Research trends on artificial intelligence and endoscopy in digestive diseases: A bibliometric analysis from 1990 to 2022.

BACKGROUND: Recently, artificial intelligence (AI) has been widely used in gastrointestinal endoscopy examinations. AIM: To comprehensively evaluate the application of AI-assisted endoscopy in detecting different digestive diseases using bibliometric analysis. METHODS: Relevant publications from the Web of Science published from 1990 to 2022 were extracted using a combination of the search terms "AI" and "endoscopy". The following information was recorded from the included publications: Title, author, institution, country, endoscopy type, disease type, performance of AI, publication, citation, journal and H-index. RESULTS: A total of 446 studies were included. The number of articles reached its peak in 2021, and the annual citation numbers increased after 2006. China, the United States and Japan were dominant countries in this field, accounting for 28.7%, 16.8%, and 15.7% of publications, respectively. The Tada Tomohiro Institute of Gastroenterology and Proctology was the most influential institution. "Cancer" and "polyps" were the hotspots in this field. Colorectal polyps were the most concerning and researched disease, followed by gastric cancer and gastrointestinal bleeding. Conventional endoscopy was the most common type of examination. The accuracy of AI in detecting Barrett's esophagus, colorectal polyps and gastric cancer from 2018 to 2022 is 87.6%, 93.7% and 88.3%, respectively. The detection rates of adenoma and gastrointestinal bleeding from 2018 to 2022 are 31.3% and 96.2%, respectively. CONCLUSION: AI could improve the detection rate of digestive tract diseases and a convolutional neural network-based diagnosis program for endoscopic images shows promising results.

Fat in mass in primary liver lesions: are usual MRI diagnostic criteria of both malignant and benign liver lesions still applicable?

PURPOSE: To evaluate the accuracy of the usual MRI diagnostic patterns of primary liver lesions applied to the diagnosis of pathologically proven fat-containing liver lesions. MATERIALS AND METHODS: This monocentric IRB approved retrospective study included all patients with pathologically proven focal liver lesions and documented intra-tumoral fat on pathology and who underwent preoperative liver MRI for characterization. Both liver morphology and usual lesion MRI features were analyzed and their distribution correlated to the final pathological result (Khi-2 or Fisher exact tests, Student t-test or Mann-Whitney U test, as appropriate). The Sensitivity (Se) and Specificity (Sp) of MRI patterns known to be associated to both Hepatocellular Carcinoma (HCC), Focal Nodular Hyperplasia (FNH), and Hepatocellular Adenoma (HCA) subtypes were evaluated. RESULTS: Between March 2014 and November 2021, 66 lesions were included, corresponding to: 26 HCC, 32 HCA, 6 FNH and 2 hepatic angiomyolipoma (HAML). All lesions developed on a dysmorphic liver were HCC. A non-rim arterial phase hyperenhancement with a washout and an enhancing capsule had a 98% specificity for HCC diagnosis; A homogeneous dropout of signal on the opposed phase had a sensitivity of 92% and a specificity of 89% for the diagnosis of HNF1alpha inactivated subtype of HCA (HHCA). The FNH pattern was specific at 100% for the diagnosis of FNH with a 40% Se. Finally, the accuracy of inflammatory hepatocellular adenoma (IHCA) pattern had a low 60% Se but a high 89% Sp for IHCA diagnosis. CONCLUSION: Known MRI patterns remain reliable for the accurate diagnosis of primary liver tumors on MRI even in fat-containing lesions.

[Development and thoughts of digestive endoscopy in children]. / å„¿ç«¥æ¶ˆåŒ–å† é•œçš„å‘å±•ä¸Žæ€è€ƒ.

After nearly 40 years of development, digestive endoscopy in children has been widely applied, and it has helped to expand the spectrum of pediatric digestive system diseases and greatly improve the diagnosis and treatment of pediatric digestive system diseases. Pediatric digestive endoscopy has become a subject. However, there are some problems such as the unbalanced development of pediatric digestive endoscopy across China, the lack of homogeneity in diagnosis and treatment system, the tendency of adult-oriented diagnosis and treatment techniques, and the localization of training quality, which affect the standardized and healthy development of pediatric digestive endoscopy. The diagnosis and treatment with digestive endoscopy in children should adhere to both pediatric characteristics and technological innovation to propose the concept of comfort, emphasize the importance of standardization (including the space and process for endoscopic diagnosis and treatment, perioperative evaluation, training mode, and access qualification), standardize the minimally invasive techniques, and develop artificial intelligence. It is of great importance to formulate related consensus statements and guidelines on the basis of medical safety and the features of the growth and development of children, so as to achieve the high-quality development of pediatric digestive endoscopy, effectively improve the diagnosis and treatment levels of pediatric digestive endoscopy, and bring benefits to more pediatric patients.

Application and prospect of artificial intelligence in digestive endoscopy.

INTRODUCTION: With the progress of science and technology, artificial intelligence represented by deep learning has gradually begun to be applied in the medical field. Artificial intelligence has been applied to benign gastrointestinal lesions, tumors, early cancer, inflammatory bowel disease, gallbladder, pancreas, and other diseases. This review summarizes the latest research results on artificial intelligence in digestive endoscopy and discusses the prospect of artificial intelligence in digestive system diseases. AREAS COVERED: We retrieved relevant documents on artificial intelligence in digestive tract diseases from PubMed and Medline. This review elaborates on the knowledge of computer-aided diagnosis in digestive endoscopy. EXPERT OPINION: Artificial intelligence significantly improves diagnostic accuracy, reduces physicians' workload, and provides a shred of evidence for clinical diagnosis and treatment. Shortly, artificial intelligence will have high application value in the field of medicine.

Fast and accurate automated recognition of the dominant cells from fecal images based on Faster R-CNN.

Fecal samples can easily be collected and are representative of a person's current health state; therefore, the demand for routine fecal examination has increased sharply. However, manual operation may pollute the samples, and low efficiency limits the general examination speed; therefore, automatic analysis is needed. Nevertheless, recognition exhaustion time and accuracy remain major challenges in automatic testing. Here, we introduce a fast and efficient cell-detection algorithm based on the Faster-R-CNN technique: the Resnet-152 convolutional neural network architecture. Additionally, a region proposal network and a network combined with principal component analysis are proposed for cell location and recognition in microscopic images. Our algorithm achieved a mean average precision of 84% and a 723 ms detection time per sample for 40,560 fecal images. Thus, this approach may provide a solid theoretical basis for real-time detection in routine clinical examinations while accelerating the process to satisfy increasing demand.

The Correlation Between UGT1A1 Gene Phenotypes and the Clinical Prognosis of Advanced Colorectal Cancer After FOLFIRI Therapy.

Objective: This study investigated the correlations between the different phenotypes of the uridine diphosphate glucuronyl transferase (UGT) 1A1 gene and the treatment of advanced colorectal cancer after the FOLFIRI regimen. Materials and Methods: A total of 240 advanced colorectal cancer patients with stage IV colon cancer or recurrence after radical surgery between January 2014 and December 2018 were included in a retrospective study. All participants were treated with the FOLFIRI regimen until the disease progressed or an intolerable level of toxicity occurred. Results: In this study, three phenotypes of the UGT1A1 gene promoter were found: the homozygous wild type (TA6/6 type, 78.3%), the heterozygous mutant type (TA6/7 type, 19.6%), and the homozygous mutant type (TA7/7 type, 2.1%). Compared with TA6/7 and TA6/6, the risk of nonresponse to FOLFIRI chemotherapy increased by 16%, but the difference was not significant. The risk of death increased by 24%, and there was no significant difference. There was a risk of hematologic and nonhematologic adverse reactions occurring in TA6/7 and TA6/6, and the total risk of adverse reactions increased by 9.3773 times among patients with more than two metastatic organs. Compared with patients with TA6/6, the risk of toxic side-effects increased by 42.8066 times (p = 0.0259) for patients with TA6/7. Among patients who received FOLFIRI chemotherapy for more than four cycles, the proportion with TA6/7 was greater than that with TA6/6. Compared with those with TA6/6, patients with TA6/7 showed a higher risk of hematologic toxicity (22.3246 times, p = 0.0035). Conclusion: The TA6/7 in patients with advanced colorectal cancer had more than two metastatic organs, and received FOLFIRI chemotherapy for more than four cycles compared with TA6/6 patients. Furthermore, the risk of hematologic and nonhematologic adverse reactions significantly increased, and the risk of digestive-tract and hematologic toxicity was more significant.

Disaster Diagnoses in Geriatric Patients with Abdominal Pain.

Care of geriatric patients with abdominal pain can pose significant diagnostic and therapeutic challenges to emergency physicians. Older adults rarely present with classic signs, symptoms, and laboratory abnormalities. The incidence of life-threatening emergencies, including abdominal aortic aneurysm, mesenteric ischemia, perforated viscus, and other surgical emergencies, is high. This article explores the evaluation and management of several important causes of abdominal pain in geriatric patients with an emphasis on high-risk presentations.

Assessment of Exposure to Mycotoxins in Spanish Children through the Analysis of Their Levels in Plasma Samples.

In this study, we present, for the first time in Spain, the levels of 19 mycotoxins in plasma samples from healthy and sick children (digestive, autism spectrum (ASD), and attention deficit hyperactivity (ADHD) disorders) (n = 79, aged 2-16). The samples were analyzed by liquid chromatography-mass spectrometry (triple quadrupole) (LC-MS/MS). To detect Phase II metabolites, the samples were reanalyzed after pre-treatment with ß-glucuronidase/arylsulfatase. The most prevalent mycotoxin was ochratoxin A (OTA) in all groups of children, before and after enzyme treatment. In healthy children, the incidence of OTA was 92.5% in both cases and higher than in sick children before (36.7% in digestive disorders, 50% in ASD, and 14.3% in ADHD) and also after the enzymatic treatment (76.6 % in digestive disorders, 50% in ASD, and 85.7% in ADHD). OTA levels increased in over 40% of healthy children after enzymatic treatment, and this increase in incidence and levels was also observed in all sick children. This suggests the presence of OTA conjugates in plasma. In addition, differences in OTA metabolism may be assumed. OTA levels are higher in healthy children, even after enzymatic treatment (mean OTA value for healthy children 3.29 ng/mL, 1.90 ng/mL for digestive disorders, 1.90 ng/mL for ASD, and 0.82 ng/mL for ADHD). Ochratoxin B appears only in the samples of healthy children with a low incidence (11.4%), always co-occurring with OTA. Sterigmatocystin (STER) was detected after enzymatic hydrolysis with a high incidence in all groups, especially in sick children (98.7% in healthy children and 100% in patients). This supports glucuronidation as a pathway for STER metabolism in children. Although other mycotoxins were studied (aflatoxins B1, B2, G1, G2, and M1; T-2 and HT-2 toxins; deoxynivalenol, deepoxy-deoxynivalenol, 3-acetyldeoxynivalenol, 15-acetyldeoxynivalenol; zearalenone; nivalenol; fusarenon-X; neosolaniol; and diacetoxyscirpenol), they were not detected either before or after enzymatic treatment in any of the groups of children. In conclusion, OTA and STER should be highly considered in the risk assessment of mycotoxins. Studies concerning their sources of exposure, toxicokinetics, and the relationship between plasma levels and toxic effects are of utmost importance in children.

Comparison of Diagnosis Accuracy between a Backpropagation Artificial Neural Network Model and Linear Regression in Digestive Disease Patients: an Empirical Research.

INTRODUCTION: A Noninvasive diagnosis model for digestive diseases is the vital issue for the current clinical research. Our systematic review is aimed at demonstrating diagnosis accuracy between the BP-ANN algorithm and linear regression in digestive disease patients, including their activation function and data structure. METHODS: We reported the systematic review according to the PRISMA guidelines. We searched related articles from seven electronic scholarly databases for comparison of the diagnosis accuracy focusing on BP-ANN and linear regression. The characteristics, patient number, input/output marker, diagnosis accuracy, and results/conclusions related to comparison were extracted independently based on inclusion criteria. RESULTS: Nine articles met all the criteria and were enrolled in our review. Of those enrolled articles, the publishing year ranged from 1991 to 2017. The sample size ranged from 42 to 3222 digestive disease patients, and all of the patients showed comparable biomarkers between the BP-ANN algorithm and linear regression. According to our study, 8 literature demonstrated that the BP-ANN model is superior to linear regression in predicting the disease outcome based on AUROC results. One literature reported linear regression to be superior to BP-ANN for the early diagnosis of colorectal cancer. CONCLUSION: The BP-ANN algorithm and linear regression both had high capacity in fitting the diagnostic model and BP-ANN displayed more prediction accuracy for the noninvasive diagnosis model of digestive diseases. We compared the activation functions and data structure between BP-ANN and linear regression for fitting the diagnosis model, and the data suggested that BP-ANN was a comprehensive recommendation algorithm.

Extra-pulmonary complications of 45 critically ill patients with COVID-19 in Yichang, Hubei province, China: A single-centered, retrospective, observation study.

ABSTRACT: Mortality of critically ill patients with coronavirus disease 2019 (COVID-19) was high. Aims to examine whether time from symptoms onset to intensive care unit (ICU) admission affects incidence of extra-pulmonary complications and prognosis in order to provide a new insight for reducing the mortality. A single-centered, retrospective, observational study investigated 45 critically ill patients with COVID-19 hospitalized in ICU of The Third People's Hospital of Yichang from January 17 to March 29, 2020. Patients were divided into 2 groups according to time from symptoms onset to ICU admission (>7 and ≤7 days) and into 2 groups according to prognosis (survivors and non-survivors). Epidemiological, clinical, laboratory, radiological characteristics and treatment data were studied. Compared with patients who admitted to the ICU since symptoms onset ≤7 days (55.6%), patients who admitted to the ICU since symptoms onset >7 days (44.4%) were more likely to have extra-pulmonary complications (19 [95.0%] vs 16 [64.0%], P = .034), including acute kidney injury, cardiac injury, acute heart failure, liver dysfunction, gastrointestinal hemorrhage, hyperamylasemia, and hypernatremia. The incidence rates of acute respiratory distress syndrome, pneumothorax, and hospital-acquired pneumonia had no difference between the 2 groups. Except activated partial thromboplastin and Na+ concentration, the laboratory findings were worse in group of time from symptoms onset to ICU admission >7 days. There was no difference in mortality between the 2 groups. Of the 45 cases in the ICU, 19 (42.2%) were non-survivors, and 16 (35.6%) were with hospital-acquired pneumonia. Among these non-survivors, hospital-acquired pneumonia was up to 12 (63.2%) besides higher incidence of extra-pulmonary complications. However, hospital-acquired pneumonia occurred in only 4 (15.4%) survivors. Critically ill patients with COVID-19 who admitted to ICU at once might get benefit from intensive care via lower rate of extra-pulmonary complications.

Postoperative complications and mortality following emergency digestive surgery during the COVID-19 pandemic: A multicenter collaborative retrospective cohort study protocol (COVID-CIR).

ABSTRACT: Infection with the SARS-CoV-2 virus seems to contribute significantly to increased postoperative complications and mortality after emergency surgical procedures. Additionally, the fear of COVID-19 contagion delays the consultation of patients, resulting in the deterioration of their acute diseases by the time of consultation. In the specific case of urgent digestive surgery patients, both factors significantly worsen the postoperative course and prognosis. Main working hypothesis: infection by COVID-19 increases postoperative 30-day-mortality for any cause in patients submitted to emergency/urgent general or gastrointestinal surgery. Likewise, hospital collapse during the first wave of the COVID-19 pandemic increased 30-day-mortality for any cause. Hence, the main objective of this study is to estimate the cumulative incidence of mortality at 30-days-after-surgery. Secondary objectives are: to estimate the cumulative incidence of postoperative complications and to develop a specific postoperative risk propensity model for COVID-19-infected patients.A multicenter, observational retrospective cohort study (COVID-CIR-study) will be carried out in consecutive patients operated on for urgent digestive pathology. Two cohorts will be defined: the "pandemic" cohort, which will include all patients (classified as COVID-19-positive or -negative) operated on for emergency digestive pathology during the months of March to June 2020; and the "control" cohort, which will include all patients operated on for emergency digestive pathology during the months of March to June 2019. Information will be gathered on demographic characteristics, clinical and analytical parameters, scores on the usual prognostic scales for quality management in a General Surgery service (POSSUM, P-POSSUM and LUCENTUM scores), prognostic factors applicable to all patients, specific prognostic factors for patients infected with SARS-CoV-2, postoperative morbidity and mortality (at 30 and 90 postoperative days). The main objective is to estimate the cumulative incidence of mortality at 30 days after surgery. As secondary objectives, to estimate the cumulative incidence of postoperative complications and to develop a specific postoperative risk propensity model for SARS-CoV-2 infected patients.The protocol (version1.0, April 20th 2020) was approved by the local Institutional Review Board (Ethic-and-Clinical-Investigation-Committee, code PR169/20, date 05/05/20). The study findings will be submitted to peer-reviewed journals and presented at relevant national and international scientific meetings.ClinicalTrials.gov Identifier: NCT04479150 (July 21, 2020).

Comparison of Aerodigestive and Nonaerodigestive Provider Responses to Clinical Case Vignettes.

OBJECTIVE: To evaluate differences in practice patterns between aerodigestive and nonaerodigestive providers in pediatric gastroenterology when diagnosing and treating common aerodigestive complaints. STUDY DESIGN: A questionnaire comprised of clinical vignettes with multiple-choice questions was distributed to both aerodigestive and nonaerodigestive pediatric gastroenterologists. Vignettes focused on management of commonly encountered general gastroenterology and aerodigestive issues, such as gastroesophageal (GE) reflux, aspiration, and feeding difficulties. Tests of equal proportions were used to compare rates of testing and empiric therapy within and across groups. Multivariate analysis was used to assess differences in response rates between aerodigestive and nonaerodigestive providers. RESULTS: A total of 88 pediatric gastroenterologists from 18 institutions completed the questionnaire. There were 35 aerodigestive gastroenterology providers and 53 nonaerodigestive gastroenterology providers. The nonaerodigestive group included 31 general gastroenterologists and 22 providers with self-identified subspecialty gastroenterology expertise. Aerodigestive specialists were more likely than nonaerodigestive gastroenterologists to pursue testing over empiric therapy in cases involving isolated respiratory symptoms (P < .05); aerodigestive providers were more likely to recommend pH-impedance testing, videofluoroscopic swallow studies, and upper gastrointestinal barium study (P < .05 for each test) depending on the referring physician. For vignettes involving infant GE reflux, both groups chose empiric treatments more frequently than testing (P < .001), although aerodigestive providers were more likely than nonaerodigestive providers to pursue testing like upper gastrointestinal barium studies (P < .05). CONCLUSIONS: Although some practice patterns were similar between groups, aerodigestive providers pursued more testing than nonaerodigestive providers in several clinical scenarios including infants with respiratory symptoms and GE reflux.