The purpose of this clinical study is to compare the accuracy of linear endoscopic ultrasonography and magnified narrowband endoscopy in the diagnosis of invasive depth of early rectal cancer, and to provide more powerful evidence for the choice of surgical methods for early rectal cancer. Patients with early rectal cancer who will be examined by endoscopic ultrasonography or magnifying narrowband endoscopy in the department of gastroenterology and general surgery will be examined by linear endoscopic ultrasonography or magnifying narrowband endoscopy to examine the depth of invasion of early rectal cancer, and the results are compared with the postoperative pathological results of the patients as the gold standard. It provides strong evidence that the accuracy of linear endoscopic ultrasonography in judging the invasive depth of early colorectal cancer is not inferior to that of magnifying narrowband endoscopy.

1. Research purpose. 1.1 main purpose: to provide a more accurate preoperative evaluation method for the selection of surgical methods for early rectal cancer by comparing the accuracy of linear endoscopic ultrasonography and magnifying narrowband endoscopy in the diagnosis of invasive depth of early rectal cancer.

   1.2 Secondary purpose: to compare the sensitivity, specificity and examination time of linear endoscopic ultrasonography and magnified narrowband endoscopy in the diagnosis of invasive depth of early rectal cancer.

2. Experimental design. This study is a single-center randomized, self-controlled clinical trial. The subjects were selected according to the inclusion and exclusion criteria, and all subjects who met the inclusion criteria were required to sign an informed consent form before joining the group. The team of doctors is composed of four experienced endoscopes from the Department of Gastroenterology of Nanfang Hospital. Patients were randomly divided into group A (patients underwent linear endoscopic ultrasonography followed by magnifying narrowband endoscopy) and group B (patients underwent magnifying narrowband endoscopy first and then linear endoscopic ultrasonography). The examination results, examination time and complications were recorded in both groups. After examination, continue to follow up the patient's operation until the patient's surgical and pathological results are obtained.

   The main observation indicators of this clinical study include the accuracy of linear endoscopic ultrasonography and magnifying narrowband endoscopy in the diagnosis of the invasive depth of early rectal cancer, in which the surgical and pathological results are the gold standard, and the examination time of the two methods are also compared.

3. data analysis. After examination, continue to follow up the patient's operation until the patient's surgical and pathological results are obtained. Patients who did not undergo surgery and patients whose surgical and pathological results were not consistent with rectal cancer were excluded. The remaining patients took the pathological results as the gold standard to calculate the accuracy, sensitivity and specificity of linear endoscopic ultrasonography and endoscopic narrowband imaging in the diagnosis of invasive depth of early rectal cancer, and the results were statistically analyzed by chi-square test.

4. Safety monitoring, reporting and medical treatment. 4.1 adverse event (AE) definition. An adverse event refers to any adverse medical event that occurs after the patient or subject is examined in this study, which does not necessarily have a causal relationship with the examination. Therefore, adverse events can be any adverse physical signs (including abnormal laboratory results), symptoms, or diseases associated with the use of this study test, regardless of whether or not there is a causal relationship with the study test. Adverse events include serious adverse events ((Serious Adverse Event, SAE)) and non-serious adverse events.

4.2 SAE definition.

SAE refers to medical events during clinical trials, such as requiring hospitalization or prolonging hospitalization, disability, affecting working ability, endangering life or death, leading to congenital deformities and so on. Includes the following medical events:

1. events leading to death;
2. a life-threatening event (defined as the risk of the subject's immediate death at the time of the event);
3. events requiring hospitalization or prolonging the duration of hospitalization;
4. events that can lead to permanent or severe disability / insufficiency / affect work ability;
5. congenital abnormality or birth defect; Other important medical events (defined as events that endanger the subject or require intervention to prevent the occurrence of any of the above).

4.3. Adverse event recording, collection, reporting and handling. 4.3.1 Collection, reporting and processing of AE. After signing the informed consent form and prior to the inspection of this study, all AE related to the operating procedures specified in the test scheme should be recorded in the CRF.

The records of the AE shall include the description of AE and all related symptoms, time of onset, severity, duration, relevance to the trial drug, action taken, and final outcome and outcome. AE records must use medical terms, and if the subjects' symptoms and signs can be summed up by a common cause, the diagnosis should be recorded as far as possible. Except for the indicators related to disease progression, all clinical events and laboratory adverse reactions with clinical significance can be dealt with with reference to the evaluation criteria for common adverse events ((CTCAE) version 5.0). Adverse reactions caused by the treatment will be recorded by the researchers.

4.3.2 Collection and reporting of SAE. From the time the subjects signed the informed consent form to the completion of the study, all SAE, regardless of the cause or relevance to the inspection of this study, need to be reported by using the SAE report form. In the event of SAE, researchers should immediately take appropriate treatment measures to ensure the safety of the subjects, and report to the applicant for drug registration, the State Drug Administration, the Provincial Food and Drug Administration, the ethics committee of the corresponding clinical trial center and the Medical Department of the Medical Administration and Hospital Administration within 24 hours, and promptly report to the ethics committee of the team leader. As far as possible, the first report should include the following: the source of the report, the items examined in this study, the name of serious adverse events, the time of occurrence, severity, duration, relevance to experimental drugs, measures taken and outcomes.

4.3.3 AE severity criteria.

The researchers will evaluate the severity of the disease based on the five-level criteria developed by the NCI CTCAE5.0 version:

Grade 1, mild; asymptomatic or mild signs; only clinical or diagnostic observation without medical intervention; Grade 2, moderate; age-related limited functions of daily life (such as cooking, shopping, phone calls, etc.); Grade 3, serious or medically important but not immediately life-threatening; leading to or prolonging the length of hospital stay; disability; activities of daily self-care are limited (activities of daily self-care: bathing, dressing, undressing, eating, going to the toilet, taking medicine, etc., but not bedridden); Level 4, life-threatening and in need of emergency treatment; Level 5, AE-related death. 4.3.4 other responsibilities of researchers during follow-up of serious adverse events.

Researchers should examine and treat serious adverse events according to clinical judgment, including necessary clinical laboratory examination, physical examination and so on. Any examination results or other updated SAE-related information must be followed up. The time limit and procedure of the follow-up report are the same as those of the initial report.

5. Trial termination / suspension criteria. 5.1 the organizer has the right to terminate / suspend this study. Before terminating / suspending a clinical study, the sponsor must notify the researcher, the ethics committee and the relevant regulatory authorities and state the reasons. After the early termination / suspension of the study, the resumption of the study must be reviewed and approved by the Ethics Committee.

5.2 termination / suspension requested by the Ethics Committee.

6. The rule to end clinical trials.

The experiment ends when all the subjects meet the following conditions:

1. include at least the number of valid cases.
2. serious safety problems or serious adverse events occur.
3. there are major mistakes in the test scheme.
4. the organizer asks for termination.

7. Data management. 7.1 data Management. The main contents are as follows: 1) researchers must ensure that the data are true, complete and accurate; 2) when making any corrections, the test records can only be underlined, side notes on the revised data, explain the reasons, signed and dated by the researchers, and shall not be scrubbed or overwritten the original records; 3) the items of laboratory inspection are complete. 7.2. Data recording and file saving. The data about the subjects on the case report form should be recorded in the way of the subject code, and the subjects can only be identified by the subject code or their initials.

This study uses EXCEL for data management: from the data entry to the verification requirements of the source data, to the query and answer of the quality control data, and finally to the operation of data locking and export. After confirming that there is no doubt about the data, the parties sign the database locking application form, and the database is locked by the data administrator. After the database is locked, the analysis database is exported by the data administrator and handed over to the statisticians for statistical analysis. The locked data can no longer be edited, and the problems found after the database is locked can be corrected in the statistical analysis program after confirmation.

8. Statistic analysis. 8.1 sample size determination. After consulting the relevant literature, investigators found that the accuracy, sensitivity and specificity of magnifying narrowband endoscopy in the diagnosis of the depth of early rectal cancer invasion were about 90%, 75% and 90%, respectively. According to the previous research results of our team, the accuracy, sensitivity and specificity of linear endoscopic ultrasound in the diagnosis of early rectal cancer invasion were about 93%, 75% and 96%, respectively. Therefore, investigators speculate that the accuracy of linear endoscopic ultrasonography in the diagnosis of invasive depth of early rectal cancer is not inferior to that of magnifying narrowband endoscopy, and the non-inferiority threshold is set at 10%. By setting the bilateral test level is 0.025, the test efficiency (power) is 80%, according to the 1:1 ratio, calculated by PASS16.0 software, each group needs Nylon 73, plus the shedding rate is 10%, each group has at least 81 cases, because this study is a self-control study, so a total of 81 patients with early rectal cancer are needed.

8.2 Statistical methods. The measurement data are described by mean and standard deviation, and the counting data are described by examples and percentage. Chi-square test was used to evaluate the diagnostic accuracy, sensitivity and specificity of the two methods. A P value less than or equal to 0.05 will be considered to have a statistically significant difference between the test group and the control group. SPSS20.0 statistical software was used to analyze the indexes of the two groups respectively.

8.3 Statistical software and general requirements: The observational indexes of the two groups were statistically analyzed by SPSS20.0 statistical software.

Inclusion Criteria:

1. Colonoscopic diagnosis of early rectal cancer, the longest diameter >10mm.
2. Age 18 Mel 80 years old;
3. The subjects are willing to participate in this clinical trial and sign the relevant informed consent form.

Exclusion Criteria:

1. The coagulation function of the subjects was abnormal and uncontrolled: international standardized ratio ((INR))>2 or platelet count <75000 platelets / μ L;
2. Three months before entering the group, there were uncontrolled angina pectoris and / or myocardial infarction, congestive heart failure, lung (respiratory failure), kidney, pancreas, liver disease or other diseases were judged by the researchers to be unable to participate in the study.
3. Pregnant woman

In order to protect privacy, investigators intend not to disclose the information of the participants.

All the subjects who met the inclusion criteria were required to sign an informed consent form before joining the group, but they did not know their own grouping. The team of doctors is composed of four experienced endoscopes from the Department of Gastroenterology of Nanfang Hospital. Patients were randomly divided into group A (patients underwent linear endoscopic ultrasonography followed by magnifying narrowband endoscopy) and group B (patients underwent magnifying narrowband endoscopy first and then linear endoscopic ultrasonography). Different examination items are examined by randomly selected doctors, and the two doctors do not know the diagnosis of the other test. The examination results, examination time and complications were recorded in both groups. After examination, continue to follow up the patient's operation until the patient's surgical and pathological results are obtained.