lung knot
Section, single-shot kidney lump.European Society of Radiology proposes guide to the effective communication of the critical value of dept. of radiology, from Systematic and implementation
Level gives feasible method；They emphasize that the clinical supervisor that is sent to that dept. of radiology has a responsibility for ensuring to report promptly and accurately cures
It is raw.The tertiary hospitals of China's health and Family Planning Committee are evaluated detailed rules for the implementation (2011 editions) traditional Chinese medicine Image Management and are continued
Improving in chapters and sections requires the timely specification of Medical imaging diagnostic reports, emergency treatment report time to be accurate to minute.Chinese Hospital Association
Target six is to strengthen clinical critical value reporting system, specification inspection, image, electrocardiogram in patient safety target 2014-2015 editions
In critical value effectively report.Each mechanism especially emphasizes the timeliness and validity of critical value report executing, but for radiation
The method that University of Science and Technology's data deficiency science embodies, it is difficult to meet the state of an illness demand of individual.
Traditional natural language processing techniques use syntactic analysis and phrase structure rule.American-European countries researcher is to dept. of radiology
The research that report carries out natural language processing is more, but there is not yet the correlative study that Chinese natural language is handled.The scholars such as Ewoud
Summarize five big applications of the natural language processing in medical image report analysis：Diagnostic monitoring, epidemiological study queue are built
Vertical, case return visit, quality control, clinical information support etc..Diagnostic result in report is monitored in real time, is conducive in time
Clinical supervisor doctor is reminded to carry out clinical decision.Current diagnosis monitoring concentrates on monitoring appendicitis, acute lung injury, pneumonia, blood
Bolt embolism class diseases, malignant tumour etc., which part monitor for critical value.Lakhani research teams using it is rule-based from
Image report of the right language processing techniques to 9 kinds of critical illness such as pneumothorax, acute pulmonary embolism, acute appendicitis, ectopic pregnancy
It accuses text to be detected, accuracy is in 81%-100%.And it carries out natural language processing using deep learning and is based on corpus
Construction and statistical theory application, will further improve accuracy.Nearly 10 years with the continuous propulsion of medical information, radiation
Section's data volume constantly increases, and providing data for the computer-aided diagnosis system based on deep learning supports.The accumulation of data,
The support of hardware, the breakthrough of algorithm provide sufficient and necessary condition for critical value intelligent information excavation in dept. of radiology's picture and text.
Invention content
The purpose of the present invention is to provide a kind of critical value diagnosis and therapy systems of intelligence, are transmitted with solving the critical value of existing disease
The busy problem of speed.
In order to solve the above technical problems, the present invention provides a kind of critical value diagnosis and therapy system of intelligence, the critical value of intelligence is examined
Treatment system includes critical value screening module and critical value transfer module, wherein：
The critical value screening module establishes critical Value Data library, and the critical Value Data library is trained and is tested
Card, obtains critical value-based algorithm；
Critical Value Data library is divided into critical value positive data library and the negative data of critical value by the critical value screening module
Library, and critical value positive data library is randomly divided into positive training library and positive verification library by a certain percentage, by a certain percentage will
The negative database of critical value is randomly divided into negative training library and negative verification library, is trained using positive training library and negative training library
Critical value-based algorithm verifies critical value-based algorithm using positive verification library and negative verification library；
Check whether the data of radiology image and radiology report have critical illness by the critical value-based algorithm, if
With critical illness, then and after being verified start warning device, and clinic is sent to by the critical value transfer module
Doctor.
Optionally, in the critical value diagnosis and therapy system of intelligence, the critical value includes the critical value of pulmonary embolism, master
The critical value of artery dissection and the critical value of pneumothorax.
Optionally, in the critical value diagnosis and therapy system of intelligence, the critical value screening module includes deep learning figure
As identification module and natural language processing module, wherein：
The deep learning picture recognition module carries out the critical value information of image to radiology image and extracts, the nature language
Say that processing module carries out the critical value information of word to radiology report and extracts；
The critical Value Data library is trained and is verified by the critical value information of image and the critical value information of word, is obtained
To critical value-based algorithm.
Optionally, in the critical value diagnosis and therapy system of intelligence, before establishing critical Value Data library, the critical value sieve
It looks into module and removes privacy information in radiology image and radiology report.
Optionally, in the critical value diagnosis and therapy system of intelligence, the critical value positive data library includes dynamic with lung
The radiology image and radiology report of arteries and veins embolism, dissection of aorta and pneumothorax situation, the negative database of critical value include
Radiology image and radiology report without pulmonary embolism, dissection of aorta and pneumothorax situation.
Optionally, in the critical value diagnosis and therapy system of intelligence, positive training library and negative training library training danger are utilized
Before anxious value-based algorithm, the critical value-based algorithm is built using the positive training library and the negative training library.
Optionally, in the critical value diagnosis and therapy system of intelligence, the critical value screening module is by recycling nerve net
Network and convolutional neural networks build the critical value-based algorithm.
Optionally, in the critical value diagnosis and therapy system of intelligence, the positive verification library and the negative verification are utilized
Before the critical value-based algorithm is examined in library, the data that doctor verifies in the positive verification library and the negative verification library are correct
's.
Optionally, in the critical value diagnosis and therapy system of intelligence, the ratio is 9:1.
It optionally, can to the highest feature progress of the critical value-based algorithm in the critical value diagnosis and therapy system of intelligence
It is handled depending on change, to carry out visualization processing to the critical value of the radiology image.
Optionally, in the critical value diagnosis and therapy system of intelligence, critical value positive data library and critical value the moon are being established
Property database before, the data in critical Value Data library are manually marked, positive data and negative data are divided into, will be negative
Data are put into the negative database of critical value, and the illness of positive data is further carried out automatic marking, and by manual verification come
Confirm whether automatic marking is correct, and positive data is then put into critical value positive data library.
In the critical value diagnosis and therapy system of intelligence provided by the invention, critical value screening module generates in medical image and uploads clothes
Screening is carried out to it at the first time after business device, finds suspicious critical value, responsibility dept. of radiology doctor is notified by Internet technology at once
Teacher, radiologist notify responsibility clinician by development of Mobile Internet technology at once after confirming, realize the critical value of disease
Quickly confirm and transmit, gains time for treatment.
In addition, the critical value-based algorithm excavated by establishing the critical value information of dept. of radiology's picture and text based on deep learning, real
Now following effect：Pulmonary embolism, dissection of aorta, pneumothorax intelligent recognition accuracy rate 99.9% or so in radiology report, it is false
Negative rate is less than 0.1%.For pulmonary embolism digital image recognition accuracy rate 93% or so, false negative rate is less than 3%.Aorta
For interlayer digital image recognition accuracy rate 95% or so, false negative rate is less than 1%.Pneumothorax intelligent recognition accuracy rate is on 98% left side
The right side, false negative rate are less than 1%.
Description of the drawings
Fig. 1 is one embodiment of the invention intelligently critical value diagnosis and therapy system schematic diagram.
Specific implementation mode
Make below in conjunction with the drawings and specific embodiments value diagnosis and therapy system critical to intelligence proposed by the present invention further detailed
Explanation.According to following explanation and claims, advantages and features of the invention will become apparent from.It should be noted that attached drawing is adopted
Use with very simplified form and non-accurate ratio, only to it is convenient, lucidly aid in illustrating the embodiment of the present invention
Purpose.
Core of the invention thought is to provide a kind of critical value diagnosis and therapy system of intelligence, to solve the critical value of existing disease
The busy problem of transmission speed.
To realize above-mentioned thought, the present invention provides a kind of critical value diagnosis and therapy system of intelligence, the critical value screening module
Critical Value Data library is established, and the critical Value Data library is trained and is verified, obtains critical value-based algorithm；The critical value
Critical Value Data library is divided into critical value positive data library and the negative database of critical value by screening module, and by a certain percentage will danger
Anxious value positive data library is randomly divided into positive training library and positive verification library, by a certain percentage that the negative database of critical value is random
It is divided into negative training library and negative verification library, trains critical value-based algorithm using positive training library and negative training library, utilize the positive
Verification library and negative verification library verify critical value-based algorithm；Radiology image and radiology report are checked by the critical value-based algorithm
Data whether there is critical illness, if having critical illness, and start warning device after being verified, and described in
Critical value transfer module is sent to clinician.
The present embodiment provides a kind of critical value diagnosis and therapy systems of intelligence, as shown in Figure 1, the critical value diagnosis and therapy system packet of the intelligence
Critical value screening module and critical value transfer module are included, wherein：The critical value screening module establishes critical Value Data library, and right
The critical Value Data library is trained and verifies, and obtains critical value-based algorithm；The critical value screening module is by critical Value Data
Library is divided into critical value positive data library and the negative database of critical value, and (ratio is 9 by a certain percentage:1) by critical value
Positive data library is randomly divided into positive training library and positive verification library, and (ratio is 9 by a certain percentage:1) by critical value the moon
Property database be randomly divided into negative training library and negative verification library, train critical value calculation using positive training library and negative training library
Method verifies critical value-based algorithm using positive verification library and negative verification library；Radiology image is checked by the critical value-based algorithm
Whether there is critical illness with the data of radiology report, if having critical illness, and starts alarm dress after being verified
It sets, and clinician is sent to by the critical value transfer module.
The critical value diagnosis and therapy system of the intelligence can also include a critical value setup module, the critical value setup module
For being arranged to identify the keyword of the critical value of medicine, it is high that the critical value of medicine refers to a certain or certain class medical inspection result
Degree is abnormal, and patient may be in the rim condition being in peril of one's life, for example, the critical value include the critical value of pulmonary embolism,
The critical value of dissection of aorta and the critical value of pneumothorax.Alternatively, when the critical value diagnosis and therapy system of intelligence is used for different department, different department
The corresponding different keyword.
Specifically, in the critical value diagnosis and therapy system of intelligence, the critical value screening module includes deep learning figure
As identification module and natural language processing module, wherein：The deep learning picture recognition module carries out figure to radiology image
As the extraction of critical value information, the natural language processing module carries out the critical value information of word to radiology report and extracts, to report
The keyword occurred in announcement is identified, and recognition result is extracted；Pass through the critical value information of image and the critical value of word
Information is trained and verifies to the critical Value Data library, obtains critical value-based algorithm.
In addition to the privacy of protection patient is establishing critical Value Data in the critical value diagnosis and therapy system of intelligence
Before library, the critical value screening module removes the privacy information in radiology image and radiology report, such as patient's name, trouble
Person's gender, patient age, application section office etc..
Further, in the critical value diagnosis and therapy system of intelligence, the critical value positive data library includes having lung
The radiology image and radiology report of arterial embolism, dissection of aorta and pneumothorax situation, the negative database packet of critical value
Include radiology image and radiology report without pulmonary embolism, dissection of aorta and pneumothorax situation.Utilize positive training
Before critical value-based algorithm is trained in library and negative training library, built using the positive training library and the negative training library described critical
Value-based algorithm.The critical value screening module builds the critical value-based algorithm by Recognition with Recurrent Neural Network and convolutional neural networks.Profit
Before examining the critical value-based algorithm with the positive verification library and the negative verification library, doctor verify the positive verification library and
Data in the feminine gender verification library are correct.
In addition, in the critical value diagnosis and therapy system of intelligence, the highest feature of the critical value-based algorithm is carried out visual
Change is handled, to carry out visualization processing to the critical value of the radiology image.Establishing critical value positive data library and critical
Before the negative database of value, the data in critical Value Data library are manually marked, are divided into positive data and negative data, it will
Negative data are put into the negative database of critical value, the illness of positive data are further carried out automatic marking, and by manually testing
It demonstrate,proves to confirm whether automatic marking is correct, positive data is then put into critical value positive data library.
In the critical value diagnosis and therapy system of intelligence provided by the invention, critical value screening module generates in medical image and uploads clothes
Screening is carried out to it at the first time after business device, finds suspicious critical value, responsibility dept. of radiology doctor is notified by Internet technology at once
Teacher, radiologist notify responsibility clinician by development of Mobile Internet technology at once after confirming, realize the critical value of disease
Quickly confirm and transmit, gains time for treatment.
In addition, the critical value-based algorithm excavated by establishing the critical value information of dept. of radiology's picture and text based on deep learning, real
Now following effect：Pulmonary embolism, dissection of aorta, pneumothorax intelligent recognition accuracy rate 99.9% or so in radiology report, it is false
Negative rate is less than 0.1%.For pulmonary embolism digital image recognition accuracy rate 93% or so, false negative rate is less than 3%.Aorta
For interlayer digital image recognition accuracy rate 95% or so, false negative rate is less than 1%.Pneumothorax intelligent recognition accuracy rate is on 98% left side
The right side, false negative rate are less than 1%.
As shown in Figure 1, for the critical value diagnosis and therapy system of intelligence, some radiology images and report are obtained first, and
Privacy information is removed, these images and report are put into newly-established critical Value Data library, doctor is by manual sort, by image
It is divided into positive data and negative data with report, negative data is directly put into negative database, to positive data into advancing one
The mark of step marks its specific illness, and is then marked automatically by manual verification again with critical value-based algorithm automatic marking again
Whether note is accurate, to optimize critical value-based algorithm, optimizes illness mark completion of the verification up to all positive data repeatedly with this, will
It is put into positive data library.The data in positive data library are divided into 10 equal portions, a data composition positive verification library, nine parts of data
The data of negative database are divided into 10 equal portions, a negative verification library of data composition, nine parts of data groups by the positive training library of composition
At feminine gender training library.The data in positive training library and negative training library are merged, are trained using the data in two trained libraries critical
Value-based algorithm makes critical value-based algorithm carry out deep learning, until its own can judge critical value illness, at this time by positive verification library and
The data of negative verification library merge, and whether the data mark of the artificial nuclear tests card database of doctor is correct, if correctly, it will verification
Library data issue critical value-based algorithm, and make critical value-based algorithm that verification library data are carried out with the diagnosis of critical value illness, to judge to calculate
Whether method reaches requirement, is advanced optimized if it cannot reach requirement, furthermore with visualization technique, by examining for critical value illness
It is disconnected to carry out visualization processing, keep diagnostic result more very clear.
To sum up, the various configuration of the critical value diagnosis and therapy system of intelligence is described in detail in above-described embodiment, certainly, this hair
Cited configuration in bright including but not limited to above-mentioned implementation, it is any to be become on the basis of the configuration of above-described embodiment offer
The content changed belongs to the range that the present invention is protected.Those skilled in the art can lift one according to the content of above-described embodiment
Anti- three.
Foregoing description is only the description to present pre-ferred embodiments, not to any restriction of the scope of the invention, this hair
Any change, the modification that the those of ordinary skill in bright field does according to the disclosure above content, belong to the protection of claims
Range.
Claims (11)
1. a kind of critical value diagnosis and therapy system of intelligence, which is characterized in that the critical value diagnosis and therapy system of intelligence includes critical value screening
Module and critical value transfer module, wherein：
The critical value screening module establishes critical Value Data library, and the critical Value Data library is trained and is verified, and obtains
To critical value-based algorithm；
Critical Value Data library is divided into critical value positive data library and the negative database of critical value by the critical value screening module, and
Critical value positive data library is randomly divided into positive training library and positive verification library by a certain percentage, by a certain percentage by critical value
Negative database is randomly divided into negative training library and negative verification library, and critical value is trained using positive training library and negative training library
Algorithm verifies critical value-based algorithm using positive verification library and negative verification library；
Check whether the data of radiology image and radiology report have critical illness by the critical value-based algorithm, if having
Critical illness then and after being verified starts warning device, and is sent to clinician by the critical value transfer module.
2. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that the critical value includes pulmonary embolism
The critical value of critical value, dissection of aorta and the critical value of pneumothorax.
3. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that the critical value screening module includes deep
Degree study picture recognition module and natural language processing module, wherein：
The deep learning picture recognition module carries out the critical value information of image to radiology image and extracts, at the natural language
It manages module and the critical value information extraction of word is carried out to radiology report；
The critical Value Data library is trained and is verified by the critical value information of image and the critical value information of word, is endangered
Anxious value-based algorithm.
4. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that before establishing critical Value Data library, institute
It states critical value screening module and removes privacy information in radiology image and radiology report.
5. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that the critical value positive data library includes
Radiology image and radiology report with pulmonary embolism, dissection of aorta and pneumothorax situation, the negative number of critical value
Include radiology image and radiology report without pulmonary embolism, dissection of aorta and pneumothorax situation according to library.
6. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that trained using positive training library and feminine gender
Before critical value-based algorithm is trained in library, the critical value-based algorithm is built using the positive training library and the negative training library.
7. the critical value diagnosis and therapy system of intelligence as claimed in claim 6, which is characterized in that the critical value screening module is by following
Ring neural network and convolutional neural networks build the critical value-based algorithm.
8. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that utilize the positive verification library and described
Before negative verification library examines the critical value-based algorithm, doctor verifies the data in the positive verification library and the negative verification library
It is correct.
9. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that the ratio is 9:1.
10. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that the highest of the critical value-based algorithm
Feature carries out visualization processing, to carry out visualization processing to the critical value of the radiology image.
11. the critical value diagnosis and therapy system of intelligence as described in claim 1, which is characterized in that establishing critical value positive data library
Before the negative database of critical value, the data in critical Value Data library are manually marked, positive data and feminine gender are divided into
Negative data are put into the negative database of critical value, the illness of positive data are further carried out automatic marking, and pass through by data
Manual verification confirms whether automatic marking is correct, then positive data is put into critical value positive data library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810404112.1A CN108492874A (en) | 2018-04-28 | 2018-04-28 | The critical value diagnosis and therapy system of intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810404112.1A CN108492874A (en) | 2018-04-28 | 2018-04-28 | The critical value diagnosis and therapy system of intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108492874A true CN108492874A (en) | 2018-09-04 |
Family
ID=63314312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810404112.1A Pending CN108492874A (en) | 2018-04-28 | 2018-04-28 | The critical value diagnosis and therapy system of intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108492874A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111755108A (en) * | 2020-05-11 | 2020-10-09 | 深圳市罗湖区人民医院 | Method for treating and managing patients based on critical values |
| CN112259197A (en) * | 2020-10-14 | 2021-01-22 | 北京赛迈特锐医疗科技有限公司 | Intelligent analysis system and method for acute abdomen plain film |
| CN114255859A (en) * | 2021-12-23 | 2022-03-29 | 四川大学华西医院 | Method and system for sorting emergency values of radiology department emergency examination reports |
| US11426119B2 (en) | 2020-04-10 | 2022-08-30 | Warsaw Orthopedic, Inc. | Assessment of spinal column integrity |
| US11450435B2 (en) | 2020-04-07 | 2022-09-20 | Mazor Robotics Ltd. | Spinal stenosis detection and generation of spinal decompression plan |
| US11471118B2 (en) | 2020-03-27 | 2022-10-18 | Hologic, Inc. | System and method for tracking x-ray tube focal spot position |
| US11481038B2 (en) | 2020-03-27 | 2022-10-25 | Hologic, Inc. | Gesture recognition in controlling medical hardware or software |
| US11510306B2 (en) | 2019-12-05 | 2022-11-22 | Hologic, Inc. | Systems and methods for improved x-ray tube life |
| US11694792B2 (en) | 2019-09-27 | 2023-07-04 | Hologic, Inc. | AI system for predicting reading time and reading complexity for reviewing 2D/3D breast images |
| US11883206B2 (en) | 2019-07-29 | 2024-01-30 | Hologic, Inc. | Personalized breast imaging system |
| US12414217B2 (en) | 2022-02-07 | 2025-09-09 | Hologic, Inc. | Systems and methods for adaptively controlling filament current in an X-ray tube |
| US12433678B2 (en) | 2022-10-10 | 2025-10-07 | Warsaw Orthopedic, Inc. | Devices, methods, and systems for assessing suitability of spinal implants |
| US12530860B2 (en) | 2020-11-20 | 2026-01-20 | Hologic, Inc. | Systems and methods for using AI to identify regions of interest in medical images |
| US12548676B2 (en) | 2022-09-07 | 2026-02-10 | Mazor Robotics Ltd. | Spinal stenosis detection and generation of spinal decompression plan |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992070A (en) * | 2015-07-18 | 2015-10-21 | 李建 | Medical imaging examination critical value automatic identifying and reporting system |
| CN105160361A (en) * | 2015-09-30 | 2015-12-16 | 东软集团股份有限公司 | Image identification method and apparatus |
| CN107085665A (en) * | 2017-05-24 | 2017-08-22 | 洛阳瀍盛科技有限公司 | Health Service Laboratory survey report intellectual analysis and the automatic transmission system of critical value |
| CN107220518A (en) * | 2017-06-28 | 2017-09-29 | 湖南暄程科技有限公司 | A kind of medical pre-alert notification method and system |
-
2018
- 2018-04-28 CN CN201810404112.1A patent/CN108492874A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992070A (en) * | 2015-07-18 | 2015-10-21 | 李建 | Medical imaging examination critical value automatic identifying and reporting system |
| CN105160361A (en) * | 2015-09-30 | 2015-12-16 | 东软集团股份有限公司 | Image identification method and apparatus |
| CN107085665A (en) * | 2017-05-24 | 2017-08-22 | 洛阳瀍盛科技有限公司 | Health Service Laboratory survey report intellectual analysis and the automatic transmission system of critical value |
| CN107220518A (en) * | 2017-06-28 | 2017-09-29 | 湖南暄程科技有限公司 | A kind of medical pre-alert notification method and system |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12226233B2 (en) | 2019-07-29 | 2025-02-18 | Hologic, Inc. | Personalized breast imaging system |
| US11883206B2 (en) | 2019-07-29 | 2024-01-30 | Hologic, Inc. | Personalized breast imaging system |
| US12119107B2 (en) | 2019-09-27 | 2024-10-15 | Hologic, Inc. | AI system for predicting reading time and reading complexity for reviewing 2D/3D breast images |
| US11694792B2 (en) | 2019-09-27 | 2023-07-04 | Hologic, Inc. | AI system for predicting reading time and reading complexity for reviewing 2D/3D breast images |
| US11510306B2 (en) | 2019-12-05 | 2022-11-22 | Hologic, Inc. | Systems and methods for improved x-ray tube life |
| US11471118B2 (en) | 2020-03-27 | 2022-10-18 | Hologic, Inc. | System and method for tracking x-ray tube focal spot position |
| US11481038B2 (en) | 2020-03-27 | 2022-10-25 | Hologic, Inc. | Gesture recognition in controlling medical hardware or software |
| US11450435B2 (en) | 2020-04-07 | 2022-09-20 | Mazor Robotics Ltd. | Spinal stenosis detection and generation of spinal decompression plan |
| US11426119B2 (en) | 2020-04-10 | 2022-08-30 | Warsaw Orthopedic, Inc. | Assessment of spinal column integrity |
| US12201440B2 (en) | 2020-04-10 | 2025-01-21 | Warsaw Orthopedic, Inc. | Assessment of spinal column integrity |
| CN111755108A (en) * | 2020-05-11 | 2020-10-09 | 深圳市罗湖区人民医院 | Method for treating and managing patients based on critical values |
| CN111755108B (en) * | 2020-05-11 | 2022-02-22 | 深圳市罗湖区人民医院 | Method for treating and managing patients based on critical values |
| CN112259197B (en) * | 2020-10-14 | 2025-02-18 | 北京赛迈特锐医疗科技有限公司 | Intelligent analysis system and method for abdominal plain film of acute abdomen |
| CN112259197A (en) * | 2020-10-14 | 2021-01-22 | 北京赛迈特锐医疗科技有限公司 | Intelligent analysis system and method for acute abdomen plain film |
| US12530860B2 (en) | 2020-11-20 | 2026-01-20 | Hologic, Inc. | Systems and methods for using AI to identify regions of interest in medical images |
| CN114255859A (en) * | 2021-12-23 | 2022-03-29 | 四川大学华西医院 | Method and system for sorting emergency values of radiology department emergency examination reports |
| US12414217B2 (en) | 2022-02-07 | 2025-09-09 | Hologic, Inc. | Systems and methods for adaptively controlling filament current in an X-ray tube |
| US12548676B2 (en) | 2022-09-07 | 2026-02-10 | Mazor Robotics Ltd. | Spinal stenosis detection and generation of spinal decompression plan |
| US12433678B2 (en) | 2022-10-10 | 2025-10-07 | Warsaw Orthopedic, Inc. | Devices, methods, and systems for assessing suitability of spinal implants |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108492874A (en) | The critical value diagnosis and therapy system of intelligence | |
| Back et al. | Development and validation of an automated sepsis risk assessment system | |
| Austin et al. | Early detection of severe maternal morbidity: a retrospective assessment of the role of an Early Warning Score System | |
| Yang et al. | Identification of nursing diagnosis–outcome–intervention linkages for inpatients in the obstetrics department nursing Unit in South Korea | |
| Borgmann | Abortion, the undue burden standard, and the evisceration of women's privacy | |
| Satoh et al. | The Perinatal Committee report: Review of the progress of obstetric healthcare in Japan | |
| Chen et al. | Medical implementation practice and its medical performance evaluation of a giant makeshift hospital during the COVID-19 pandemic: An innovative model response to a public health emergency in Shanghai, China | |
| Hou et al. | Msdiagnosis: A benchmark for evaluating large language models in multi-step clinical diagnosis | |
| Wu et al. | Medical liability of residents in Taiwan criminal court: an analysis of closed malpractice cases | |
| Wu et al. | Study of the relationship between ICU patient recovery and tcm treatment in acute phase: a retrospective study based on Python data mining technology | |
| Mininim et al. | Community based interventions for the reduction of maternal mortality-the role of professional health associations, non-governmental organisations and community-based organisations in delta state, Nigeria | |
| Keijzers et al. | Tertiary survey performance in a regional trauma hospital without a dedicated trauma service | |
| McHale | Medical malpractice in England-current trends | |
| Tulin | Poverty and chronic conditions during natural disasters: A glimpse at health, healing, and Hurricane Katrina | |
| Edwards | The dark side of healthcare: issues, cases, and lessons for the future | |
| England et al. | Resources to support the safe adoption of the revised National Early Warning Score (NEWS2) | |
| Ai et al. | [Retracted] Application of the Whole Optimization of Emergency Nursing Model United and Its Influence on Patients’ Stress Response and Nursing Satisfaction | |
| Nelson | Defining and Understanding ‘Health’in Health Law | |
| Farhat et al. | WACEM 2018 Abstracts | |
| Nnajiuba | Improving Trauma System Effectiveness Through Optimisation of A Major Trauma Triage Tool | |
| Fan et al. | frontiers| Frontiers in Public Health PUBLISHED 06 January 2023 | |
| Stewart | False Positives | |
| Jenkins et al. | WACEM 2018 Abstracts | |
| Yaqoob et al. | WACEM 2018 Abstracts | |
| Gauhar et al. | WACEM 2018 Abstracts |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |
|
| RJ01 | Rejection of invention patent application after publication |