• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    In the present invention a methodology is proposed to analyze the functional tremors of people, using a classification system to establish to what degree, said tremors may or may not constitute Parkinson's disease. The methodology is based on capturing data in an aerial way based on the drawing of the Archimedes spiral. In this way, a comparison can be established between three-dimensional charts of healthy patients and patients with tremors, which may or may not be due to Parkinson's symptoms. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2684568A1
    申请号:ES201700503
    申请日:2017-03-30
    公开日:2018-10-03
    发明作者:Carlos Manuel TRAVIESO GONZÁLEZ；Ciro Ángel GARCÍA MERINO
    申请人:Universidad de las Palmas de Gran Canaria；
    IPC主号:
    专利说明:

    5
    10
    fifteen
    twenty
    25
    description
    Methodology for the diagnosis of Parkinson's disease, using three-dimensional spirals
    Object of the invention
    The present invention relates to a method of performing analysis of functional tremors that people may suffer. For this, a method based on data collection is applied using sensors for detecting volumetric activity, thus capturing information in three dimensions. With this three-dimensional data acquisition when performing the Archimedes spiral, the tremors will be obtained naturally, without any element influencing them, such as footholds, when done on paper or a tablet.
    From these data obtained, it can be determined whether or not this tremor belongs to Parkinson's disease, thanks to the use of a classification system based on Euclidean distances that will calculate the degree of error against an ideal Archimedes spiral.
    Background of the invention
    Parkinson's disease being the second neurodegenerative disease that most affects people, after Alzheimer's, in terms of techniques for the detection of it, is with a very small number of work done in this field.
    There is currently a test project with some similarity to the one proposed. It is a system that uses, as in this work, the Archimedes spiral, in two dimensions, using data acquisition through an electronic device.
    On the other hand, a development carried out by professionals from several hospitals, tries to measure the motor ability of people, based on the temporal analysis of the click. It consists of a solution that measures the keystroke and release time during normal use of a computer and converts it into an engine index. To do this, it uses automatic pattern detection in the time series using set regression algorithms.
    i) https://www.neuroqwerty.com/es/
    5
    10
    fifteen
    twenty
    25
    As for the analysis of Voice quality, there are works that allow the study of tremors, produced by neurodegenerative diseases. One of these cases is software developed by the Vithas Nuestra Señora de América hospital and the Biomedicine Technological Center of the Polytechnic University of Madrid, which provide information on Glotic Closure Defect and Asymmetric Vibration of the Vocal Folds.
    ii) Pedro Gómez, Victoria Rodellar, Víctor Nieto, Rafael Martínez, Agustín Álvarez, Bartolomé Scola, Carlos Ramírez, Daniel Poletti1, Mario Fernández, "BioMet®Phon: A System to Monitor Phonation Quality in the Clinics"
    On the other hand, there are currently several medical tests, in the clinical setting, to detect Parkinson's disease:
    • Brain SPECT. Cerebral SPECT is a scintigraphic test, in which a small amount of radioactivity is used to obtain brain images, with which partial seizures can be detected.
    • Brain PET: It is a technology within the specialty of nuclear medicine. It is based on detecting and analyzing the distribution adopted by a radiopharmaceutical within the brain.
    • Transcranial ultrasound: a medical test that stimulates neuronal activity using ultrasound.
    This review of the state of the art, allows to visualize the works and the ways of work carried out so far, and demonstrates the innovation of the proposal, when working with three-dimensional information when obtaining the information of the possible users of the system.
    Summary of the invention
    The present invention relates to a method of performing analysis that allows non-invasively measure the tremors that some people may have, to assimilate
    5
    10
    fifteen
    twenty
    25
    said tremor to that caused by Parkinson's disease. This method is implemented following the following five steps:
    i) Obtaining the data through the use of sensors for detecting volumetric activity, in order to perform a non-invasive test of the representation of the Archimedes spiral.
    ii) Pre-processed data obtained, in order to avoid factors such as spatial translation, scalability or rotation.
    iii) Comparison of the 3D spirals obtained against the ideal Archimedes spiral, in order to calculate the errors between both representations.
    iv) Definition of the thresholds by which a tremor is considered as Parkinson's tremor, or functional tremor characteristic of the person or other type of disease to perform the Archimedes spiral.
    v) Obtain a result on a numerical scale, which identifies the tremor to a healthy person, or characteristic of Parkinson's disease by implementing a three-dimensional representation of the Archimedes spiral.
    Description of the figures
    Figure 1 shows in a block diagram the five steps that make up this method. The first step is the acquisition of data, which is done non-invasively through sensors. It is followed by preprocessing, where the data is transformed to avoid factors such as scaling, rotation or translation. Once the data is pre-processed, a classifying algorithm is applied, in this case the Dynamic Time Warping, the result of which will be the errors between the acquired spirals and the standard spiral. Subsequently, the threshold values are set by which the tremors will be able to be differentiated, and because of this the result of the comparison can be interpreted
    Figure 2 shows the window used as a guide, to make the test more intuitive to the user.
    Figure 3 shows one of the tests performed on a healthy patient. In it you can see the spiral captured by the sensors and represented in the X-Y plane.
    5
    10
    fifteen
    twenty
    25
    Figure 4 shows one of the tests performed on a patient with mild tremor. In it you can see the spiral captured by the sensors and represented in the X-Y plane.
    Figure 5 shows one of the tests performed on a patient with moderate tremor. In it you can see the spiral captured by the sensors and represented in the X-Y plane.
    Figure 6 shows one of the tests performed on a healthy patient. In it you can see the spiral captured by the sensors and represented in the three dimensions.
    Figure 7 shows one of the tests performed on a patient with mild tremor. In it you can see the spiral captured by the sensors and represented in the three dimensions.
    Figure 8 shows one of the tests performed on a patient with moderate tremors. In it you can see the spiral captured by the sensors and represented in the three dimensions.
    Figure 9 shows the reference spiral, the ideal Archimedes spiral.
    Figure 10 shows the ideal Archimedes spiral in three dimensions.
    Figure 11 is shown, as the image obtained at the output of the Dynamic Time Warping algorithm, when comparing a three-dimensional time series, from a healthy person, highlighting the optimal deformation path in red
    Figure 12 is shown, as the image obtained at the output of the Dynamic Time Warping algorithm, when comparing a three-dimensional time series, from a person with moderate tremors, highlighting in red the optimal deformation path
    Detailed description of a preferred embodiment of the invention
    The proposed invention consists of a method of carrying out analysis that is capable of capturing functional tremors through non-invasive sensors, in three dimensions when performing the Archimedes spiral, for a subsequent classification based on error study, as indicated in figure 1.
    The first step is the acquisition of the data that make up the spiral of Archimedes in three dimensions. For this purpose, a drawing has been created that contains the spiral to be drawn, in order to propose a guide in the test, as shown in Figure 2.
    5
    10
    fifteen
    twenty
    25
    30
    The acquisition of data is carried out aerially, through sensors for capturing the volumetric activity, performed by the person to be analyzed; Therefore, the result obtained is a time series of three components, one for each three-dimensional position.
    Once the data of the three-dimensional Archimedes spiral to be studied is obtained, a pre-processing of these data is carried out, in order to avoid factors such as magnitude, translation or rotation. For this step, it is necessary to normalize the values obtained to make them independent of the scaling of each spiral. The normalization applied responds to the following expression:
    x n]
    ^ normalized m ^ x [n ^ iUU
    Where x [n] is one of the components of the captured time series, and n the position in the vector of the time series.
    Regarding the independence in space, a first order derivative processing is applied to each of the three axes of the time series, where the discrete information of the time series make an adaptation of the concept of derivative in functions to said discrete values, as stated in the following expression:
    dx [n]
    —-— «-» x [n + 1] - x [n] dn
    Where x [n] is one of the components of the time series captured, and n the position in the vector of the time series of the Archimedes spiral representation.
    The next step is to compare the spiral obtained by non-invasive sensors, with the ideal spiral of Archimedes. For this, a classification system such as Dynamic Time Warping is used. This algorithm measures the similarity between two time series that can vary in time or speed. Through Dynamic Time Warping, the optimal alignment between series can be found twice if a time series can be "deformed" not linearly by stretching or contraction along its time axis.
    In the application of the classification system, the error is measured as the result of the measurement of the Euclidean distance, comparing the three-dimensional time series of the Archimedes spiral captured by the sensors, and the ideal time series. This measure will be the value to be established against the decision threshold. And the threshold is set as the intermediate value between the average Euclidean distances of healthy people and the average of people with Parkinson's disease.
    权利要求:
    Claims (1)
    [1]
    5
    10
    fifteen
    twenty
    1.- Method of carrying out analysis of the functional tremors that people may suffer. To do this, a method based on data collection is applied by means of non-invasive sensors in the three spatial dimensions to represent the Archimedes spiral, and from these to determine whether said tremor belongs or not to Parkinson's disease, applying the derivative from first grade to the three-dimensional series captured from the representation of the spiral and subsequently, the use of a classification system based on the calculation of errors resulting from the measurement of the Euclidean distance of the three-dimensional time series of the ideal spiral against the spiral realized, one component for each dimension of the measured space. This method consists of the following steps with invention:
    i) Obtaining the three-dimensional data of the representation of the Archimedes spiral, made by a person whose tremor is to be analyzed, by using the volumetric sensor of the air activity at the time of making the stroke.
    ii) Pre-processing of the three-dimensional data obtained from the time series when making a spiral of Archimedes. It is necessary to carry out a treatment of the data obtained from the air route, in order to minimize errors caused by the difference in sizes, rotations, or acquisitions in different positions of the space. For this, a normalization of the values is carried out as follows:
    x [n]
    Xnormalized = * 100
    With this, each of the three-dimensional captures of the Archimedes spiral representation can be treated, as initiated at the same spatial point, independent of their size, and their translation. As well as the application of the first order derivative to the normalized data to highlight its variability, using the following formula:
    dx [n]
    —-— «-» x n + 1] - x [n] dn
    iii) Comparison of the derivation of the three-dimensional Archimedes spirals obtained, with the derivation of the ideal Archimedes spiral, in order to calculate the errors between them, by studying the Euclidean distance.
    iv) Definition of the thresholds by which a time series captured by the sensors
    5 when performing the Archimedes spiral, it is considered Parkinson's tremor, or
    functional tremor of the person or another type of disease. This threshold is established as the intermediate value between the average Euclidean distances of healthy people and the average of people with Parkinson's disease.
    v) Obtain a result on a numerical scale, which identifies that the spiral of
    10 Three-dimensional archimedes captured by the sensors, and including the threshold
    established, so you can locate the average Euclidean distance of a person and see if he is healthy or has Parkinson's disease; besides observing the proximity to the threshold.
    image 1
    image2
    Figure 1
    5
    image3
    Figure 2
    image4
    image5
    image6
    Figure 5
    image7
    image8
    Figure 7
    image9
    image10
    Figure 9
    image11
    image12
    Figure 11
    image13
    类似技术:
    公开号 | 公开日 | 专利标题
    ES2673117T3|2018-06-19|Breathing data analysis
    KR101492803B1|2015-02-12|Apparatus and method for breast tumor detection using tactile and near infrared hybrid imaging
    KR100979506B1|2010-09-02|Diagnosis device of Sasang constitution
    RU2392855C1|2010-06-27|Method of digital diagnostics of vertebral deformations
    ES2684568B1|2019-09-09|Methodology for the diagnosis of Parkinson's disease, using three-dimensional spirals
    CN110772255A|2020-02-11|Method for measuring human body scoliosis angle based on posture and position sensor
    Johnson2014|Detection of Parkinson disease rest tremor
    Zhou et al.2018|Development of the circumduction metric for identification of cervical motion impairment
    KR101309013B1|2013-09-17|Diagnosis Device For Face Form Using Facial Image and Cephalometric Image
    JP6703323B2|2020-06-03|ROI setting technology for biological image inspection
    KR20160057024A|2016-05-23|Markerless 3D Object Tracking Apparatus and Method therefor
    Kutilek et al.2010|Comparison of methods of measurement of head position in neurological practice
    TWI616183B|2018-03-01|Non-invasive skin image detection method
    EP3376414B1|2021-02-24|Joint movement detection system and method, and dynamic assessment method and system for knee joint
    Rosado et al.2015|A novel framework for supervised mobile assessment and risk triage of skin lesions
    De Luca et al.2018|Towards fully instrumented and automated assessment of motor function tests
    Gabriel et al.2018|Development and clinical application of Vertebral Metrics: using a stereo vision system to assess the spine
    KR102248251B1|2021-05-04|Medical diagnosis support apparatus using body heat information
    ES2725927B2|2020-11-26|SYSTEM FOR OBTAINING USEFUL DATA FOR ANALYSIS OF BODY MORPHOMETRY AND ASSOCIATED METHOD
    TWI681755B|2020-01-11|System and method for measuring scoliosis
    RU2633330C1|2017-10-11|Method for determination of inclination of lower jaw condylar process on jaws tomograms
    KR102222059B1|2021-03-03|Object position recognition diagnostic device that outputs the object position and the diagnosis result using a diagnostic device with marker
    ROVINI2018|Study and development of novel diagnostic and therapeutic approaches for Parkinson's disease based on ICT and machine learning
    US10258237B2|2019-04-16|Photobiomedical measurement apparatus
    BR102019015290A2|2021-02-02|METHOD FOR MONITORING AND ANALYZING MOVEMENTS OF JOINTS AND BODY PARTS
    同族专利:
    公开号 | 公开日
    ES2684568B1|2019-09-09|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20080045804A1|2005-05-02|2008-02-21|Williams Mark E|Systems, devices, and methods for interpreting movement|
    WO2012098388A1|2011-01-18|2012-07-26|The University Of York|Signal processing method and apparatus|
    法律状态:
    2018-10-03| BA2A| Patent application published|Ref document number: 2684568 Country of ref document: ES Kind code of ref document: A1 Effective date: 20181003 |
    2019-09-09| FG2A| Definitive protection|Ref document number: 2684568 Country of ref document: ES Kind code of ref document: B1 Effective date: 20190909 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201700503A|ES2684568B1|2017-03-30|2017-03-30|Methodology for the diagnosis of Parkinson's disease, using three-dimensional spirals|ES201700503A| ES2684568B1|2017-03-30|2017-03-30|Methodology for the diagnosis of Parkinson's disease, using three-dimensional spirals|
    [返回顶部]