Artificial intelligence in small intestinal diseases:Application and prospects

The small intestine is located in the middle of the gastrointestinal tract,so small intestinal diseases are more difficult to diagnose than other gastrointestinal diseases.However,with the extensive application of artificial intelligence in the field of small intestinal diseases,with its efficient learning capacities and computational power,artificial intelligence plays an important role in the auxiliary diagnosis and prognosis prediction based on the capsule endoscopy and other examination methods,which improves the accuracy of diagnosis and prediction and reduces the workload of doctors.In this review,a comprehensive retrieval was performed on articles published up to October 2020 from PubMed and other databases.Thereby the application status of artificial intelligence in small intestinal diseases was systematically introduced,and the challenges and prospects in this field were also analyzed.

Application of double-balloon enteroscopy in the diagnosis and therapy of small intestinal diseases

Background X-ray of the digestive tract and radionuclide examination could not accurately detect diseases of the small intestine. Double-balloon enteroscopy has been used to increase the detection rate of these diseases in addition to endoscopic biopsy and therapy. The purpose of this study was to determine the value of double-balloon enteroscopy in the diagnosis and treatment of diseases of the small intestine. Methods A total of 258 double-balloon enteroscopies via the mouth and/or anus were performed in 208 patients. If no lesion was detected on one side （mouth or anus）, examination on the other side （anus or mouth） was made. If active small intestinal bleeding was detected, endoscopic hemostasis was done to treat the bleeding. Results In the 208 patients, 50 were subjected to double-balloon enteroscopy via both mouth and anus. Lesions were detected in 151 patients, giving a rate of 72.6% （151/208）. The detection rates for obscure digestive tract bleeding, diarrhea, abdominal pain and weight loss were 90.2% （92/102）, 64.9% （24/37）, 48.5% （16/33） and 43.3% （13/30）, respectively. Lesions of the 151 patients were confirmed by endoscopic biopsy, surgery, clinical studies, and follow-up. In the 102 patients with bleeding of the digestive tract, active bleeding was detected in 27 patients. Endoscopic hemostasis was successful in 25 of them （92.6%, 25/27）. No serious complications occurred in all the patients, the average time for the procedure was 100 minutes. Conclusions Double-balloon enteroscopy is safe, effective in the diagnosis of diseases of the small intestine in addition to endoscopic therapy.

The role of capsule endoscopy combined with double-balloon enteroscopy in diagnosis of small bowel diseases

Background The diagnosis of small bowel diseases remains relatively inefficient using traditional imaging techniques. Capsule endoscopy （CE） and double-balloon enteroscopy （DBE） are two novel methods of enteroscopy for examining the entire small bowel. The aim of this study was to evaluate the detection rate and diagnostic accuracy of CE and DBE in patients with suspected small bowel diseases and to investigate the clinical significance of combined use of these two novel modalities. Methods Two hundred and eighteen patients were evaluated for suspected small bowel disease, including 116 with obscure gastrointestinal bleeding and 102 with obscure abdominal pain or chronic diarrhea. One hundred and sixty-five out of these patients underwent CE first and 53 patients underwent DBE （under anesthesia with propofol） first. DBE was recommended after negative or equivocal evaluation on CE and vise versa. Introduction of the endoscope during DBE was either orally or anally and the patients were referred for a second procedure using the opposite route several days later when no abnormalities were found on the first procedure. The detection rates, diagnostic accuracy, tolerance and frequency of adverse events of these two modalities were then analyzed. Results Failure of the procedure was seen in one patient with CE and in two patients with DBE. Sixty-four DBE procedures were carded out in 51 patients; by the oral route in 34 cases, the anal route in 4 and both routes in 13 cases. The overall detection rate of small bowel diseases using CE （72.0%, 118/164） was superior to that with DBE （41.2%, 21/51）; X2=16.1218, P〈0.0001. The diagnostic rate （51.8%, 85/164） was also higher than that with the latter procedure （39.2%, 20/51）, but was not significantly different （X^2=2.4771, P〉0.05）. Furthermore, the detection rate of small bowel diseases in patients with obscure gastrointestinal bleeding using CE （88.0%, 88/100） was superior to that of DBE （60.0%, 9/15）; X2=7.7457, P=0.0054. Lesions were detected by DBE in 1 out of 4 patients in whom CE had a negative result. Suspected findings by CE were confirmed by DBE combined with biopsy in 12 out of 15 patients. On the other hand, small bowel lesions were identified by CE in all 3 patients after negative evaluations by DBE. There were no severe complications during or after either of the two procedures. Conclusions The detection rate of small bowel diseases by CE is very high. CE should be selected for the initial diagnosis in patients with suspected small bowel diseases, especially in patients with obscure gastrointestinal bleeding. DBE appears to be inferior to CE in the diagnosis of small bowel diseases. However, it was shown that abnormalities could still be identified by DBE in patients with normal images or used to confirm suspected findings from CE. DBE can also serve as a good complementary approach after an initial diagnostic imaging using CE.