TR202022620A2 - ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE - Google Patents
ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE Download PDFInfo
- Publication number
- TR202022620A2 TR202022620A2 TR2020/22620A TR202022620A TR202022620A2 TR 202022620 A2 TR202022620 A2 TR 202022620A2 TR 2020/22620 A TR2020/22620 A TR 2020/22620A TR 202022620 A TR202022620 A TR 202022620A TR 202022620 A2 TR202022620 A2 TR 202022620A2
- Authority
- TR
- Turkey
- Prior art keywords
- impedance
- measurement
- ann
- stage
- layer
- Prior art date
Links
- 238000002847 impedance measurement Methods 0.000 title claims abstract description 16
- 238000005259 measurement Methods 0.000 claims abstract description 24
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 238000012805 post-processing Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 13
- 210000004027 cell Anatomy 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000000034 method Methods 0.000 description 5
- 239000000463 material Substances 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- HVYWMOMLDIMFJA-DPAQBDIFSA-N cholesterol Chemical compound C1C=C2C[C@@H](O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2 HVYWMOMLDIMFJA-DPAQBDIFSA-N 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000012010 growth Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 208000000453 Skin Neoplasms Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 238000004113 cell culture Methods 0.000 description 1
- 230000010261 cell growth Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 235000012000 cholesterol Nutrition 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000005094 fruit set Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000002489 impedance cardiography Methods 0.000 description 1
- 238000001566 impedance spectroscopy Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 201000000849 skin cancer Diseases 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/053—Measuring electrical impedance or conductance of a portion of the body
- A61B5/0531—Measuring skin impedance
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2560/00—Constructional details of operational features of apparatus; Accessories for medical measuring apparatus
- A61B2560/02—Operational features
- A61B2560/0223—Operational features of calibration, e.g. protocols for calibrating sensors
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Theoretical Computer Science (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Signal Processing (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Psychiatry (AREA)
- Physiology (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Power Engineering (AREA)
- Radiology & Medical Imaging (AREA)
- Probability & Statistics with Applications (AREA)
- Dermatology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Fuzzy Systems (AREA)
- Measurement Of Resistance Or Impedance (AREA)
Abstract
Buluş, farklı değer aralıklarında ölçümler yapabilmek için farklı değerlerde kalibrasyon empedanslarına ihtiyaç duyan AD5933 entegre devresinin, kalibrasyon ihtiyacının üstesinden gelmek amacıyla yapay sinir ağı tabanlı sinyal işleme sonrası algoritması ile ilgilidir. Buluşta, sabit bir kalibrasyon empedansı kullanılarak yüksek hassasiyetli empedans ölçümü için uygulamaya özel bir yapay sinir ağı topolojisi geliştirilmiş ve eğitilmiştir. Ölçüm sistemi, nominal cilt empedansı aralığında çalışmak üzere tasarlanmıştır.The invention relates to an artificial neural network-based post-processing algorithm in order to overcome the need for calibration of the AD5933 integrated circuit, which requires different calibration impedances in order to make measurements in different value ranges. In the invention, an application-specific artificial neural network topology has been developed and trained for high precision impedance measurement using a fixed calibration impedance. The measuring system is designed to operate within the nominal skin impedance range.
Description
TARIFNAME CILT EMPEDANS ARALIGI IÇIN YSA TABANLI TEK KALIBRASYONLU EMPEDANS ÖLÇÜM SISTEMI TEKNIK ALAN Bulus; farkli deger araliklarinda ölçümler yapabilmek için farkli degerlerde kalibrasyon empedanslarina ihtiyaç duyan AD5933 entegre devresinin, kalibrasyon ihtiyacinin üstesinden gelmek amaciyla yapay sinir agi tabanli sinyal isleme sonrasi algoritmasi ile ilgilidir. Bulusta, sabit bir kalibrasyon empedansi kullanilarak yüksek hassasiyetli empedans ölçümü için uygulamaya özel bir yapay sinir agi topolojisi gelistirilmis ve egitilmistir. Ölçüm sistemi, nominal cilt empedansi araliginda çalismak üzere tasarlanmistir. ÖNCEKI TEKNIK Yüksek hassasiyetli empedans ölçümü, biyolojik hücre analizinden yakit hücresi testine kadar çok yaygin olarak kullanilan bir yöntemdir. Empedans ölçümü, hücrelerin, dokularin ve materyallerin elektriksel karakterizasyonu için basit, kullanisli ve güçlü bir yöntemdir. DESCRIPTION YSA-BASED SINGLE CALIBRATION FOR SKIN IMPEDANCE RANGE IMPEDANCE MEASUREMENT SYSTEM TECHNICAL FIELD Meet; in different values in order to make measurements in different value ranges. Calibration of AD5933 integrated circuit, which needs calibration impedances, artificial neural network-based signal processing post-processing to overcome the need related to the algorithm. In the invention, using a fixed calibration impedance, high An application-specific neural network topology for precision impedance measurement developed and trained. The measuring system is designed to operate within the nominal skin impedance range. is designed to. PRIOR ART High precision impedance measurement, fuel cell from biological cell analysis It is a widely used method until testing. Impedance measurement, cells, A simple, convenient and powerful tool for electrical characterization of tissues and materials. method.
Genis bir frekans araliginda kararli ve yüksek hassasiyetli empedans ölçümü karmasik bir islem oldugundan, geleneksel olarak kullanilan cihazlar hantal ve pahalidir. Digertaraftan, cilt kanseri gibi bazi durumlarda, cilt empedansinin sürekli ve düzenli olarak izlenmesi tedavi etkinligini artirmaktadir. Günlük ve yüksek hassasiyetli empedans ölçümü, hastaligin seyri ve anormal dokunun fizyolojisi hakkinda bilgiler içerir. Bu durumda basit, kompakt ve düsük maliyetli bir empedans ölçüm sistemiyle dokularin düzenli olarak takibi, hastalarin kendileri tarafindan kullanilabilir. Stable and high precision impedance measurement over a wide frequency range Since it is a complex process, traditionally used devices are cumbersome and it is expensive. On the other hand, in some cases, such as skin cancer, the skin impedance is constantly and Regular monitoring increases the effectiveness of treatment. everyday and high precision information about impedance measurement, course of disease, and physiology of abnormal tissue includes. In this case, with a simple, compact and low-cost impedance measuring system regular monitoring of tissues can be used by the patients themselves.
Yari iletken teknolojisindeki ve gömülü sistemlerdeki gelismeler, daha küçük, daha ucuz ve tasinabilir tek çipli empedans ölçüm cihazlari üretmeyi mümkün kilmistir. Advances in semiconductor technology and embedded systems have resulted in smaller, It has made it possible to produce cheaper and portable single-chip impedance measuring devices.
Yüksek hassasiyetli, tek çipli ag analizörü olan AD ve dijital sinyal isleme (DSP) yeteneklerine sahip böyle bir empedans ölçüm entegre devresidir. Literatürde, empedans kardiyografisi, biyoempedans ölçümü, tek hücre ölçümü, hücre kültürü büyümesi izleme, biyosensörler ve teknik nesne teshisi gibi çesitli arastirma türlerinde AD5933 entegre devresi kullanilmaktadir. Snajdrova ve ark. ve Shlyakhotka ve ark. çalismalarinda, kardiyak output ve strok hacmini degerlendirmek için AD5933 entegre devresi tabanli empedans kardiyografi analizörleri gelistirmislerdir. Her ikisi de basit birer yapiya sahip olan cihazlar, sürekli ölçümler elde etmek için uygundur ve kabul edilebilir sonuçlar saglamaktadir. Seoane ve ark., biyomedikal uygulamalar için dört elektrotlu bir topolojiye sahip olan ve AD5933 entegre devresini kullanan bir analog devre sunmustur. Bu cihaz, dogru empedans ölçüm sonuçlari ve dinamik yük araligi açisindan iyi bir performans göstermistir. Chen ve ark., L929 hücrelerinin büyümesini izlemek için tasinabilir bir empedans ölçüm sistemi tasarlamistir. Önerilen bu cihaz, düsük hücre konsantrasyonlarinda hücre büyümesini gözlemlemek için bir mikroislemci, AD5933 entegre devresi ve iç içe geçmis mikroelektrotlardan olusmaktadir. Diger bir çalismada, el tipi empedans spektroskopi uygulamalari için kullanimi kolay, düsük maliyetli ve sekiz kanalli bir empedans analizörü gelistirilmistir. Biyomedikal uygulamalar için bu cihaz, bagimsiz ve tasinabilir olmasi nedeniyle bir öncü olmustur. Hoja ve Lentka, 10 mHz ile 100 kHz frekans araliginda 10 O ve 10 GO arasindaki empedansin ölçülmesine izin veren telemetrik bir analizör sunmustur. Bu çalismada, AD5933 entegre devresi bir empedans dönüstürücü olarak kullanilmistir. AD5933 entegre devresi tabanli biyo-empedans ölçüm sistemleri literatürde yaygin olarak bulunmaktadir. Margo ve ark. AD5933 için yeni bir ek devre sunmustur. Devre, elektrot ve cilt arasinda olusan artefaktlari azaltmak için dört elektrot kullanimina olanak saglamaktadir. Hui ve Ding'in çalismasinda, düsük güçlü ve tasinabilir bir biyoelektrik empedans analizörü sunulmustur. Ferreira ve ark., transtorasik elektriksel biyoimpedans ölçüm kaydi için analog bir devre eklentisi ile AD5933 entegre devresini kullanmistir. Yang ve ark., farkli insanlari ayirt edebilmek için bir avuç içi biyoimpedans spektroskopi sistemi önermistir. Literatürde “cole parametresi” tahminlerine veya cihaz tasinabilirligine odaklanan biyoimpedans ölçüm sistemleri de mevcuttur. Bu sistemlerin kontrolü, ARM? gibi yaygin olarak kullanilan mikro denetleyiciler tarafindan yapilmistir. AD, a high-precision, single-chip network analyzer and integrated such an impedance measurement with digital signal processing (DSP) capabilities. is the circuit. In the literature, impedance cardiography, bioimpedance measurement, single cell measurement, cell culture growth monitoring, biosensors and technical object detection. AD5933 integrated circuit is used in various types of research. Snajdrova et al. and Shlyakhotka et al. studies, cardiac output and stroke volume AD5933 integrated circuit based impedance cardiograph to evaluate developed the analyzers. Devices, both of which have a simple structure, are constantly It is suitable for obtaining measurements and provides acceptable results. Seoane et al., which has a four-electrode topology for biomedical applications and He presented an analog circuit using the AD5933 integrated circuit. This device is correct Good performance in terms of impedance measurement results and dynamic load range. has shown. Chen et al. use a portable tool to monitor the growth of L929 cells. Impedance measurement system has been designed. This recommended device, low cell A microprocessor for observing cell growth at concentrations of AD5933 It consists of integrated circuit and nested microelectrodes. In another study, It is easy to use, low cost, and suitable for handheld impedance spectroscopy applications. An eight-channel impedance analyzer has been developed. For biomedical applications this The device has been a pioneer because it is independent and portable. Hoja and Lentka, 10 Impedance between 10 O and 10 GO in the frequency range mHz to 100 kHz. presented a telemetric analyzer that allows the measurement of In this study, AD5933 The integrated circuit is used as an impedance converter. AD5933 integrated circuit-based bio-impedance measurement systems are widely used in the literature. are available. Margo et al. Presented a new additional circuit for AD5933. circuit, electrode It allows the use of four electrodes to reduce artifacts between the skin and the skin. it provides. In Hui and Ding's work, a low-power and portable bioelectric impedance analyzer is presented. Ferreira et al., transthoracic electrical AD5933 integrated circuit with an analog circuit plug-in for bioimpedance measurement recording has used. Yang et al., a palm bioimpedance to distinguish different people spectroscopy system. In the literature, “cole parameter” estimations or device There are also bioimpedance measurement systems that focus on portability. These systems control, ARM? It is made by widely used microcontrollers such as
AD5933 entegre devresi tabanli empedans ölçümü, meyve yaslanmasi ve olgunluk tespiti üzerine de uygulanmistir. Görüldügü gibi, AD5933 entegre devresi birçok farkli alanlardaki uygulamalarda kullanilabilmektedir. Bu anlamda AD5933 entegre devresi, pH sensöründen, kolesterol konsantrasyonunun ölçümüne, hava-su kalitesi izleme sisteminden, dijital LCR metrenin gerçeklestirilmesine, girdap akimi testinden ve AC sifir-potansiyel devrelerine kadar çok farkli alanlarda kendine yer bulmustur. AD5933 integrated circuit based impedance measurement, fruit set and maturity applied to detection. As can be seen, the AD5933 integrated circuit has many different can be used in applications in the field. In this sense, the AD5933 integrated circuit, From the pH sensor to the measurement of cholesterol concentration, air-water quality monitoring system, the realization of the digital LCR meter, the eddy current test and the AC It has found a place for itself in many different fields, from zero-potential circuits.
AD5933 entegre devresi ile birçok basarili uygulama ve çalisma olmasina ragmen, en önemli dezavantaji dogrusal olmayan kalibrasyon gereksinimi ve kalibrasyon etkilerinden kaynaklanan ölçüm hatasidir. Ölçüm bilgilerinin DDS'den DFT modülüne dogrudan dijital olarak aktarimi, ölçüm sonuçlarinda problem olusturmaktadir. Bu nedenle, kazanç faktörünü elde etmek için her ölçümden önce bir kalibrasyon adimi gerçeklestirilmelidir. Kazanç faktörü, degeri bilinmeyen, ölçülecek birimin empedansini belirlemek için kullanilir. Kalibrasyon direncinin degeri, bilinmeyen empedansin beklenen araligina bagli olarak degisir. Teorik olarak, bir malzemenin empedansi ölçülürken, o malzemenin gerçek empedans degerine daha yakin kalibrasyon direnci kullanmak, daha dogru ölçüm saglar. Chen ve ark., mikro rölelerden ve yüksek hassasiyetli dirençlerden olusan bir anahtarlama dizisi kullanarak kalibrasyon sorununun üstesinden gelmeye çalismistir. Ölçüm sistemi kalibrasyonu, uyarlanabilir geri besleme yaklasimiyla iyilestirilmesine ragmen, röle ve direnç dizileri sistemin karmasikligini artirmaktadir. Simic çalismasinda, uygun kalibrasyon direnci seçimi için iki analog çoklayici kullanarak kendi kendine kalibrasyon yapan bir sistem önermistir. Sistemin kalibrasyonunda kullanilan hassas dirençlerin sinirliligi nedeniyle yalnizca düsük frekanslarda dogru ölçümler gözlenmistir. Benzer sekilde, Abo Bakr ve ark. ölçüm birimini farkli frekans araliklari için paralel RC aglari kullanarak kalibre ederken, Breniuc ve ark., kalibrasyon direnci seçimi için bir çoklayici kullanmistir. Although there are many successful applications and studies with the AD5933 integrated circuit, However, its most important disadvantage is the need for non-linear calibration and is the measurement error caused by the calibration effects. DFT of measurement information from DDS digital transfer directly to the module, problem in measurement results forms. Therefore, before each measurement to obtain the gain factor, a calibration step must be performed. Gain factor, value unknown, to be measured Used to determine the impedance of the unit. The value of the calibration resistor, depends on the expected range of unknown impedance. Theoretically, a When measuring the impedance of the material, it is more important to the actual impedance value of that material. Using a close calibration resistor provides a more accurate measurement. Chen et al., micro using a switching array of relays and high-precision resistors tried to overcome the calibration problem. Measurement system calibration, Although improved by adaptive feedback approach, relay and resistor arrays increases the complexity of the system. In simic operation, suitable calibration resistor A self-calibrating system using two analog multiplexers for selection has suggested. Due to the limitation of sensitive resistors used in the calibration of the system only accurate measurements at low frequencies have been observed. Similarly, Abo Bakr and arc. calibrate the unit of measurement for different frequency ranges using parallel RC networks. while Breniuc et al. used a multiplexer for calibration resistor selection.
Genis bir deger araliginda, ölçülecek olan her bir test malzemesinin empedans degeriyle yakin bir degerde kalibrasyon direnci kullanilamayacagindan, ölçüm hatalari sürekli olarak varligini sürdürecektir. The impedance of each test material to be measured over a wide range of values Since the calibration resistor cannot be used with a value close to its value, measurement errors may occur. will continue to exist continuously.
BULUSUN KISA AÇIKLAMASI Yapay sinir agi (YSA) tabanli bu özgün ölçüm sistemi sayesinde, önceki tekniklerin aksine tek bir sabit kalibrasyon direnci kullanilarak, AD5933 entegre devresi ile yüksek hassasiyetli empedans ölçümleri elde etmek mümkün olacaktir. Literatürde AD5933 entegre devresi kalibrasyon probleminin üstesinden gelmek ve ölçüm dogrulugunu artirmak için böyle bir çalisma bulunmamaktadir. Önerilen sistemin bir diger avantaji, genis bir frekans araliginda yüksek ölçüm dogrulugudur. Yeni bir sinyal isleme sonrasi algoritmasi olan bulus, AD5933 entegre devresiden elde edilen gerçek (Re) ve sanal (Im) verileri kullanir. AD5933 entegre devresinin farkli frekanslardaki ölçüm çiktilari, yapay sinir aginin girdileri olarak alinir. Bu veriler ile üstel biçimde Öklid uzakliginin hesaplanmasi, toplama ve normallestirme adimlarinin ardindan, çok katmanli algilayici yapay sinir agi (MLPNN) yürütülür. Sinir agi egitiminin ardindan test ölçümleri yapilir. Test sonuçlari, ölçüm sisteminin verimliligini kanitlamistir. BRIEF DESCRIPTION OF THE INVENTION Thanks to this unique measurement system based on artificial neural network (ANN), the previous By using a single fixed calibration resistor, the AD5933 integrated circuit It will be possible to obtain high precision impedance measurements with in the literature Overcoming AD5933 integrated circuit calibration problem and measuring There is no such study to increase its accuracy. One of the proposed system another advantage is high measurement accuracy over a wide frequency range. a new signal The invention, which is the post-processing algorithm, is the real data obtained from the AD5933 integrated circuit. It uses (Re) and virtual (Im) data. of the AD5933 integrated circuit at different frequencies. measurement outputs are taken as inputs of the artificial neural network. Euclidean in exponential form with these data Calculation of the distance, addition and normalization steps, layered perceptron artificial neural network (MLPNN) is executed. Testing after neural network training measurements are made. The test results have proven the efficiency of the measuring system.
SEKILLERIN ANLAMI Sekil 1. Egitim algoritmasinin akis semasi Sekil 2. Bulusun veri toplama ve empedans ölçümünün akis semasi Sekil 3. Bulusun test platformu devresi Sekil 4. Önerilen ölçüm sisteminin baski devresi (PCB) BULUSUN DETAYLI AÇIKLAMASI Yüksek dogrulukla empedans degerini tahmin etmek için gelistirilen Yapay Sinir Aglari (YSA) tabanli bulus, özgün bir YSA topolojisi içerir. Topoloji iki ana bölümden olusur. Ilk bölüm, radyaI-bazIi-fonksiyon (RBF) tabanli regresyonu gerçeklestirirken; ikinci bölüm, hata-geri yayilimli ögrenmeli bir çok katmanli algilayici sinir agi (MLPNN)'dir. Ilk bölümde, Öklid uzakligi hesaplama katmanini toplama ve normallestirme katmanlari izlemektedir. Tek bir frekans için tüm gerçek ve sanal ölçüm verileri Öklid mesafe hesaplama katmanindaki iki belirli nörona uygulanir. Ayrica, bir frekanstaki ölçüm için bir toplama ve normallestirme katmani nöronu atanir. Üstel aktivasyonlu öklid uzakligi hesaplama nöronlari, her bir frekansin gerçek ve sanal girislerini iliskilendirmek için agirlik degerlerine de sahiptir. En küçük kareler (LS) algoritmasi, bu katmanin agirlik ve üslü çarpan terimlerini hesaplamak için kullanilir. MEANING OF SHAPES Figure 1. Flowchart of the training algorithm Figure 2. Flowchart of data collection and impedance measurement of the invention Figure 3. Test platform circuit of the invention Figure 4. Printed circuit (PCB) of the proposed measuring system DETAILED DESCRIPTION OF THE INVENTION Artificial Neural developed to estimate impedance value with high accuracy. Networks (ANN) based invention includes a unique ANN topology. Topology consists of two main parts. occurs. While the first section performs the radial-based-function (RBF)-based regression; The second part is a multilayer perceptron neural network with error-backpropagation learning. (MLPNN). In the first section, summing the Euclidean distance calculation layer and followed by normalization layers. All real and virtual measurement for a single frequency data is applied to two specific neurons in the Euclidean distance computation layer. Also, a An addition and normalization layer neuron is assigned for the measurement in frequency. exponential activated euclidean distance calculation neurons, real and virtual It also has weight values to associate its inputs. least squares (LS) algorithm is used to calculate the weight and exponential factor terms of this layer.
Topolojinin ikinci bölümünde, girdilerden çiktilara eslestirilmesinin gerçeklestirilmesi için korelasyon asamasi yer alir. MLPNN'nin geri yayilim ögrenme algoritmasi ikinci bölümde uygulanmaktadir. Buradaki asamalar birbiri ardina baglansa da birbirinden bagimsiz olarak egitilirler. Her kademenin hedef degeri, gerekli olan empedans çikisidir. Ilk asamanin kullanilmasindan sonra, veriler daha iyi hata minimizasyonu için ikinci asamaya uygulanir. In the second part of the topology, performing the mapping from inputs to outputs There is a correlation stage for MLPNN's back propagation learning algorithm is the second applied in the section. Although the stages here are connected one after another, they are trained independently. The target value of each step, the required impedance is the output. After using the first stage, the data is stored for better error minimization. applied to the second stage.
Egri uydurma, bir veri kümesi için en popüler modelleme yaklasimlarindan biridir. Bu yaklasim, veri setine maksimum uyuma sahip bir egri veya matematiksel fonksiyon olusturur. Fonksiyon, veri kümesiyle çok yakin ardisik degisiklik miktarini temsil eden matematiksel bir modeli ifade eder. Fonksiyon seçimi, nicel veri setinin grafik bilgilerinin yorumlanmasina dayanir. Curve fitting is one of the most popular modeling approaches for a dataset. is one. This approach is a curve or mathematical model with maximum fit to the data set. creates a function. The function calculates the amount of sequential change very closely with the dataset. Represents a mathematical model that represents Function selection, quantitative data set based on the interpretation of graphic information.
LS algoritmasi, egri uydurmanin en etkili yöntemlerinden biridir. Hata fonksiyonu minimizasyonunun kareleri toplamina dayanir. LS algoritmasi, türevlenebilir herhangi bir fonksiyon için kullanilabilir. En yaygin kullanilan fonksiyonlar polinom, üstel, Gauss ve Fourier'dir. Bu çalismada, LS algoritmasinda iki terimli üstel fonksiyon modeli, denklem (1 )'de verilen verileri uydurmak için kullanilmistir. The LS algorithm is one of the most effective methods of curve fitting. error function based on the sum of squares of the minimization. LS algorithm, any differentiable can be used for a function. The most commonly used functions are polynomial, exponential, Gaussian and Fourier. In this study, the binomial exponential function model in the LS algorithm, It is used to fit the data given in equation (1).
LS algoritmasi, egri uydurmanin en etkili yöntemlerinden biridir. Hata fonksiyonu minimizasyonunun kareleri toplamina dayanir. LS algoritmasi, türevlenebilir herhangi bir fonksyon için kullanilabilir. En yaygin kullanilan fonksiyonlar polinom, üstel, Gauss ve Fourier'dir. Bu çalismada, LS algoritmasinda iki terimli üstel fonksiyon modeli, denklem (1 )'de verilen verileri uydurmak için kullanilmistir. x!. = ai..eb"'""g" + credimg" (0.1) burada al., bi., ci., di. model katsayilari iken magl. denklem (2)'de gösterildigi gibi her bir Re ve lm deger çiftinin genligidir. magj = `lReJZ+11rnj2 (0.2) Burada hata fonksiyonu denklem (3)'teki gibi tanimlanmaktadir. The LS algorithm is one of the most effective methods of curve fitting. error function based on the sum of squares of the minimization. LS algorithm, any differentiable can be used for a function. The most commonly used functions are polynomial, exponential, Gaussian and Fourier. In this study, the binomial exponential function model in the LS algorithm, It is used to fit the data given in equation (1). x!. = ai..eb"'""g" + credimg" (0.1) here al., bi., ci., di. model coefficients while magl. As shown in equation (2), each is the amplitude of a Re and lm value pair. magj = `lReJZ+11rnj2 (0.2) Here, the error function is defined as in equation (3).
E((i,b, C, 01') : IÜ!' - (0.', BMW" + Quad-”m“ )) (0.3) En küçük hata, hata fonksiyonunun her bir degiskene göre kismi türevinin çözülmesiyle elde edilebilir. Elde edilen a,b,c,d katsayilari, Tablo 1'de gösterildigi gibi yapay sinir aginin ön isleme adimi için kullanilir.E((i,b, C, 01') : IÜ!' - (0.', BMW" + Quad-”m“ )) (0.3) The smallest error is calculated by solving the partial derivative of the error function with respect to each variable. obtainable. Obtained a,b,c,d coefficients, as shown in Table 1, It is used for the preprocessing step of the network.
Claims (3)
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TR2020/22620A TR202022620A2 (en) | 2020-12-31 | 2020-12-31 | ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE |
| PCT/TR2020/051504 WO2022146270A1 (en) | 2020-12-31 | 2020-12-31 | An ann based single calibration impedance measurement system for skin impedance range |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TR2020/22620A TR202022620A2 (en) | 2020-12-31 | 2020-12-31 | ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TR202022620A2 true TR202022620A2 (en) | 2021-01-21 |
Family
ID=75573025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TR2020/22620A TR202022620A2 (en) | 2020-12-31 | 2020-12-31 | ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE |
Country Status (2)
| Country | Link |
|---|---|
| TR (1) | TR202022620A2 (en) |
| WO (1) | WO2022146270A1 (en) |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AUPR343401A0 (en) * | 2001-02-28 | 2001-03-29 | Nguyen, Hung | Modelling and design for early warning systems using physiological responses |
| US10888273B2 (en) * | 2015-05-20 | 2021-01-12 | North Carolina State University | Personal hydration monitor |
-
2020
- 2020-12-31 TR TR2020/22620A patent/TR202022620A2/en unknown
- 2020-12-31 WO PCT/TR2020/051504 patent/WO2022146270A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022146270A1 (en) | 2022-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10070800B2 (en) | Impedance measurement process | |
| US8099250B2 (en) | Impedance parameter values | |
| EP2348987B1 (en) | Impedance measurement process | |
| Chowdhury et al. | Design and development of microcontroller based instrumentation for studying complex bioelectrical impedance of fruits using electrical impedance spectroscopy | |
| Al-Ali et al. | Design and implementation of portable impedance analyzers | |
| US20180014748A1 (en) | Post processing system and post processing method for electrical impedance tomography images | |
| CN107003296A (en) | Hemolysis test method and system | |
| JP6813563B2 (en) | Determining fluid level | |
| Sanchez et al. | Propagation of measurement errors through body composition equations for body impedance analysis | |
| Kapoulea et al. | Simple implementations of fractional-order driving-point impedances: Application to biological tissue models | |
| Pfeiffer et al. | Elliptical Fitting as an Alternative Approach to Complex Nonlinear Least Squares Regression for Modeling Electrochemical Impedance Spectroscopy. | |
| Khoma et al. | Synthesis of dynamic errors correcting algorithm for auto–balancing bridge methods | |
| Serrano-Finetti et al. | Toward noninvasive monitoring of plant leaf water content by electrical impedance spectroscopy | |
| TR202022620A2 (en) | ANN-BASED SINGLE CALIBRATION IMPEDANCE MEASUREMENT SYSTEM FOR SKIN IMPEDANCE RANGE | |
| Thulasi et al. | Portable impedance measurement device for sweat based glucose detection | |
| CN106618570A (en) | Skin biochemical index detection method and system based on biological dielectric spectrum | |
| Istanbullu et al. | An ANN-based single calibration impedance measurement system for skin impedance range | |
| Todorov et al. | Design of a Flexible, Wearable Interdigitated Capacitive Sensor for Monitoring Biomarkers of Atopic Dermatitis | |
| Juang et al. | Portable RDFT-based EIS system design with a low-complexity impedance calculation | |
| Al-Ali et al. | Bio-impedance measurement and applications | |
| Al-Ali | Design and implementation of a magnitude only bio-impedance analyzer | |
| Copot et al. | Fractional order modeling of diffusion processes: A new approach for glucose concentration estimation | |
| CN113080923B (en) | Equivalent bioelectrical impedance measuring method based on bridge method | |
| AU2006275299B2 (en) | Impedance parameter values | |
| CN109363662A (en) | A kind of Portable thrombus detection device |