Method and device to support decision-making based on instances

FIELD: information technologies.

SUBSTANCE: method to support decision-making based on instances includes a stage of calculation of remoteness from likeness between the input case of requesting and the set of instances for extraction of similar cases, using the set of standard criteria and their weights for assessment of likeness. Then, in accordance with the method, a user is provided with similar instances and a set of standard criteria and weights. And also an input is received from the user, including a variable weight for one of the set of standard criteria and/or one new criterion in addition to the set of standard criteria. Besides, the calculation of remoteness from likeness is varied with a new set of criteria and weights for extraction of instances similar from the point of view of the user. At the same time a new set of criteria and weights is generated on the basis of clustering on the basis of likeness for variation of calculation of the remoteness from likeness by means of start-up of a genetic learning logic.

EFFECT: creation of a basis system of input estimates of likeness for adaptation of actual value of likeness to similar users with another experience or other opinion.

11 cl, 3 dwg

 

The scope of the invention

The present invention relates to support decision-making on the basis of the cases, and more particularly to a method and apparatus for use in individualized support based cases, for example, in medical applications, such as computer-aided diagnosis (CADx).

The level of technology

Radiologists have to read a lot of images of the scans produced by computed tomography (CT), x-ray, magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), etc. This can lead to "information overload" radiologists. On the other hand, radiologists may mistakenly interpret the scans, which can lead to delays in treatment or unnecessary biopsies. Information overload is a potentially worsens this problem. In such situations, as a consequence, are increasingly being used in the system of decision support, such as computer diagnostics, to improve both workflow and patient outcomes.

The prior art systems computer diagnostics is that clinicians have knowledge of the experience to the situations they encountered earlier. One way by which the system decision support can help clinical doctor in the resolution of the diagnosis, in the example CT scan for lung cancer, is to offer the previous images that have been diagnosed and is like new. The scan can be created by the same or any other techniques, such as x-rays, displaying magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), etc. Paradigm, based on the example (i.e. event-based), is that the knots with a known diagnosis extract from the database of previous cases and represent the radiologist. This is the main premise of the CADx system based on the example.

Patent application WO, entitled "Controlled clinical physician based on the example of computer-aided diagnosis", filed by Koninklijke Philips Electronics N.V with application number IB2007/052307 and not yet published, describes a method and apparatus for optimization of a managed clinical physician based on the example computer system diagnostics. According to the patent application WO, optimization of diagnostics, based on the example (i.e. event-based), is achieved by clustering of interest volume (VOI) in the database in the corresponding clusters according to the subjective assessment of similarity. The optimal set of features of interest volume then choose appropriate examples so that an objective assessment of similarity, based on wybran the x signs, United in clusters VOI database in place of the sign to match the clustering on a subjective basis. Suitable examples are displayed next to the diagnosed VOI to compare clinical doctor.

The invention

Up to the present time for most of the existing systems of support of making decisions on the basis of cases such as the CADx system, it is assumed that something like one radiologist, will also be similar to another radiologist. However, because the likeness is very subjective, it is likely that it will be different with different doctors, even with the same skill level. There is therefore a need to develop reliable metrics of similarity for retrieval, which are personalized for each user. The problem of similarity detection refers to the problem of establishing the true value of similarity. To support the decision, the term "true value" usually means "correct answer", which would give a perfect system. For example, in the computer diagnostic system, which evaluates whether a lung nodule is benign or malignant, the true value for the data set is clinically proven diagnosis for each bundle of light: it is either benign or malignant, which often proved about Ascom biopsy and histopathology analysis. In contrast to the true values for malignant or benign true value of similarity is fuzzy and black and white, and is different for each user. Different radiologists may have different opinions regarding similarity. Thus, the goal is to still create a computer system that will be able to agree with the true value; however, in itself the true meaning is much less clear. The likeness is very subjective and varies from user to user. For example, for a single user pen and ballpoint pen are similar in that they both write. However, for another user, they are two completely different entities with completely different properties.

As suggested here, the system support decision making on the basis of cases is trained on input data from multiple radiologists to install the base system, and the system then enables the radiologist to improve the basic system, based on his/her input data. These inputs are used to either specify the weight indication to calculate the distance from the similarity directly, or to create new clusters, the true meaning of similarity, in order to adapt the calculation of the distance from the similarity to a variety of users, R is slicznym experience and/or different views.

Therefore, in accordance with one aspect of the present invention is a method containing the steps are:

perform the calculation of the distance from the similarity between the introduced version of the query and the set of cases in the database to retrieve such cases, using a set of typical features and their weights to assess the similarity;

provide the user with such cases and a set of typical features and scales;

receive user input comprising a modified weight for at least one from a set of typical symptoms and/or at least one new trait in addition to a set of typical signs; and

change the calculation of the distance from the similarity with the new set of features and weights for retrieving cases similar from the point of view of the user.

When changing the weights for the characteristics or the inclusion of new features to calculate the distance from the similarity, the proposed method directly modifies the operating parameters of the system, and thus fits the retrieval of similar cases for different users with different experience and/or opinion.

In the embodiment, the method further includes the steps in which: take an introductory assessment, user-defined set of cases for clustering by similarity; and generate a new set of signs iveson based clustering by similarity to modify the calculation of the distance from the similarity by running the learning algorithm.

When combining the modified true value specified by the user and generating a new set of features and weights for changes to the calculation of the distance from the similarity, the proposed method indirectly modifies the operating parameters of the system, and thus also fits the true meaning of similarity to other users with a different experience and/or opinion.

In the embodiment, the method additionally includes the stage at which generate a new set of features and weights as the new basis for calculation of the distance from the similarity on the basis of personal preferences signs and scales collected from a group of users.

Thus, the new basis is defined as a specific group of users, such as specific for the hospital. Additionally, the difference between the personal settings for each user and the new basis of the true value is estimated to identify deviations. When inexperienced users will want to learn from experienced users, they can use the settings for advanced users.

In accordance with another aspect of the present invention, provided is a device containing:

the extraction block, configured to perform the calculation of the distance from p is dobie between the introduced case query and a set of cases to retrieve similar cases, using a set of standard signs and scales for assessment of similarity;

the block representation made with the possibility of presentation to the user of such cases and a set of typical signs and scales;

the reception unit, configured to receive from a user input that includes the modified weight for at least one from a set of typical signs or at least one new trait in addition to a set of typical signs; and

block changes with changes to the calculation of the distance from the similarity with a new set of signs and/or weights to retrieve cases that are similar from the point of view of the user.

In the embodiment, the reception unit is additionally configured to receive input estimates, user-defined set of cases for clustering by similarity, and the changing unit is additionally configured to generate a new set of features and weights based on clustering by similarity to modify the calculation of the distance from the similarity by running the learning algorithm.

Modifications and changes of the invention as defined in independent clauses, which correspond to the described modifications and changes can be carried out by a specialist in the art based on the present description.

DESCRIPTION of DRAWINGS

The above and other objects and features of the present invention will be better understood from the subsequent detailed description, considered together with the attached drawings, on which:

Figure 1 is a block diagram of a sequence of operations that provides an overview of an exemplary variant of the method in accordance with the invention.

Figure 2 is a block diagram of a sequence of operations that shows a sample implementation of process improvements in the method in accordance with the invention.

Figure 3 is a block diagram showing an exemplary version of the exercise device 300 in accordance with the invention.

The same reference positions are used to indicate similar parts in the figures.

DETAILED DESCRIPTION

Figure 1 is a block diagram of a sequence of operations that provides an overview of an exemplary variant of the method in accordance with the invention. In accordance with this method, the system of support of decision-making on the basis of the first cases to be trained on input data from multiple radiologists in order to install the base system (step 100). The system then adapts to the particular radiologist process improvements based on input radiologist (step 101).

Training basis can occur either during product development, or what about the install time in the hospital, a training module, built-in system, for example, using the method described in patent application WO IB2007/052307. Once the system is trained for General fill, defines a set of standard signs and weights used to calculate the distance from the similarity. These signs and weight are used for the objective assessment of the similarity between the diagnosed cases (i.e., the requested cases) and the cases in the database.

When the radiologist first logs in, he/she may want to personalize the true meaning and function similarity for their own use with process improvements provided by the invention. Improving the true meaning of similarity and function similarity based on inputs radiologist that can directly change the weight indication or provide new clusters the true meaning of similarity, or a combination of both. When the radiologist is satisfied with the results of retrieval on the basis of adjusted true value and functions of similarity after several iterations, the improvement process is stopped and the system is now personalized for a specific radiologist, who will extract relevant similar cases, as they are understood by the radiologist.

Figure 2 is a flowchart of the sequence of operations showing the approximate variant implementation, developing PR is the process improvements for example, the stage 101 in figure 1. It is assumed that the system support decision making on the basis of cases was developed and studied as described above. According Fig. 2, the method includes the step 210 of receiving the case of a request from a user, which may be a radiologist, a doctor or a new doctor.

Then the method further includes the step 220 perform the calculation of the distance from the similarity between the introduced case and set of cases in the database. Such cases are extracted using a set of typical signs and scales for assessment of similarity.

Each case, that is, images and information, associative associated with the medical object, an associative link with the set of features that characterize this case. These characteristics may include characteristics such as the effective diameter, degree of circularity, the contrast, the average value of brightness, angularity, field density, the standard deviation of the pixels, the ratio of the radial gradient, etc. Clinical data specific to the patient, such as age, history of cancer, etc. can also be a sign. Under the "standard signs", we mean a list of the types of signs that are defined in advance, but for each case are clearly different values associated with each sign. Typical signs and choosing the appropriate weights are used to calculate the distance from the similarity to objectively assess the similarity between a case in the query and the cases which is stored in the database in accordance with user selection. If you change the signs or weight, that is, change the calculation of the distance from the similarity, the system operating parameters, that is, the extracted results are changed accordingly.

Such cases can be retrieved from its own database of the hospital, including cases that were previously evaluated, diagnosed, or treated in this hospital, from your own user-defined reports, including cases that are marked as diagnosed by that particular user, or from a pre-selected training set including cases Packed baseline system. Case requested by the user, and the cases retrieved from the database, may include images and/or text. For example, the case requested by the user, can be diagnostic image (or set of images) of the medical object, such as a lung nodule, and the retrieved cases can be a lung nodules that were diagnosed as malignant or benign. Case query and the extracted results can be equipped with an additional texts, such as information from the Electronic Health Report a case of a request or retrieved cases.

The method further includes the step 230 views user is the user for his/her subjective assessment of such cases and a set of typical signs and weights, used in the calculation of the distance from the similarity to an objective assessment of the similarity between the requested case and cases in the database selected by the user.

The method further includes the step 235 receiving input from the user indicating whether further adaptation. The user evaluates the retrieved similar cases and determines whether or not he/she has to improve a host system. If the user is satisfied with the retrieved results (i.e. the true value of similarity or calculation of the distance from the similarity adequately adapted to the user), the system does not need further adaptation. Otherwise, the user has the possibility to adapt the system to his/her needs.

The method further includes the step 240 of receiving user input that includes the modified weight for at least one from a set of typical signs or at least one new trait in addition to a set of typical characteristics. User input reflects personal opinion or preference similarity between cases. In the embodiment, the user can explicitly change the typical characteristics and/or their weight. For example, the user enters information, which will change the weight for an existing sign in the base system, or ubavlyaet new characteristic, for example, changing the size of the tumor due to the treatment, which was excluded from the basic system.

In an alternative embodiment, the change characteristics can be performed implicitly by the training session. For example, the system can accept input provided modified true values: a different set of cases, for example a set of images of nodules of the lung, other than those originally used in the baseline system, which must be classified by the user as being similar or not, with this case the query. Alternatively, the user may be presented with a set of nodules of the lung, and he/she will assess which ones are similar and which are not. The user can also specify what attributes derived from the cases used for clustering. Thus, the user creates a new true value of similarity, which he or she prefers.

The method further includes the step 250 changes to the calculation of the distance from the similarity with the new set of features and weights for retrieving cases similar from the point of view of the user. When the weight for characteristics directly modified, or new signs directly added user input, calculating a distance from the similarity will change with the new set of features and weights, directly to the public leading to changes in the operating characteristics of the system.

When the true value is changed, a new set of features and weights will be generated based on user input, i.e. clustering similarity, to change the function of similarity by performing the learning algorithm. With a new set of features and weights, the computation of the distance from the similarity will change, indirectly resulting in changes to the performance of the system.

The details of how to generate or select a new set of features and weights can be found in the description of the patent application WO IB2007/052307, in which the genetic algorithm (GA) is used to find the optimal space-based, and, preferably, optimal pointwise similarity criterion for use in the optimal space characteristic. Optimal space characteristic must be inferred from the totality of signs, which signs-participants can be derived from subjective ranking of the signs on the stage clustering. The set of features may be more inclusive to include any recoverable signs images, or relevant clinical data related to VOI in the database. Candidates similarity criteria may differ from each other only in the pointwise measure of remoteness.

Each time a new chromosome is created by reproduction in GA, estimated health x is omoomi. As each chromosome represents a set of characteristics and preferably a relevant indicator of the distance, and as the most fit chromosome is selected during meet the stopping criterion, each health assessment chromosome can be considered as an iteration of the iterative process. Iterative, then, it selects a set of signs and, preferably, an indicator of remoteness.

Once the set of features and weights for the similarity function is changed, the user can additionally request system using the case of the request. Case the request may be the same that was used in the previous query or new input by the user. In this situation, the calculation of the distance from the similarity will be performed using a new set of features and weights that reflect personal true meaning of similarity to the user, i.e. the user's opinion regarding the similarity between cases. In this situation, the process returns to step S110 for further query.

According the occasion of the request and the updated set of features and weights, the computation of the distance from the similarity is performed again to evaluate the similarity between the requested case and cases in the database to retrieve cases that are similar to the user. The extracted cases pre is delivered to the user with new features and weights, used to calculate the distance from the similarity. The user can view and evaluate the retrieved similar cases again and determine the need for further improvement. If the user is still not satisfied with the retrieved result, he/she may cause the system to rerun the process of improvement. After many iterations of process improvement, the user might be satisfied with the retrieved cases, and the system is now upgraded or personalized for the user.

As the only true similarity value adapts to the user, the process proceeds to step S255, on which the user saves the settings, that is updated signs and weight for future use. The user can be allowed to have more than one set of settings personalization for use in various application areas, for installation of check boxes to allow or restrict access to his/her own settings for use by other users and to upload their personal settings on his/her computer as needed.

The advantage is that a variant of the method allows all the retrieved cases to change in real time so that the user can easily adjust isplace the Oia, and allows the user to view two or more sets of queries and the retrieved results when the correction weight. Next, the process may provide additional management of an experienced user, for example, weighting the characteristics manually or semi-automatically.

The advantage of the above-mentioned embodiments is that the method includes the step of personalization by the user of the number of retrieved cases and data or images that he would like to see for each retrieved case, that is what clinical information he would like to have presented. At any stage during the improvement process or during use of the system, the user can further personalize the system, indicating that a particular case should never be removed when he or she uses the system, because, for example, the user has concluded that it is "atypical" case.

The improvement process can be performed for the first time when the user is logged on or at any time in the future at his/her discretion. For example, it may be desirable to change the settings, personalization, that is, the signs and weight, used to calculate the distance from the similarity, because the user gets more experience.

In d the natives embodiment, the method further includes the step 260 of generating a new set of signs and weight as the new basis for calculation of the distance from the similarity. The step of generating a new set of features and weights based on a set of personal settings of signs and scales collected from a group of users, such as physician groups, using the system in the hospital. Personalized signs and weights could be set explicitly by directly changing signs/Libra or implicitly through the training session, and the new basis can be defined as specific for the hospital.

At step 265 assessment calculated the difference between the new basis and personal settings for each user to identify deviations. When inexperienced users will want to learn from experienced users, and they can use the settings for advanced users.

The above-mentioned method, as shown in Fig. 1 and 2, may be implemented by software or hardware, or a combination of both.

Fig. 3 is a block diagram showing a sample implementation of the personalized device 300 of decision support on the basis of the cases according to the invention. The device 300 includes:

block 310 extraction performed with the opportunity to perform the calculation of the distance from the similarity between the introduced case query and a set of instances from the database to retrieve similar the cases, using a set of standard signs and scales for assessment of similarity, that is, to perform the function of step 220 retrieval;

block 320 views made with the possibility of presentation of such cases and a set of typical signs and weights of the user, i.e. to perform the function of step 230 views;

the reception unit 330, configured to receive from a user input that includes the modified weight for at least one from a set of typical signs or at least one new trait in addition to a set of typical features, that is, to perform the function of step 240 admission; and

unit 340 changes made with the possibility of changes to the calculation of the distance from the similarity with the new set of features and weights for retrieving cases similar from the point of view of the user, i.e. to perform the function of step 250 changes.

The device 300 may also contain base data 303 includes cases for extraction and the internal bus 305 to collect blocks in the device 300.

In the embodiment, the reception unit 330 is additionally configured to receive input estimates, user-defined set of cases for clustering by similarity, and additionally capable of receiving input characteristics that the user used for clustering by similarity.

Another is the version of the implementation unit 340 changes made with the possibility of generating a new set of features and weights based on clustering by similarity to modify the calculation of the distance from the similarity by running algorithm learning.

In another embodiment, the device 300 further comprises a block 345 control, configured to control the iterative stage on which to perform the calculation of the distance from the similarity using the updated set of features and weights for retrieving cases similar from the point of view of the user.

In yet another embodiment, the device 300 further comprises a unit 350 estimates made with the possibility of generating a new set of features and weights as the new basis for calculation of the distance from the similarity on the basis of personal preferences signs and scales collected from a group of users. Block 350 advanced assessment designed to assess the difference between the new basis and each personal user preference for characteristics and scales, to identify deviations.

Specialist in the art will understand that the invention can be enhanced by other features such as a flexible user interface and authentication management. The invention can be integrated into Informatics radiology or products of health Informatics as a sign or as a separate add-on module. The invention may also be implemented as a stand-alone product workstation-based CADx cases.

The image is a buy can be used for computer-assisted diagnosis, together with any methods of imaging. In particular the invention can be used to aid in the diagnosis of various diseases or to confirm suspected diagnoses during the diagnosis process performed by radiologists. Other applications include training, diagnosis in emergency and computer control therapy on the basis of the cases.

It should be noted that the above embodiments of illustrate, but not limit the invention, and that the specialists in the art will be able to develop alternative implementation, without departing from the scope of the applied claims. In the claims, any reference signs items placed in parentheses shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claim or in the description. The use of the singular does not exclude the presence of many such elements. The invention can be carried out by the module hardware comprising several distinct elements, and the module is programmed computer. In the claims relating to the device, listing several blocks, some of these modules may be embodied by one and the same element of AP is aratoga hardware or software. The use of the words "first", "second" and "third" and so on does not imply any ordering. These words should be interpreted as a name.

1. The method of decision support on the basis of cases containing phases in which:
perform (220) calculation of the distance from the similarity between the introduced case query and a set of cases to retrieve such cases, using a set of typical features and their weights to assess the similarity;
represent (230) to the user such cases and a set of typical features and scales;
accept (240) from the user input comprising the modified weight for at least one from a set of typical symptoms and/or at least one new trait in addition to a set of typical features;
change (250) calculation of the distance from the similarity with the new set
signs and weights to retrieve cases that are similar from the point of view of the user, while
at step (240) take an introductory assessment specified by the user
many cases for clustering by similarity, and
at step (250) generate a new set of features and weights based on clustering by similarity to modify the calculation of the distance from the similarity by running a genetic learning algorithm.

2. The method according to claim 1, wherein the step of generating additional is that take input is of Reznikov, the user used for clustering by similarity.

3. The method according to claim 1, additionally containing phase, which take input from the user to exclude the case from the retrieved cases are similar from the point of view of the user.

4. The method according to any one of claims 1 to 3, additionally containing an iterative stage on which to perform the calculation of the distance from the similarity using the updated set of features and weights for retrieving cases similar from the point of view of the user.

5. The method according to claim 4, additionally containing phase, which generate (260) a new set of features and weights as the new basis for calculation of the distance from the similarity on the basis of personal preferences signs and weights collected for a group of users.

6. The method according to claim 5, additionally containing phase (265), which estimate the difference between the new basis and the personal preferences of each user for the characteristics and scales to identify deviations.

7. The method according to claim 1, containing a stage at which selects cases from the database or from a pre-selected training set, which includes cases Packed reference system.

8. Device support decision making on the basis of cases containing:
block (310) extraction performed with the ability to perform computation of returns the items from the similarity between the introduced case query and a set of cases to retrieve similar cases, using a set of standard signs and scales for assessment of similarity;
block (320) representations made with the possibility of presentation to the user of such cases and a set of typical signs and scales;
block (330) receiving, configured to receive from a user input that includes the modified weight for at least one from a set of typical signs or at least one new trait in addition to a set of typical signs; and
block (340) changes with changes to the calculation of the distance from the similarity with the new set of features and weights by presenting new the true meaning of similarity for retrieval of cases that are similar from the point of view of the user, while
unit (330) for receiving is configured to receive an introductory assessments, user-defined set of cases for clustering by similarity, and
block (340) changes made with the possibility of generating a new set of features and weights based on clustering by similarity to modify the calculation of the distance from the similarity by running a genetic learning algorithm.

9. The device according to claim 8, in which the block (330) receiving additionally configured to receive input characteristics that the user used for clustering by similarity.

10. The device according to claim 8 or 9, further containing block is (345) management made with the ability to control the iterative stage on which to perform the calculation of the distance from the similarity using the updated set of features and weights for retrieving cases similar from the point of view of the user.

11. The device according to claim 10, further containing block (350) assessments made with the possibility of generating a new set of features and weights as the new basis for calculation of the distance from the similarity on the basis of personal preferences signs and scales collected from a group of users.



 

Same patents:

FIELD: information technology.

SUBSTANCE: image display device has an image display unit configured to display a still image or a moving image, an image capturing unit configured to capture an image in a direction in which the image display unit displays a still image or a moving image, and a face recognition unit configured to execute a face recognition process during which, through given intervals, the face of the user contained in the image captured by the image capturing unit is recognised, wherein the given intervals vary according to whether the face of the user is contained in the image captured by the image capturing unit, and if the recognised face in its entirety or a part thereof does not change position over a given time interval during the face recognition process, the face recognition unit does not consider the recognised face the face of the user.

EFFECT: automatic control of the operating mode of an image display device.

22 cl, 16 dwg

FIELD: information technologies.

SUBSTANCE: method includes sending a document past an image signal generator with simultaneous serial illumination of the document using several types of illumination, at the same time for each line scanned by the image signal generator a pre-specified type of illumination is used, which makes it possible to generate a document image with alternation of several types of lines. Such image with alternation of several types of lines is exposed to multi-dimensional transformation function, as a result a combined measurement is produced, which is then projected to an empirically determined vector of suitability, in order to produce a scalar value of suitability for this document. This value is compared to the empirically established value of suitability for identification of whether this document is suitable enough to stay in circulation or it shall be removed from circulation and destroyed.

EFFECT: increased accuracy of document suitability inspection.

15 cl, 10 dwg

FIELD: physics.

SUBSTANCE: disclosed is a document photosensor which comprises upper and lower measuring assemblies (1 and 2) lying on opposite sides of a passageway (55) along which a document (50) is transported. The upper measuring assembly (1) comprises an upper substrate (11), an upper light-emitting diode (LED) chip (21) and an upper photodiode chip (37) mounted on the surface of the upper substrate (11). The lower measuring assembly (2) comprises a lower substrate (12), a lower LED chip (31) and a lower photodiode chip (38) mounted on the surface of the lower substrate (12). These chips (21, 37, 31 and 38) are precisely located on the upper and lower substrates (11 and 12) so that the photodiode chips (37 and 38) accurately detect light emitted by the LED chips (21 and 31) and penetrates through a particular point on the document (50), which improves bill validation.

EFFECT: high accuracy of detecting light-emitting elements.

15 cl, 23 dwg

FIELD: information technology.

SUBSTANCE: light control system has a light source (4) for illuminating a region (10), a pointing device (6) for projecting location information (11) onto the region (16), and a receiver (5) for detecting location information (11), the system being configured to determine the position of location information (11) on the region (10), and the controller being configured to control illumination of the region (10) depending on said location information (11).

EFFECT: easier interaction of a user with a light source.

9 cl, 3 dwg

FIELD: information technologies.

SUBSTANCE: in the method a test colour sample is printed on a device of anaglyph images printing, functions of spectral transmission are assessed for the left and right filter of glasses for a printed test sample, a stereomate is generated, an anaglyph image is generated for a stereomate, and the anaglyph image is printed.

EFFECT: adaptation for printing of an anaglyph image with preservation of 3D perception of a printed image for specific stereo glasses.

7 cl, 7 dwg

FIELD: information technology.

SUBSTANCE: method for perception of images of microscopic objects involves forming a photoelectric converter (PEC) matrix using circular or hexagonal insulated PEC in form of a dense hexagonal package thereof, said PEC are grouped and identical hexagonal PEC rings which are not connected to each other are formed in the matrix; PEC of each hexagonal PEC ring are combined and a cluster is formed; the clusters are grouped in form of a series of adjacent clusters and a hexagonal ring of clusters is formed; the set of hexagonal rings of clusters form a hexagonal-cluster-structured matrix on which the image field of objects is projected; a closed hexagonal path of movement of image field elements is formed, which, in the matrix plane, is identical to PEC distribution in the hexagonal PEC rings of the clusters, wherein after each step of movement of the image field, the signal at the output of all hexagonal PEC rings of clusters of the matrix is detected; at the PEC lying at the geometric centre of a cluster, and at the PEC of a cluster on which at the current step the dominant element is projected, signals of dominant element are detected as signals of points.

EFFECT: eliminating interference of diffraction light scattering on perception of adjacent points, high resolution and faster perception and processing of image elements, providing three-dimensional perception of image field elements in real time.

9 dwg, 2 tbl

FIELD: process engineering.

SUBSTANCE: proposed method comprises embossing identification signs on packing foil and/or package in hot-pressing jointly therewith and applying logos by embossing. Note here that set of marks is embossed as a pattern at conveyor, read by appropriate device, compared with pattern and estimated by image processing method. Said pattern comprises symbols while embossing roller or rollers represent roller processed to pattern. Proposed device comprises, at least, two embossing rollers, read device and estimation device. Note here that one of said embossing rollers is provided with drive. Roller comprises some teeth to apply identification signs feature different shape, height or surface structure while other teeth are used hot-pressing. One of said rollers is furnished with rings extending in circle.

EFFECT: simplified identification, fabrication and expanded range of used foil.

12 cl, 13 dwg

FIELD: information technology.

SUBSTANCE: method involves forming images, labelling a plurality of documents while forming on each document said image, having differences which are unique for each document, wherein most images made on said documents have a certain physical characteristic which provides copy protection and which satisfies a predetermined criterion, and said characteristic of most copies of these images which can be made does not satisfy said criterion; characteristics of said differences are obtained for each said document in order to obtain a unique image of the labelling and said unique image is stored.

EFFECT: wider range of physical properties of material of documents used during identification thereof.

26 cl, 32 dwg

FIELD: medicine.

SUBSTANCE: system comprises a digital image input unit, a system parameter input unit, a final image reader and memory; an image morphological filtration module, a separator for local minima of image intensity; a measurer of distance from each image pixel to the nearest background pixel; a separator for local maxima of image intensity; an OR element; a noise object border filter; a morphological image gradient detector; a border mask unit of the required objects by segmentation; a calculator of selected object coordinates; a unit of obtaining the optical and geometrical characteristics of the detected objects and the coordinate database, the optical and geometrical characteristics of the detected objects.

EFFECT: higher accuracy of object selection on the images containing a number of irregularly distributed small spatial objects.

5 cl, 8 dwg

FIELD: information technology.

SUBSTANCE: display includes the following: a display unit which includes a first and a second display region, which can establish a first and a second coordinate system; memory which stores a display file containing image data, coordinate range data and data of a table of coordinate values; and a processor which performs: indication processing of image files, which indicates the image file; display processing which (i) reads said data, (ii) controls display for part of the image from the data image of said image, (iii) establishes the first coordinate system in a range which overlaps part of the image, and (iv) controls display of the constructed points of the first coordinate system, wherein data of the table of coordinate values correct the coordinate value of a third coordinate axis with coordinate values in the first coordinate system, and the display processing establishes the second coordinate system in the second display region and controls display of the constructed points.

EFFECT: broader functional capabilities of the display device and machine-readable medium by providing a higher training effect.

17 cl, 24 dwg

FIELD: information technologies.

SUBSTANCE: under clinical conditions, when at any time there are several patients, there are central stations (10) of patient control, for instance, nursing units, for combination of the collected information relative to physiological parameters of patients. Data is displayed in several subwindows (22) of the display (18) of the control station (10). Due to certain limitations for dimensions of the display (18), it is often difficult to distinguish data displayed in subwindows (22), or even display all collected data. The user may expand any such subwindow (22) into a scale-variable subwindow (32), which provides for more functions than any other subwindow (22), without full coverage or adjustment of size of any other subwindow (22).

EFFECT: improved access to information.

12 cl, 6 dwg

FIELD: chemistry.

SUBSTANCE: method of operating a device for measuring an analyte, having a display, a user interface, a processor, memory and user interface buttons includes steps of: measuring the analyte in the body fluid of a user using the analyte measuring device; displaying a value representing the analyte; prompting the user to select an indicator for linking with the displayed value; and pressing one of the user interface buttons only once to select an indicator linked with the value of the analyte, and storing the selected indicator together with the displayed value in the memory of the device. The group of inventions also relates to a method of operating the measuring device, which additionally includes a step of ignoring activation of any of the user interface buttons except the selected button.

EFFECT: more intuitive and easier use of the device for measuring an analyte, eg a glucometer.

20 cl, 12 dwg

FIELD: information technology.

SUBSTANCE: method of extracting a plurality of data layers from a set (5) of data of medical images, wherein the method includes the following steps: a) displaying an indicator (10, 20) associated with the plurality of data layers; b) selecting the indicator (10, 20) based on user input; and c) extracting the plurality of data layers associated with the indicator when said indicator is selected; wherein the link between the indicator and the plurality of layers is based on segmentation of the set of data of medical images, wherein the indicator is an object obtained during segmentation of the set of data of medical images, and the plurality of data layers include object data, wherein the object data are contained in the plurality of layers.

EFFECT: reducing the amount of data transmission.

12 cl, 7 dwg

FIELD: physics.

SUBSTANCE: invention discloses a computer implemented method and system for conducting a geologic basin analysis in order to determine the accumulation of hydrocarbons in a subsurface region of interest. According to the disclosure, a basin analysis project is defined within a subsurface region. At least one basin analysis cycle is applied to the basin analysis project and the results of the basin analysis are integrated to generate basin analysis project results for the basin. The project results are used to optimise and manage the performance of technical tasks required to determine the accumulation of hydrocarbons in the subsurface region of interest.

EFFECT: high accuracy and information value of survey data.

20 cl, 26 dwg

FIELD: radio engineering, communications.

SUBSTANCE: device comprises P units of maximum signal separation, P units of activation function calculation and P groups of membership function values generation units.

EFFECT: increased accuracy of recognition when recognising objects with separate low or partially distorted areas.

1 dwg

FIELD: information technology.

SUBSTANCE: apparatus has a synchronisation unit 1, an integrated unit 2, a switch 3, units for controlling and linearising transfer characteristics of multichannel converters 4, counters for counting the number of times a fault detection subunit 5 is switched, a control unit 6, memory units 7 and 8. The output of the synchronisation unit is connected to the input of an interfacing unit, and a multidimensional sequence generator is in form of a multichannel device of a matrix structure with feedback, and the data output of the interfacing unit fully conforms to data connections, and its data output is connected to the input of the switch.

EFFECT: high accuracy of simulation by combining control of transfer characteristics and statistical estimation of the frequency index of the effect of the set of destabilising factors.

2 dwg

FIELD: medicine.

SUBSTANCE: ECG analysis system comprising: an input device for receiving an ECG enquiry from a patient, and a control system configured to extract the information relevant to the patient, to consolidate the received enquiry and the extracted information into an updated enquiry, and to send the updated enquiry to an ECG apparatus; the extracted information contains the previous ECG result selected from a variety of the previous ECGs with the use of the best-previous ECG algorithm. The method describes the system operation procedure.

EFFECT: higher accuracy and effectiveness of the automated ECG analysis providing more information to ECG specialists and transcriptionists.

13 cl, 2 dwg

FIELD: medicine.

SUBSTANCE: drug delivery system includes: a drug delivery point comprising at least one protected compartment configured to store the drugs, a controller responding to the status information on patient's admission to hospital and configured to: to prescribe at least one of these protected compartments for administering and protecting the drugs for the patient; to give a selective access to the drugs for patients in at least one protected compartment when the status information on patient's admission to hospital indicates that the patient is currently admitted, and to clock the access to retrieve the drugs for the patient in at least one protected compartment when the status information on patient's admission to hospital indicates that the patient is not currently admitted, wherein the controller is further configured to generate a notification to the pharmacy console to remove the drugs from at least one protected compartment prescribed to the patient once the patient is discharged from hospital.

EFFECT: patient-specific drug delivery.

15 cl, 8 dwg

FIELD: medicine.

SUBSTANCE: invention relates to medical equipment. System of visualisation in vivo contains capsule with animal, which includes one or more holders or bed, cylindrical cover, sensors of physiological parameters, identifier and docking interface, providing control, heating and anesthesia for one or more animals. System contains one first means of visualisation for collection in vivo data of animals in the field of visualisation of the first visualisation means, which includes docking port and reading device. System contains reconstructing processor, which reconstructs presented image from in vivo data for visualisation, installation unit or preparation unit, which includes docking port. System contains research work station, which includes electronic interface for interaction with the first visualisation means, user interface, which provides user with possibility of projecting new research and modification of existing research. System contains display devices, which displays vital parameters of animals during visualisation and preparation, and system of animal control and anesthesia for control of vital signals from animals during influence with sedative medication and for provision of regulated introduction of anesthesia into object via several docking interfaces.

EFFECT: application of the invention will make it possible to assess several biological parameters simultaneously within one visualisation procedure.

13 cl, 7 dwg

FIELD: medicine.

SUBSTANCE: once the new content arrived, a medical assistant mark it on a template. Then a planner (36) looks through all the copies of the previous content and replaces it by the new content as smooth as possible. The content replacement data base (40) interprets the previous data in view of the new content so that the previous data shall not be considered as deceptive or outdated.

EFFECT: enhanced presentation of the required information for the patients due to creation of the medical-sanitary care network.

19 cl, 2 dwg

FIELD: computers.

SUBSTANCE: device has decoder, registers, AND groups elements, delay elements, memory blocks, counter, trigger, signs input block, comparators.

EFFECT: higher productiveness.

2 dwg

Up!