CN117932472A - Method for classifying fetal heart rate deceleration based on fast Fourier transform - Google Patents
Method for classifying fetal heart rate deceleration based on fast Fourier transform Download PDFInfo
- Publication number
- CN117932472A CN117932472A CN202410133749.7A CN202410133749A CN117932472A CN 117932472 A CN117932472 A CN 117932472A CN 202410133749 A CN202410133749 A CN 202410133749A CN 117932472 A CN117932472 A CN 117932472A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- fetal heart
- fourier transform
- fast fourier
- classifying
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 208000031427 Foetal heart rate deceleration Diseases 0.000 title claims description 29
- 210000002458 fetal heart Anatomy 0.000 claims abstract description 18
- 208000036029 Uterine contractions during pregnancy Diseases 0.000 claims abstract description 13
- 238000013135 deep learning Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000000737 periodic effect Effects 0.000 claims 1
- 238000010183 spectrum analysis Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 11
- 238000012544 monitoring process Methods 0.000 abstract description 10
- 230000008569 process Effects 0.000 abstract description 4
- 239000012634 fragment Substances 0.000 abstract description 3
- 238000010276 construction Methods 0.000 abstract 1
- 239000000284 extract Substances 0.000 abstract 1
- 238000005215 recombination Methods 0.000 abstract 1
- 230000006798 recombination Effects 0.000 abstract 1
- 230000001605 fetal effect Effects 0.000 description 7
- 210000003754 fetus Anatomy 0.000 description 6
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 5
- 229910052760 oxygen Inorganic materials 0.000 description 5
- 239000001301 oxygen Substances 0.000 description 5
- 206010033799 Paralysis Diseases 0.000 description 4
- 230000004913 activation Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 230000008774 maternal effect Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 108010076504 Protein Sorting Signals Proteins 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000002490 cerebral effect Effects 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000003340 mental effect Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000002035 prolonged effect Effects 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 208000013824 Acidemia Diseases 0.000 description 1
- 208000010444 Acidosis Diseases 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 208000006893 Fetal Hypoxia Diseases 0.000 description 1
- 230000003187 abdominal effect Effects 0.000 description 1
- 210000003403 autonomic nervous system Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 208000029028 brain injury Diseases 0.000 description 1
- 230000005792 cardiovascular activity Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000001802 infusion Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007472 neurodevelopment Effects 0.000 description 1
- 230000008288 physiological mechanism Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention provides a method for extracting frequency domain features based on Fast Fourier Transform (FFT) and carrying out deceleration classification by combining deep learning. The method comprises the following steps: 1. firstly, the extracted decelerated Fetal Heart Rate (FHR) and Uterine Contraction (UC) signals are intercepted into fragments with fixed length so as to be convenient to process; 2. and (3) extracting the frequency domain characteristics of the signal segments based on the signal segments intercepted in the step (1) by using fast Fourier transform. The high-frequency signals separated from the frequency domain are converted into a two-dimensional input form suitable for deep learning through dimension recombination; 3. based on the two-dimensional signal data obtained in the step 2, the invention adopts a depth learning model constructed by a Inception module in the field of image processing to carry out iterative training, accurately extracts deceleration characteristics from signals, and carries out fine classification. The invention provides a high-efficiency and low-cost solution for the deceleration recognition system, and has obvious advantages in the simplicity of model construction and execution efficiency. The application of the invention can significantly improve the performance of the related medical system in the aspects of deceleration monitoring and classification.
Description
Technical Field
The invention relates to the technical field of medical monitoring and signal processing, in particular to a method for analyzing and classifying a Fetal Heart Rate (FHR) deceleration event by utilizing Fast Fourier Transform (FFT).
Background
Electronic Fetal Monitoring (EFM) is a common obstetrical technique in the medical arts and has been used to assess fetal health since the 60 s of the 20 th century. Fetal heart Gong Sutu (CTG) it monitors records Fetal Heart Rate (FHR) and maternal Uterine Contractions (UC) signals. Currently, ultrasonic doppler transducers are widely used for FHR monitoring due to their practical and anti-interference capabilities. Meanwhile, the UC signal is acquired by a pressure sensor in the abdominal band. These monitoring means not only reveal cardiovascular activity, but also indirectly reflect the state of the fetal autonomic nervous system, are critical for assessing neurological development, and can assist in detecting neonatal acidemia risk during delivery.
In clinical practice, physicians typically rely on experience to interpret CTG signals and decide accordingly whether to perform a medical intervention. Timely proper intervention is critical in rescuing the fetus and preventing serious consequences. Conversely, untimely interventions may be harmful to the fetus, while unnecessary interventions may increase the physical and mental and economic burden of the pregnant woman. It has been studied that about 50% of neonatal brain injuries can be prevented by accurately interpreting CTG and performing appropriate rescue. Whereas CTG analysis monitors fetal conditions by assessing two main signals-characteristics of Fetal Heart Rate (FHR) and Uterine Contractions (UC), such as FHR baseline, range of variation, acceleration and deceleration patterns. The deceleration is a key index for judging whether the fetus is possibly lack of oxygen, and if the lack of oxygen is continuous, serious consequences such as neonatal paralysis or cerebral paralysis can be caused, medical intervention is needed in time. Medical measures may include caesarean section and aspiration delivery, etc., or non-invasive posture adjustment, oxygen delivery and adjustment of drug infusion rates. Deceleration can be subdivided into early deceleration, late deceleration, variable deceleration, and prolonged deceleration. Before birth, CTG shows less and slight retardation, while the time-of-birth retardation is more frequent and of greater concern.
Continuous monitoring of fetal heart rate deceleration is critical to early identification of fetal hypoxia. Challenges include the handling of large amounts of data and the rarity of deceleration events. In addition, experts suggest that clinicians need to understand the physiological mechanism of deceleration and CTG patterns, otherwise manual analysis alone may result in wasted medical resources. However, CTG analysis still has problems of low consistency of judgment by doctors and high misdiagnosis rate. Therefore, it is important to develop a method for classifying fetal heart rate deceleration based on a Fast Fourier Transform (FFT). The method can report abnormal changes in time by monitoring the CTG with the aid of a computer, and effectively improves the working efficiency of doctors and the accuracy of diagnosis. Such advances can significantly improve the protection of maternal and infant health, especially in resource-constrained medical environments.
Disclosure of Invention
The invention provides a deceleration classification method capable of optimizing the problems, aiming at solving the problems of low efficiency and accuracy and complex fetal heart signal preprocessing process in the existing deceleration classification.
Based on the fetal heart rate signal processing problem, the invention further contemplates extracting the period and frequency in the signal with a fourier transform, where f (t) is the original signal of the input. In this scenario the problem is represented as a given series of deceleration segments, and then fourier transformation is used to derive the amplitude information F (ω) of the fetal heart rate signal at different frequencies. Then, in order to obtain the amplitude information of each frequency component in the frequency domain, this is achieved by calculating the absolute value of F (ω) and averaging it, which can be expressed as follows:
Amp＝Avg(|F(ω)|) (2)
Thus, the significant frequency represented by top-k is then represented as the first k frequency components of the average amplitude set Amp with the greatest amplitude. Considering the sparseness of the frequency domain, the invention selects only the first k amplitude values in order to avoid noise from meaningless high frequency bands. These amplitudes have the greatest effect on the feature recognition of the entire signal. The method calculates the top-k amplitude value in the signal in advance to obtain a top-k amplitude set. In the deep learning feature extraction used later, the one-dimensional signal problem is converted into the two-dimensional problem by superposing top-k vibration, so that the final classification is more accurate.
In order to achieve the technical aim, the invention provides a method for classifying fetal heart rate deceleration based on fast Fourier transform, which is characterized by comprising the following steps:
Step 1, receiving a signal data set S of an original fetal heart rate time sequence;
step 2, intercepting or filling the data set S obtained in the step 1 into a preset continuous fragment set S' according to the length L;
Step 2.1, constructing a problem set S' according to the deceleration segment, and transforming the fetal heart rate deceleration segment by using Fourier transform, and outputting an amplitude set { f 1,...,fn-1,fn } according to the result;
step 2.2, calculating the period information p i of the amplitude of the signal by dividing the number of time steps of the signal by each frequency component, expressed as:
where f k denotes the resulting amplitude, T denotes the number of time steps of the original fetal heart rate deceleration signal segment, and p i is the period corresponding to each frequency component.
Step 2.3, if the period set { p 1,...,pn-1,pn } and the signal sequence L are not proportional, performing zero expansion to the least common multiple of the period, which can be expressed as:
Step 2.4, calculating to obtain an average amplitude set, obtaining the maximum k amplitudes according to the top-k, obtaining a top-k amplitude set, and during problem selection, reshaping the original 1D signal into a 2D map according to the top-k amplitude set to adapt to subsequent feature extraction, wherein the method is expressed as:
X2D＝Concatfi,pi(X1d),i∈{1,...,k} (6)
Step 2.5, based on the 2D mapping of step 2.4, as shown in FIG. 2, performing feature learning on the 2D variable by using a convolutional neural network in deep learning;
Step 2.6, based on the feature learning in step 2.5, performing pattern recognition and classification processing on time sequence data by using the features learned by the network, wherein the deep learning network comprises a plurality of convolution layers and an activation function, and extracting complex time frequency features from the segments as shown in fig. 4;
X'1D＝Reshape(Inception(X2D)) (7)
And 3, according to the method in the step 2, the trained network is remodeled into a one-dimensional signal, and the prediction output of the model is processed through a Softmax function to obtain a final classification result.
Step 4, according to the method described in step 3, the accuracy of the invention is estimated using different estimation functions, which can be expressed as:
The invention has the following beneficial effects:
1. The classification accuracy is improved: the traditional fetal heart rate deceleration classification method is often limited by factors such as poor signal quality, and the like, so that the classification result is not accurate enough. The invention adopts a Fourier transform-based method, and obviously reduces the dependence on the quality of the original signal by extracting the top-k frequency of the signal. This innovative design allows higher classification accuracy to be achieved in the face of poor signal quality.
2. Efficiency is improved: the quality of the raw fetal heart rate signal is often poor, and previous experiments often required a filtering pre-process to ensure signal quality. Such preprocessing may lead to signal distortion, affecting the final result. The method based on Fourier transform directly acquires the frequency characteristics from the original signal, thereby reducing the distortion possibly brought by preprocessing. The processing mode can improve the processing efficiency and ensure the authenticity of the signal.
3. Better use is made of a deep learning model: conventional signal processing is typically limited to deep learning on one-dimensional inputs. By converting the fetal heart rate signal into two dimensions, the invention innovatively introduces a computer vision backbone, providing more possibilities for representation learning of the signal. In addition, the mature trunk structure in the computer vision community is selected, so that advanced experience in the field of image processing is effectively borrowed. This integration makes the deep learning model more flexible and better performing, providing new possibilities for the combination of time series analysis and computer vision.
Drawings
In order to more clearly illustrate the embodiments of the present invention, the drawings that are required to be used in the embodiments will be briefly described.
FIG. 1 is a flow chart of the overall architecture of the method of the present invention;
FIG. 2 is a schematic diagram of different decelerations;
FIG. 3 is a schematic diagram of a Fourier transform process used in the present invention;
FIG. 4 is a flow chart diagram of deep learning used in the present invention;
Detailed Description
The following describes the technical solution in the embodiment of the present invention in detail and completely with reference to the drawings in the embodiment of the present invention.
Electronic Fetal Monitoring (EFM) is a common obstetric technique in the medical arts that monitors fetal heart Gong Sutu (CTG) to record Fetal Heart Rate (FHR) and maternal Uterine Contractions (UC) signals. Currently, ultrasonic doppler transducers are widely used for FHR monitoring due to their practical and anti-interference capabilities. Physicians typically rely on experience to interpret CTG signals and decide accordingly whether to perform a medical intervention. Timely proper intervention is critical in rescuing the fetus and preventing serious consequences. Conversely, untimely interventions may be harmful to the fetus, while unnecessary interventions may increase the physical and mental and economic burden of the pregnant woman. CTG analysis monitors fetal conditions by assessing characteristics of the two main signals, fetal Heart Rate (FHR) and Uterine Contractions (UC), such as FHR baseline, range of variation, acceleration and deceleration patterns. The deceleration is a key index for judging whether the fetus is possibly lack of oxygen, and if the lack of oxygen is continuous, serious consequences such as neonatal paralysis or cerebral paralysis can be caused, medical intervention is needed in time. Deceleration can be subdivided into early deceleration, late deceleration, variable deceleration and prolonged deceleration types as in fig. 2. Before birth, CTG shows less and slight retardation, while the time-of-birth retardation is more frequent and of greater concern. Therefore, it is important to develop a method for classifying fetal heart rate deceleration based on a Fast Fourier Transform (FFT). The method can report abnormal changes in time by monitoring the CTG with the aid of a computer, and effectively improves the working efficiency of doctors and the accuracy of diagnosis. Fig. 1 is a flowchart of an overall architecture for classifying fetal heart rate deceleration based on a fast fourier transform according to the present embodiment.
The flow of this embodiment is as follows:
Step 1, receiving a signal data set S of an original fetal heart rate time sequence extracted by Doppler ultrasound;
Step2, based on the fetal heart rate data set S obtained in the step 1, after extracting a deceleration segment, intercepting the deceleration segment according to the length of 1200, and if the length exceeds 1200, filling zero into a preset continuous segment set S';
Step 2.1, constructing a problem set S' according to the deceleration segment, and transforming the fetal heart rate deceleration segment by using Fourier transform, and outputting an amplitude set { f 1,...,fn-1,fn } according to the result;
Step 2.2, calculating the period information p i of the amplitude of the signal by dividing the time step number of the signal by each frequency component;
Step 2.3, if the period set { p 1,...,pn-1,pn } is not proportional to the signal sequence length 1200, performing zero expansion to the least common multiple of the period;
Step 2.4, calculating to obtain an average amplitude set, obtaining the maximum k amplitudes according to the top-k, obtaining the top-k amplitude set, and remolding an original one-dimensional signal into a two-dimensional map in a stacking mode for feature extraction according to the top-k amplitude set when a problem is selected;
Step 2.5, based on the two-dimensional mapping in step 2.4, as shown in fig. 4, the deep learning network comprises a plurality of convolution layers and an activation function, and the TimeBlock module is used for learning features of the remodeled two-dimensional mapping, and the main structure of the deep learning network comprises:
(1) The first layer is a convolutional layer, which receives the original input signal, using a convolutional kernel size of 1*1, 3*3, 5*5, 7*7, 9*9, 11 x 11, respectively, with a step size of 1, which is Zero-filled. Group normalization was performed using GeLU as an activation function, and a final output was a matrix with a channel number of 64.
(2) The second layer is a convolutional layer that receives the output from the first layer using a convolutional kernel size of 1*1, 3*3, 5*5, 7*7, 9*9, 11 x 11, respectively, with a step size of 1, the layer being Zero-filled, group normalized, geLU as an activation function, and a final output size of a matrix of 64 channels.
And 2.6, based on the feature learning in the step 2.5, performing pattern recognition and classification processing on time sequence data by utilizing the features learned by the network, extracting complex time frequency features from the fragments, wherein the main flow is to splice the results obtained after TimeBlock processing by using residual connection, and then performing layer normalization processing on each layer by using LayerNorm.
Step 3, according to the method in the step 2, the trained network is remodeled into a one-dimensional signal, the prediction output of the model is processed through a Softmax function, and a final classification result is obtained, wherein the main flow is that global and local information are connected by using a full connection layer, and then the result is predicted through the Softmax function;
Step 4, according to the method in the step 3, using different evaluation functions to evaluate the accuracy of the invention;
Through the above steps, the deceleration classification method based on fourier changes can play a role in the fetal state assessment problem at the time of actual birth. The method utilizes top-k frequency to extract fetal heart rate signal characteristics, can utilize a model in the field of vision, improves classification quality and efficiency, and further improves working efficiency of doctors.
The foregoing has shown and described the basic principles, features and embodiments of the invention. The scope of the invention is not limited thereto, and it should be understood by those skilled in the art that the present invention and equivalents and modifications thereof are intended to be included within the scope of the present invention.
Claims (9)
1. A method for classifying fetal heart rate deceleration based on fast fourier transform, which is characterized in that a signal data set S containing a Fetal Heart Rate (FHR) and Uterine Contraction (UC) signal time sequence is received, and the data set is classified after being subjected to fast fourier transform and then is subjected to feature extraction by using a convolutional neural network.
2. Method for classifying a fetal heart rate deceleration based on a fast fourier transform according to claim 1, wherein consecutive segments S' of preset length L are truncated or zero-padded according to the accepted data set S to normalize the input data and to make further analysis.
3. Method for classifying a fetal heart rate deceleration based on a fast fourier transform according to claim 2, characterized in that a fourier transform is applied to the successive segments S' for a spectral analysis to extract the signal amplitude F (w) for identifying the characteristic frequency of the fetal heart rate deceleration time.
4. A method of classifying fetal heart rate deceleration based on a fast fourier transform as claimed in claim 3, wherein the significant period of the amplitude information F (w) is further identified and the segment expansion is performed when the significant period is not an integer ratio of segment length, ensuring the integrity of the periodic features.
5. The method for classifying fetal heart rate deceleration based on the fast fourier transform as claimed in claim 4, wherein the segment S' is length-adjusted to coincide with the least common multiple of the salient period.
6. The method for classifying fetal heart rate deceleration based on a fast fourier transform as recited in claim 5, wherein k significant frequencies are selected from the set of amplitudes and the one-dimensional signal is reconstructed as a two-dimensional matrix.
7. The method for classifying fetal heart rate deceleration based on fast fourier transform as recited in claim 6, wherein a convolutional neural network is used to perform deep feature learning on the two-dimensional matrix, and fine time-frequency features of fetal heart rate deceleration are extracted.
8. The method of classifying fetal heart rate deceleration based on a fast fourier transform as recited in claim 7, wherein the extracted features are processed using a Softmax function to achieve a classified prediction of fetal heart rate deceleration.
9. The method for classifying fetal heart rate deceleration based on the fast fourier transform as recited in claim 8, wherein the prediction accuracy is evaluated using different evaluation functions according to the output of the deep learning network, so as to ensure the effectiveness and reliability of the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410133749.7A CN117932472A (en) | 2024-01-31 | 2024-01-31 | Method for classifying fetal heart rate deceleration based on fast Fourier transform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410133749.7A CN117932472A (en) | 2024-01-31 | 2024-01-31 | Method for classifying fetal heart rate deceleration based on fast Fourier transform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117932472A true CN117932472A (en) | 2024-04-26 |
Family
ID=90750738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410133749.7A Pending CN117932472A (en) | 2024-01-31 | 2024-01-31 | Method for classifying fetal heart rate deceleration based on fast Fourier transform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117932472A (en) |
-
2024
- 2024-01-31 CN CN202410133749.7A patent/CN117932472A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jager et al. | Characterization and automatic classification of preterm and term uterine records | |
| Fergus et al. | Machine learning ensemble modelling to classify caesarean section and vaginal delivery types using Cardiotocography traces | |
| Mischi et al. | Dedicated entropy measures for early assessment of pregnancy progression from single-channel electrohysterography | |
| Romano et al. | Software for computerised analysis of cardiotocographic traces | |
| Lee et al. | Fetal QRS detection based on convolutional neural networks in noninvasive fetal electrocardiogram | |
| CN101972145A (en) | Fetus electrocardio blind separation method based on relative sparsity of time domain of source signal | |
| Xu et al. | Review on EHG signal analysis and its application in preterm diagnosis | |
| Taylor et al. | The development and validation of an algorithm for real‐time computerised fetal heart rate monitoring in labour | |
| WO2004072822A2 (en) | Quantitative fetal heart rate and cardiotocographic monitoring system and related method thereof | |
| Lu et al. | Estimation of the foetal heart rate baseline based on singular spectrum analysis and empirical mode decomposition | |
| Zarmehri et al. | On the prediction of foetal acidaemia: a spectral analysis-based approach | |
| Shahrdad et al. | Detection of preterm labor by partitioning and clustering the EHG signal | |
| Rotariu et al. | Spectral analysis of fetal heart rate variability associated with fetal acidosis and base deficit values | |
| Subramaniam et al. | A review of significant researches on prediction of preterm birth using uterine electromyogram signal | |
| Anisha et al. | Detection of fetal cardiac anomaly from composite abdominal electrocardiogram | |
| CN113768519A (en) | Method for analyzing consciousness level of patient based on deep learning and resting state electroencephalogram data | |
| Romagnoli et al. | Digital cardiotocography: What is the optimal sampling frequency? | |
| Zhong et al. | CTGNet: Automatic analysis of fetal heart rate from cardiotocograph using artificial intelligence | |
| CN117932472A (en) | Method for classifying fetal heart rate deceleration based on fast Fourier transform | |
| Das et al. | A novel approach for extraction and analysis of variability of baseline | |
| Chen et al. | Toward automated analysis of fetal phonocardiograms: Comparing heartbeat detection from fetal doppler and digital stethoscope signals | |
| Kimura et al. | Time-frequency analysis of fetal heartbeat fluctuation using wavelet transform | |
| Fergus et al. | Prediction of intrapartum hypoxia from cardiotocography data using machine learning | |
| Tian et al. | A comprehensive evaluation of the predictive abilities of fetal electrocardiogram-derived parameters during labor in newborn acidemia: our institutional experience | |
| CN113133771B (en) | Uterine electromyographic signal analysis and early birth prediction method based on time-frequency domain entropy characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |