Fundamentals and Practices of Sensing Technologies
by Dr. Keiji Taniguchi, Hon. Professor of
Dr. Masahiro Ueda, Honorary Professor, Faculty of Education and Regional Studies
Dr. Ningfeng Zeng, an Engineer of Sysmex Corporation
Global Medical Instrument Corporation),
Dr. Kazuhiko Ishikawa, Assistant Professor
of Education and Regional Studies,
[Editorfs Note: This paper is presented as Part XV and is the final installment of a series of sections from the new book gFundamentals and Practices of Sensing Technologiesh]
Chapter 7 Biomedical Sensors
In this chapter, principles of biomedical sensors are described: in section7.1,
measurements of human pulse waves; in section 7.2, measurements of hemoglobin; in
section 7.3, measurements of blood glucose levels; in section 7.4, detections of blast
cells in blood; in section 7.5, measurements of body fat; in section 7.6, urinary
component detections; and in section7.7 sensors for health management support
7.1 Measurement of Human Pulse Waves
7.1.1 Human Pulse Waves
Pulse diagnosis, as shown in Fig. 7.1, is one of the Chinese traditional diagnoses, which also includes a visual examination and a health interview. The pulse diagnosis is the most important and traditional diagnostic method. Yin and Yang, Truth and Falsehood, for organs in the human body (which are parameters of the health states defined by Chinese medicine) may be known by feeling the pulse based on Chinese traditional medicine, as shown in Fig. 7.1. Some diseases may, further, be predicted nowadays with the help of a diagnosis based on the speed, strength and rhythm of the pulse.
The most important aspect of this method is that pulse diagnosis provides a non-invasion measurement and imposes no pain on the human body. This section introduces a sensor for the pulse wave, a tool which is becoming more and more popular and is being applied in preventive medicine.
Fig.7.1 Pulse Diagnosis
7.1.2 Principle of Measurement of Pulse Waves
A sensor being used to measure the blood circulation of a fingertip is shown in Fig. 7.2. The blood circulation here represents that of the whole body because in the fingertip the blood that is pumped from the heart turns around. The fingertip, therefore, is a very effective place to measure blood circulation.
Fig. 7.2 Measuring blood circulation by a sensor
The measuring method is shown in Fig. 7.2. A near-infrared light from a LED irradiates a fingertip. A part of the light penetrates into the fingertip and is scattered in the capillary blood vessels. The scattered light is then received by a light-sensitive sensor iPhoto diodej.
The near-infrared light with a wavelength between 0.7 and 2.5 Κm has properties similar to visible light.
Power supply unit
Power supply unit
Light Emitting Diode
Light Emitting Diode
Fig. 7.3 Block diagram of the measurement of blood circulation
A block diagram of the measurement of blood circulation is shown in Fig.7.3. The electrical signal obtained from a light-sensitive sensor for detecting pulse waves is amplified by the amplification section. The amplified signal is, then, converted into a digital signal by an analog-to-digital (A/D) converter. This digital signal is, finally, carried to a personal computer.
The personal computer records every one beat of the capillary blood vessel and then calculates the time progress of the blood circulation. An acceleration pulse can be processed instantaneously by using special software. Diseases can, finally, be predicted by using a graph of the recorded data, as shown in Figs. 7.4 and 7.5(1).
Fig. 7.4 Pulse waves
7.1.3 Relationship between Pulse Waves and Age of Blood Vessels
The relationship between the pulse waves and aging is evaluated as follows.
A waveform of the acceleration pulse waves is shown in Fig. 7.6. The parameters a, b and c can be measured. The tendencies of the changing values are known from many samples of pulse waves which are obtained from different age people. As a result, the values of a and b may be getting smaller, the values of c may be getting larger, when people are getting older. Based on the parameters of the waveform, an index which is created to evaluate the health state of blood vessel, the index (IBV) is calculated by the following formula.
IBV = (-b+c)/a
The relationship between the IBV and ages is shown in Fig.7.7. This result was obtained from a group of men. Because there is a strong correlation between the IBV and ages, the correlation can be used to examine the state of health.
Ages 70s 60s 50s 40s 30s 20s 10s
Fig. 7.7 Relationship between IBV and ages
7.2 Measurement of the Hemoglobin
7.2.1 Hemoglobin (HGB)
Hemoglobin (abbreviated HGB) is the iron-containing oxygen-transport metalloprotein in the red blood cells of human, the structure of hemoglobin is shown in Fig.7.8. The protein makes up about 97% of the red cellfs dry content. Hemoglobin transports oxygen from the lungs to the rest of the body, such as to the human brain, muscle and organs, where it releases the oxygen for cell use.
Fig. 7.8 Structure of Hemoglobin
HGB is composed of a pigment called hem and protein called globin, so it is called hemoglobin.
7.2.2 Relation between HGB and Human Health
The normal value of hemoglobin in
Pregnant women, old people, children tend to be low. The normal values are different in different countries and areas.
When the hemoglobin of the required amount is not enough, the transportation of the oxygen is not enough. It may cause fast heartbeat to fasten the@circulation of the blood and makes the person out of breath. Furthermore, the people probably suffer from various anemia such as sideropenia anemia, aplastic anemia, hemolytic anemia, and chronic bleeding-related anemia or leukemia.
A HGB measurement is, therefore, very important to human health so that it is a regular item of health examination. It also is very important to measure the health condition of the athlete.
Here, two methods for HGB measurements are introduced.
7.2.3 Principle of HGB Measurement
@@SLS@( Sodium Lauryl Sulfate) hemoglobin method is a method for HGB measurement.
Whole Blood Hemolytic reagent
Photo diode Photo
Sensor HGB cell
Fig. 7.9 Simple structure for measuring SLS-Hb
The red blood cell is hemolyzed by interface activity action of SLS( Sodium Lauryl Sulfate C12H25SO4Na)and hemoglobin is released. A change of the three-dimensional structure happens, and Fe3 + of the hem is generated as the of SLS is combined with globin in the oxidation hemoglobin. Fe3 + is combined with a hydrophilic group of SLS since Fe3 + has not binding capacity with oxygen. In the same time, oxygen is released, and SLS-Hb is generated immediately.
As shown in Fig. 7.9, the light of wavelength 550}15nm (or 540}5nm) from a photo diode passes through a glass filter and then the HGB cell, and finally, the rest of light is received by the photo sensor in the other side. The transmitted light is changed into electric current. If the stronger current is obtained, it means the lower density of the SLS-Hb.
The measurement flow is described as follows:
In the first step, the HGB cell is washed and cleaned.
In the second step, the bland value (B_value) is measured, where there are nothing in HGB cell after being cleaned.
In the third step, the whole blood is mixed with SLS and dilute solution.
In the fourth step, the mixture is agitated for a while.
In the fifth step, the mixture value (SLS-Hb_value) is measured in HGB cell.
In the final step, HGB is calculated by using the following expression.
HGB_value = SLS-Hb_value - B_value
Those steps can be done automatically by using sequence programs.
B. Non-invasive Method
(A) Near-infrared Spectroscopic Imaging Method
The Near-infrared Spectroscopic Imaging Method is based on the light absorbance feature.
Fig. 7.10 Structure of measurement
Fig. 7.10 expresses the measurement setup. Near-infrared light is used as a light source. The light passes thought a human finger and an optical lens. An image can be captured by a CCD camera on the other side.
(B) Selection of Light Source
Fig. 7.11 Feature of absorbance
Because the HGB value includes both of Hb and HbO2, both of them are needed to be measured. The feature of absorbance of Hb and HbO2 are shown in Fig.7.11 (2). The milimolar absorptives of oxygenated and deoxygenated hemoglobin are almost the same, when the wave length 805nm is applied. By using this light source, it is easier to discriminate the HGB from others in CCD image. On the other hand, the density of HGB in CCD image is not changed, even the proportion of Hb and HbO2 is changed. Furthermore, this wavelength is good enough to penetrate human fingers from a lot of experiments. Therefore, wave length 805nm is selected for HGB measurement.
(C) Data Processing Algorithms
Here, an algorithm for the estimation of Hgb is described.
Fig. 7.12 The procedure to obtain the information of vessel for Hgb estimation
Fig. 7.12 shows the procedure to obtain the information on the vessel of a human
finger for Hgb estimation.
A human finger is radiated by the near infrared light. The difference between the finger tissue and vessel can be obtained by the image on a CCD camera placed on the other side as shown in Fig. (a), since the light transmission rate for vessel is different from the rate for finger tissue. The Hgb can, then, be estimated from this image by means of the intensity and absorbance, as follows:
HGB ΰ KEh/wn
(D) Result for Hgb
Fig. 7.13 The correlation between the predicted values obtained
from the non-invasion and the SLS_Hb
The result is shown in Fig. 7.13(2). A good correlation was obtained between the predicted hemoglobin concentrations using this method and the reference values obtained by the SLS hemoglobin method using an automated blood cell counter (r=0.86, n=54). These results show that this noninvasive method can be applied to estimate the value of Hgb. It can, further, be used as a noninvasive method for anemia screening since this method has the advantage of not requiring blood sampling and no pain to human.
7.3 Measurement of Blood Glucose Level
7.3.1 Blood Glucose Level
Blood glucose level is also called blood sugar level.
Glucose levels rise after meals for an hour or two, and are usually the lowest in the morning, before the first meal of the day. The level of glucose usually changes as shown in Fig. 7.14
Fig. 7.14 Change of the glucose level of a day
The measurement of glucose level can be used for a diabetic screening. And it is very important parameter for the determination of metabolic syndrome.
Diabetes is a chronic disease that affects the body's ability to produce or respond to insulin, the hormone that allows glucose to enter the body's cells and be stored or used for energy. Many diabetics require insulin injections, and all must carefully monitor and manage their blood glucose levels. Therefore the measurement of glucose level is very important. Here two methods are introduced.
Fig. 7. 15 Diagnosis criteria of WHO for diabetes
7.3.2 Principle of Measurement of Blood Glucose Level
A. Invasive Method
(A) Principle of Invasive Method
There are many methods that can be used for glucose measurement, Here, an enzymatic electrode method is described. The advantages of this method are listed as follows:
1. Even there is no oxygen, the measurement is possible.
2. The measurement does not depend on the special reagent, if there are
the enzyme and the electron mediator, the measurement can be done .
3. This method is easy to operate and maintenance.
Therefore it can be used as self-monitoring of blood glucose (SMBG).
The structure of glucose sensor is shown in Fig. 7.16, there are two different electrodes, above the electrodes, there is reagent sheet, where is used for putting a measurement sample of blood. The procedure of measurement is shown in Fig. 7.17.
Fig. 7.16 Structure of a glucose sensor
Fig.7.17 Procedure for the measurement
The reaction(3) of measurement principle is shown in Fig. 7.17.
ferricyanideferricyanide ferrocyanide ions through GOD by a specific reaction. The reaction also is described in the following chemical equation(1) .
Second, when a constant voltage is applied, it reduces ferrocyanide ion into ferricyanide ion (The reaction also is described by the following Formula(2)) and an electric current is generated. This current is to the level of the glucose. So the current can be used for measurement of glucose.
Glucose + 2[Fe(CN) 5]3- à Gluconic acid + 2[Fe(CN) 5]4- (1)
2[Fe(CN) 5]4- à 2[Fe(CN) 5]3- + 2e- (2)
(B) Response Feature
The relationship between the current of sensor response and the level of glucose is shown in Fig.7.18.
Fig. 7.18 Response feature of glucose measured by using enzymatic electrode method
B. Minimal-invasive Method
@@Because enzymatic electrode method is need some blood sample which are described in last section, it causes pain or mental pain by using the puncture and there are risk of infection. Here, a minimal-invasive method is described in this section.
(A) Principle of Minimal-invasive Method
This method(4) is not measured the glucose from blood but is measured from tissue fluid under the horny layer. Because there is a good correlation between the glucose level from tissue fluid and that value from blood. As shown in Fig.7.19, there is slight fine needle array which is put above the horny layer, so a group of microporosity passes is formed, after a short time later, the level of glucose can be measured by using a high sensitive sensor. The current of sensor is converted into a digital signal using A/D converter. Then the data are processed by using a micro-processor. Finally, the result is shown in a display.
Fig. 7.19 Structure of the mimimal-invasive measurement for the glucose level
(B) Experiment Result
7.4@Detection of Blast Cells in Bloodi5j
7.4.1 Blast Cells in Blood
Usually, Blast cells do not appear in peripheral blood of healthy people before they grow up. If the blast cells appear in peripheral blood, the person is suspected as having leukemia.
The mature processes of blood cells of two examples are shown in Fig.7.20. One is a kind of white blood cell which is called an acidocyte, the other is a red blood cell. Therefore, the measurement of blast cell in peripheral blood plays a very important role in diagnosing disease.
Red blood cell
Red blood cell
Fig. 7.20 Mature processes of blood cells
iThose cells are cited from http://plaza.umin.ac.jp/ Μ WEB PHYSIOLOGYj
7.4.2 Principle of detection of Blast Cells
(A) Semiconductor Laser Diode
Nowadays, because the laser diode has specific features of which it is hard to diffuse, and its light is able to arrive to a long distance, furthermore, it is the low cost, and the low energy requirement, the laser diode is popularly applied as a light source of many sensing machines.
Here, the laser diode of the wave length of 633(nm) is applied for the flow cytometry.
(B) Flow Cytometry for Detection of Blast Cells
(a) Three dimensional illustration of a flow cytometry
(This figure is cited from Sysmex Journal Vol.29 2006)
(b) Side view of the flow cytometry
(c) Top view of the flow cytometry
Fig. 7.21 Structure of the flow cytometry of automatic hematology analyzer
The structure of the flow cytometry for detecting blast cells is shown in Fig.7.21.
A 3D structure of the flow cytometry is shown in Fig.7.21 (a), and the enlarged flow structures of side view and top view are shown in Fig7.21(b) and Fig7.21(c), respectively.
A front scattering light, side fluorescence light, and side scattering light can be obtained. The front scattering light is usually used for getting the information of size of cells, the side scattering light is used for getting the shape, density of cells and determination of existence of granulation on the surface of cells, and the side fluorescence light is used for getting the information of the quanta of DNA and RNA.
Here, a scattering diagram is used to determinate the blast cells. The information for the scattering diagram is obtained from the front scattering light and the side fluorescence light as shown in Fig.7.21. Based on many experiments and using many samples, the blast area can be detected as shown in Fig7.22. However, only this scattering diagram is not enough, because some other substances such as atypical lymphocytes may also appear in the same area of this diagram. In order to get better information of blast cells, RF/DC detection method is also applied for detecting the blast cells at the highest precision.
It is introduced in the next session.
Front scattering light
Front scattering light
Fig.7.22 Scattering diagram for blast cell detection
7.4.3 RF / DC Detection Method
Fig.7.23 Structure of RF/DC sensor
As the principle of RF/DC method is described in chapter 1, see in Fig. 1.4.
The size of the nucleus in the cell or density of the cytoplasm reflects a state of the cell inside and change. Based on the two parameters, a diagram is created as shown in Fig. 7.24.
Fig.7.24 Scattering diagram of blast cells obtained from RF / DC method
By using many samples, the area of blast cells is detected as shown in Fig.7.24.
7.5 Measurement of Human Body Fat
7.5.1 Relation between Body Fat and Human Health
There is a new word which is called gmetabolic dominoh for describing the relation between body fat and human health. It means the body fat or obesity which can cause a chain reaction associated with many kind diseases as shown in Fig.7.25. The obesity is a condition in which too much body fat has accumulated. When a person is in this condition, it will lead to high blood pressure and insulin resistance. Then, it will lead to hyperlipidemia and diabetes. Finally, it may lead to cardiovascular disorder, blindness and dementia.@Therefore, the monitor and control of body fat play an important role in preventing many diseases.
Fig.7.25 Metabolic domino
There are different kinds of methods to measure the body fat, such as BMI method which is defined by many
Because there are some limitations of those methods, they are not convenience to be used. Here, a BIA (Bioelectrical Impedance Analysis) method is introduced. Because it is the most simple, rapid and non-invasive, so this method gets popularity.
7.5.2 Bioelectrical Impedance Analysis Method
A. Relationship between Impedance and Body Fat
When a low level electric current is flown the human body, there is no danger to human being, and the impedance can measure. If there are more the water contained in a body, the current is easier to flow, because the internal water is a good conductor, an electric current is easy to flow. If there is much quantity of water contained in a body, and electrical resistance becomes smaller.
If there is a lot of body fat in the body, conductivity become smaller, so the value of impedance gets higher. In the case reverse, the value of impedance gets lower.
Because of the difference of the impedance, we can estimate the body fat percentage due to a lot of experiment data of different people.
B. Principle of Measurement Method(6)
Usually, the impedance can be calculated by the following formula.
Z =k~H/S (7.6)
where k , H and S represent a resistivity, a height of a block of tissue and an area of a block of tissue, respectively.
@@@@@@@@@@@@Fig.7.26 Calculation of resistance of human body
As a result described above, a circuit model which is shown in Fig.7.27, is used for a human being.
Fig.7.27 A circuit model of human being
V1 = I1(Z1+Z5+Z3) ( The voltage V1 is imposed on Electrode1 and Electrode 3) (7.7)
V2 = I2(Z2+Z5+Z4) ( The voltage V2 is imposed on Electrode2 and Electrode 4) (7.8)
V3 = I3(Z4+Z3) ( The voltage V3 is imposed on Electrode3 and Electrode 4) (7.9)
V4 = I4(Z1+Z2) ( The voltage V4 is imposed on Electrode1 and Electrode 2) (7.10)
V5 = I5(Z1+Z5+Z4) ( The voltage V5 is imposed on Electrode1 and Electrode 4) (7.11)
Firstly, when V1, V2, V3, V4, and V5 are applied, the currents I1, I2, I3, I4, and I5 are measured. Then the impedances ( Z1, Z2, Z3, Z4, Z5) can be calculated. Therefore, based on the measured impedances, the fat percentage can be calculated. The detail of this method is described as follows.
W = FFM + FM (7.12)
where FFM , and FM represent the weight of body which is not contained any fat (Fat Free Mass), the weight of fat body (Fat Mass)l, respectively.
Therefore, FM can be calculated by using the following formula.
FM = W - FFM (7. 13 )
Here, a precise equation is applied for FFM estimation which is shown in the following formula.
FFM = −4.104 + 0.518Ht2 / R50 + 0.231weight + 0.130X + 4.229 Gender (7.14)
Where Ht represents height of body,
R50 represents resistance (a voltage of frequency 50kHz is applied),
Weight represents the body weight,
X represents reactance,
Gender represents an adjustment coefficient (1 for men, 0 for women)
Using formulas (7.13) and (7.14), FM or fat percentage can be calculated.
This method needs the data of a human height, the weight of a human body, and the gender for the calculation.
7.6 Urinary Component Detection
7.6.1 Urinary Components and Urinary Sediments
In urinary sediments, there are many solid substances such as red blood cells, white blood corpuscles, crystals, casts, and the microbes. If those substances are more than normal value, some diseases are suspected in kidney or the organ of urinary passage. In the traditional method, these ingredients are detected by using a microscope. Here, an automatic method to detect the substance is described. An example of urinary sediment image is shown in Fig. 7.28.
By urinary sediment analysis, the following diseases can be known.
1) If the number of red blood cells increases,
2) If the number of white blood cells
High blood pressure
7.6.2 Detection method for Urinary Components
There are two methods for detecting the substances of urinary sediments.
One is the flow cytometry, another is the image processing.
A. Flow Cytometry(7)
A flow cytometry shown in Fig. 7.29 is used for detecting substances of urinary sediments. And an example detecting result was shown in Fig.7.30.
Fig.7.29 Flow cytometry for detection of urinary sediments
Based on the detected scattergram, the different substances of urinary sediments can be classified.
FSC FL2 RBC BACT WBC Monocyte Polymorphonuclear Cell
Fig.7.30 Scattergram of the Flow Cytometry
B. Image processing method for urinary sediments
An image system which is used for detecting the substances in urinary sediments, is shown in Fig.7.31. Urinary sediment samples are injected from the top of flow chamber. The urinary sediment combines with a kind of reagent, then the mixtures flow in flow chamber from top to the bottom.
In the same time, urinary images are taken by a CCD camera, these images is stored in an image memory and processed by a computer. There are limitations to detect the abnormal substances, so the processing algorithms are important to improve the recognition rate. By using different recognition algorithms(8,9,10), the abnormal substances such as transparent cast in urinary sediment can be detected.
An image of urinary sediment shown in Fig. 7.32 is a transparent cast. Although it is transparent and very difficult to determine its edge, it can be recognized after processing by our algorithms. 50 samples of abnormal WBCs and 50 other no-WBCs are also used for this test, 46 samples of abnormal WBCs can be recognized. The recognition rate is more than 90% by using the neuro-fuzzy method.
Fig.7.31 Block diagram of the flow system
The processed result of Fig7.28 is shown in Fig.7.32.
Fig. 7.32 The processed result image of urinary sediment
7.7 Sensors for Health Management Support System
7.7.1 Importance of Health Management
According to surveys
The leading cause of mortality is cancer, cardiac disease, cerebral vascular as shown in Fig. 7.33@(there are similar results in other developed countries.j
Others 32% 41.6% Others 38.3% 41.6% Pneumonia 9.9% 41.6% Cerebral vascular 11.8% Cardiac Disease 15.9% Cancer 23.1% Cancer 30.4% Cardiac Disease 27.2%
Cerebral vascular 11.8%
Cardiac Disease 15.9%
Cardiac Disease 27.2%
Fig.7.33 The leading cause of Mortality of
Lifestyle Risk indicator Cardiovascular disease& others
Fig.7.34 Pathogeny of human being
As shown in Fig.7.34, cancer, cardiac disease and cerebral vascular are caused from Obesity, Hyperpiesia, Hyperlipidmia, High blood Glucose level. Furthermore, those symptoms are almost caused from many kinds of lifestyles. Table 7.1 shows criterions of metabolic syndrome. If a person meets the criterions which are described as follows, he is diagnosed as a metabolic syndrome.
(1) The waist meets the criterion of 1) item as described in Table 7.1.
(2) Two or more of 2), 3), 4) and 5) items as described in Table 7.1 meet their criterions.
Table 7.1 Criterions of metabolic syndrome
@The first parameter is the waist which is measured by using a ruler. The second parameter is the neutral fat which can be measured by using a laser diode. The third parameter is the HDL cholesterol which can be measured by using a cholesterol sensor. The fourth parameter is the blood pressure which can be measured by using a pressure sensor. The final parameter is the blood glucose level which can be measured by using a glucose sensor.
If a person who is in the stage of metabolic syndrome(13) , he or she can easily recover by changing her lifestyle into a right lifestyle. Therefore the health management is very important to our human health. Here, a health management system is described as follows.
7.7.2 Composition and Principle of Health Management System
Here, a health management system shown in Fig.7.35 is introduced. This system is composed of software and four devices which are described as follows:
1) A device for measuring the fat percentage, the muscle percentage and body weight of human body: This principle is described in section 7.5 in chapter 7.
2) A device for blood pressure monitor and cardiac rate: This principle is described in section 2.2 in chapter 2.
3) A device for pulse wave: This principle is described in section 7.1 in chapter 7.
4) A device for measuring HGB: This principle is described in section 7.2 in chapter 7.
5) The software for health data analysis and providing advices for prevention of diseases based on humanfs health check and his or her lifestyle.
Because it is used for preventing lifestyle disease, all of devices, which are selected, are non- invasion devices.
Table 7.2 Devices and sensors
This system is not used for metabolic syndrome diagnosing. It is used for health management in order to preventive diseases.
Blood pressure monitor
Fig. 7.35 Structure of healthcare management system
Fig. 7.36 Flow diagram for processing of health management.
A flow diagram of the health management system is shown in Fig.7.36.
(1)In the first step, health check data are measured using different kinds of devices, which are described above.
(2)In the second step, a group of questionnaire data of the food analysis and lifestyle, which are provided by tested people, is input into the computer.
(3)In the third step, the data of health examination also can be acquisitioned into computer, if the people have health examination data, which are provided by examination center.
Usually, those data are updated twice a year, when the people take health examination.
(4)In the fourth step, there is a preventive medical knowledge database, which can provide different kinds of advices for difference kinds of people and health conditions.
(5) In the fifth step, the program of data analysis is executed data processing based on the input data.
Finally, the health management system program provides some warning information and good advices for your health.
(1)Sano Y. et al. The evaluation of blood circulation and its application based on acceleration plethysmograph, Science of Labor, 61(3), pp.129-143 (1985)
(2)Toshiyuki Ozawa. ,Kaoru Asano, Shigehiro Numada, Yasushi Hasui, Yasuhiro Kouchi, Ken Ishihara , Noninvasive Measurement of Hemoglobin Concentration Using the Near-infrared Spectroscopic Imaging Method. Japanese medical and Biological Engineering 43(1), pp.93-102 (2005)
(3)Measurement system of glucose, Matsushita technical journal 51(3), pp. 248~253 (2005)
(4)Maekawa Y. ,Sato T. , Okada Seiki, Hagino k. Asakura Y. kikkawa Yasuo, Kojima J. , Omiya K., Takase T. , Hirakawa M., Uematsu Ikuo, Nawata I. Development of minimal-invasive self-monitoring system for blood glucose. The 48th Annual conference of Japan Society for Medical and Biological Engineering, pp.255 (2008)
(5)A.M. Cenci, M. Maconi, and B. Casolani:Evaluation of the Diagonostic Performance of the Sysmex XT-2000i Automated Hematology Analyzer in the Detection of Immature Granulocytes, Sysmex Journal International Vol.15,No1,pp.1-6(2005)
(6)Ursula G. Kyle, Ingvar Bosaeus, Antonio D. De Lorenzo,Paul Deurenberg, Marinos Elia, Jos!e Manuel G!omez,Berit Lilienthal Heitmann, Luisa Kent-Smith, Jean-Claude Melchior,Matthias Pirlich, Hermann Scharfetter, Annemie M.W.J. Schols,Claude Pichard, Bioelectrical impedance analysis part I: review of principles and methods, Clinical Nutrition 23,pp.1226–1243(2004)
(7)UF-100 Clinical Case
Study, TOA Medical Electronics, Scientific Division,
(8)Ningfeng Zeng, Keiji Taniguchi, Sadakazu Watanabe, Yutaka Nakano, Hiroyuki Nakamoto. Fuzzy Computation for Detecting Edges of Low Contrast Substances in Urinary Sediment Images, Medical Imaging Technology, Vol.18, No.3 (2000)
(9)Ningfeng Zeng, Keiji Taniguchi, Hong Zhu, Sadakazu Watanabe, Yutaka Nakano. Noise Analysis and Noise Suppression with the Wavelet Transform for Low Contrast Urinary Sediment Images, Medical Imaging Technology, Vol.18, No.6 (2000)
Zeng, Keiji Taniguchi, Sadakazu Watanabe, Yutaka Nakano, Hiroyuki Nakamoto. A
fuzzy model for the classification of abnormal substances in urinary sediment
images, Trans. IEE
(11)The homepage of Ministry of Health Labour and Welfare,
(12)Arialdi M. Minino, M.P.H; Melonie P. Heron, Ph.D ; Sherry L. Murphy,B.S.; and Kenneth D.Kochanek, M.A.; Division of Vital Statistics, Deaths: Final Data for 2004, National Vital Statistics Reports, August 21(2007)
Diabetes Federation. A New world wide definition of the metabolic syndrome,
[ BWW Society Home Page ]
© 2011 The Bibliotheque: World Wide Society