Design of liver functional reserve monitor based on three-wavelength from red light to IR

1.Introduction

The evaluation of liver functional reserve is an important technical assurance before performing liver resection operations, which is one of the core problems of hepatic surgery [1, 2]. By the proper evaluation of liver functional reserve, the surgeons can accurately master the cutting quantity of liver, and avoid the hepatic failure after operation [3]. The existing evaluation methods of liver reserve ability include Child-Pugh grading standard, excretion test of ICG (indocyanine green), CT volume computing method and so on. In this present paper, the ICG excretion test is adopted. The ICG is a kind of dark green dye [4, 5]. After injecting ICG into veins, the serum protein will bind with it, and then be taken selectively by the liver. In free state form, the protein binding with ICG will be secreted to the cholecyst, not taking part in the circulation of intestines and liver [6, 7, 8]. The ICG is excreted without passing through the kidney. By ICG clearance measurements, some related liver function analyzers, such as Nihon Kohden DDG-3300k, Pulsion Limon, have serviced the patients for years. The method they adopted is PDD (Pulse Dye Densitometry) whose principle is as follow: When there exist two different absorptiometic matters in blood, the concentration ratio of these two matters can be determined by the transmission of light pulse acquired through using two different wavelengths of near infrared light to irradiate the issues [9, 10]. Compared with the traditional spectrophotometric methods, the existing systems that adopted PDD do not need any venipunctures except for the intravenous injection of ICG, which is convenient and painless for patients [11]. Besides, the ICG clearance is the most sensitive and earliest indicator that can reflect the liver function, and the dynamic variation of liver function can be observed in real time [12]. However, due to the testing principle based on two-wavelength near-infrared light, the slight shake of tested part such as finger, nose, will lead to an inaccurate result. If the ICG testing result is very high (> 40%), the degree of separation may be slightly poorer [13].

Degree of blood oxygen saturation is an important indicator for judging whether the respiratory system and circulating system are in anaerobic status, and this parameter is very meaningful in the clinical medicine field [14, 15]. The blood oxygen monitor acquires the pulse signal by gathering the transmission light through the human tissue, and gives the pulse blood oxygen information in real time [8]. The pulse signal, whose period is not always equal, is a kind of quasi-periodic signal, which means the pulse waveform is not strictly equal [16]. So, it is meaningful for the daily monitor how to acquire and process the pulse signal accurately and effectively [17].

The principle of the detector designed in this paper is based on pulse dye density spectrophotometric method, which is improved on the pulse blood oxygen spectrophotometric method [18]. In our research, the microcomputer and photoelectric sensor technologies are combined. One end of the detector is formed with light source driving circuit and three light-emitting diodes (λ1= 660 nm, λ2= 805 nm, λ3= 940 nm), the other end is photoelectric sensor chip OPT101. The equipment designed is not only small in size, low in consumption, better in data storage and administration, but also multi-functional in use and powerful in network function. The clinical experiment results are identical with expectation, which is very valuable for the evaluation of liver reserve ability. For the spectrophotometric method, the influence of scattering of blood is very small. So the term scattering of blood is neglected in our model. The ratio of transmission light pulsating quantity of two wavelengths can be written as follow:

Where, a denotes artery, and v denotes vein. E=a SaO2Eo+ (1 - SaO2) Er, E=v SvO2Eo+ (1 - SvO2) Er and V =ΔDV/ΔDa.

It can be seen from the formulas above there are four parameters SaO2, SvO2, V and Cd. In fact, V and SvO2 are often taken as constants in practical application. So we need only three wavelengths to acquire the real-time ICG concentration and blood oxygen saturation.

A finger-style pulse blood oxygen detection system is designed in this paper. The photoelectric sensor POT101 and ARM9 S3C2440 are used to acquire and process the signal. The whole functional diagram of the system is shown in Fig. 1.

Figure 1.

Functional diagram of acquisition and processing system for pulse blood oxygen signal and ICG concentration.

2.Design of hardware system

The hardware system is mainly composed of two modules, signal acquisition unit and mainframe circuit unit (Fig. 2). The signal acquisition unit is made of the LED and photoelectric sensor that distribute at the two sides of finger respectively. The mainframe circuit unit is made of S3C2440, LED sequential drive and control circuit, amplifier, filtering circuit and voltage follower.

Figure 2.

Structure diagram of hardware system.

2.1Signal acquisition unit

The choice of light sources is the chip L660/802/940-35B42 made by Epitex Company. In this chip, there are three light sources whose wavelengths are 660 nm, 805 nm and 940 nm respectively, which reduces the dimension of signal acquisition unit greatly, and ensures the accordance of reception scope of three-wavelength light sources with photoelectric sensors. The choice of photoelectric sensor is OPT101 made by BB Company (Fig. 3). This chip gets the photoelectric sensor and pre-amplifier integrated, and avoids the problems that usually happen in the separate design such as leakage current error, noise cross interference, gain peaking caused by the stray capacitances and so on. It is speedy for the chip to convert the optical signal to the electrical signal.

Figure 3.

Structure principle of OPT101.

3.Design of software system

The software system designed with EVC runs under the environment Window CE5.0. The frame diagram of this system is shown in Fig. 4. It includes three modules: data acquisition module, data processing and display module, database and network module.

Figure 4.

Frame diagram of software system.

3.1Data acquisition module

In this system, the S3C2440 is adopted for the 10-bit A/D conversion, and the external pulse signal is brought in the ARM board by adopting the interrupt read module. When the external interruption is triggered with the high level, the corresponding ADC reading data channels are controlled by programs. The flow diagram of data acquisition is shown in Fig. 5.

Figure 5.

Flow diagram of data acquisition.

In the software design, the ADC drive program is adopted to sample. The work timing sequence of LED is controlled by interrupt service routine, and the data acquisition process is efficiently controlled in real time. The sampling frequency is set as 50 Hz (because of the time-share sample, the practical sampling frequency is 300 Hz), and the clock frequency is set as 300 Hz. The working time sequence of LED is shown in Fig. 6.

Figure 6.

Working time sequence of LED.

Figure 7.

ICG curves of different suspected primary liver cancer patients.

Under the control of program, the three LEDs create light successively. The wavelengths are 660 nm, 805 nm and 940 nm, respectively. In Fig. 6, ON denotes the diode is in power-up state, OFF denotes cut-off state. Where, T=1 T=2 T=3 T=4 T=5 T=6 1/300 Hz ≈ 3.33 ms. For each LED, the time ratio of ON state and OFF state is 1:5, the time interval of timer is set as 1/300 Hz. The program controls the working sequence in accordance with Fig. 6.

4.Experimental results and analysis

In experiment, the ICG was injected into a vein of the patient in advance, and the ICG concentration curve and the retention rate ICGR of each patient were detected with the equipment we designed. According to the difference of ICGR15, The patients were divided to three groups: Group S1, ICGR15 < 10%; Group S2, 10% ⩽ ICGR15 ⩽ 20%; Group S3, ICGR15 > 20%. Figure 7 shows the ICG curves of different suspected patients of primary liver cancer acquired with our detection system.

Figure 7b, d and f are the ICG curves of three patients who were respectively diagnosed as liver function being normal, mild liver insufficiency and severe liver insufficiency. It can be seen in the ICG excretion experiments, if the one whose liver reserve ability is stronger, the residual quantity of ICG in veins is less, which means the value of ICGR15 is smaller. Figure 7a, c and e show if the liver reserve ability is stronger, in the excreting process of previous 1 minute, the first crest is often higher than the second one. These features are very valuable for the evaluation of liver reserve ability.

5.Conclusions

A three-wavelength liver functional reserve testing system is proposed in this paper. The results of clinical experiments show the signal of this system is stable and reliable, and can effectively assess the reserve ability of liver. With the improving of liver functional reserve detecting technology, the application prospect of portable liver functional reserve detecting equipment will be more and more extensive. The interference from human body and surroundings is reduced greatly by adopting digital filter and moving averages smoothing method to filter out the noise, and using the threshold value method to extract the feature points of pulse wave. The ICG clearance rate are affected at the blood flow volume of liver to a large extent, so any factor influencing the blood flow volume of liver will bring impact on the testing result. Besides, serum opacity and jaundice will lead to the colorimetric error. The patients with obstruction of biliary tract are not fit for the ICG clearance testing.

The pulse blood oxygen data contain much valuable information about the human body. In this paper, a liver functional reserve testing system is proposed based on three-wavelength from red light to NIR. Combined with the database and network, the remote control and management are realized. The OPT101 photoelectric sensor and S3C2440 are used in this system, which makes the equipment much faster and smaller. Besides the acquiring of the pulse blood oxygen data, this scheme can also store and manage the data through the AQLCE database. A friendly interactive interface is designed to display the related data, and the network transmission and sharing is realized by Web programming. The results of clinical experiments show this system, whose the signal is stable and reliable, can effectively evaluate the reserve ability of liver. The system presented in this paper is not only convenient to use, but also powerful in functions.