Establishment and validation of a computer-assisted colonic polyp localization system based on deep learning.

BACKGROUND: Artificial intelligence in colonoscopy is an emerging field, and its application may help colonoscopists improve inspection quality and reduce the rate of missed polyps and adenomas. Several deep learning-based computer-assisted detection (CADe) techniques were established from small single-center datasets, and unrepresentative learning materials might confine their application and generalization in wide practice. Although CADes have been reported to identify polyps in colonoscopic images and videos in real time, their diagnostic performance deserves to be further validated in clinical practice. AIM: To train and test a CADe based on multicenter high-quality images of polyps and preliminarily validate it in clinical colonoscopies. METHODS: With high-quality screening and labeling from 55 qualified colonoscopists, a dataset consisting of over 71000 images from 20 centers was used to train and test a deep learning-based CADe. In addition, the real-time diagnostic performance of CADe was tested frame by frame in 47 unaltered full-ranged videos that contained 86 histologically confirmed polyps. Finally, we conducted a self-controlled observational study to validate the diagnostic performance of CADe in real-world colonoscopy with the main outcome measure of polyps per colonoscopy in Changhai Hospital. RESULTS: The CADe was able to identify polyps in the test dataset with 95.0% sensitivity and 99.1% specificity. For colonoscopy videos, all 86 polyps were detected with 92.2% sensitivity and 93.6% specificity in frame-by-frame analysis. In the prospective validation, the sensitivity of CAD in identifying polyps was 98.4% (185/188). Folds, reflections of light and fecal fluid were the main causes of false positives in both the test dataset and clinical colonoscopies. Colonoscopists can detect more polyps (0.90 vs 0.82, P < 0.001) and adenomas (0.32 vs 0.30, P = 0.045) with the aid of CADe, particularly polyps < 5 mm and flat polyps (0.65 vs 0.57, P < 0.001; 0.74 vs 0.67, P = 0.001, respectively). However, high efficacy is not realized in colonoscopies with inadequate bowel preparation and withdrawal time (P = 0.32; P = 0.16, respectively). CONCLUSION: CADe is feasible in the clinical setting and might help endoscopists detect more polyps and adenomas, and further confirmation is warranted.

Flurbiprofen axetil for postoperative analgesia in upper abdominal surgery: a randomized, parallel controlled, double-blind, multicenter clinical study.

PURPOSE: To investigate the efficacy and safety of flurbiprofen axetil in postoperative analgesia in upper abdominal surgery. METHODS: This was a multicenter, randomized, positive drug parallel controlled double-blind clinical study. Patients undergoing upper abdominal surgery were randomly divided to receive flurbiprofen axetil or tramadol. The VAS pain scores at rest and on coughing (pulmonary function training) were assessed immediately before drug usage (T1) to evaluate the efficacy of postoperative analgesia. Repeat assessment of the VAS was performed after T1. The timing of the recovery of the gastrointestinal function and the preoperative and postoperative IL-6, cortisol, and blood glucose levels were recorded as secondary endpoints. Vital signs and the occurrence of adverse reactions were evaluated for the assessment of safety. RESULTS: A total of 240 patients were enrolled in the current study; 119 used flurbiprofen axetil for postoperative analgesia. The VAS scores at rest and on coughing did not differ between the two groups to a statistically significant extent (P > 0.05). However, the reduction of the VAS score at rest in the flurbiprofen axetil group was greater than that in the tramadol group at 4-24 h after T1. The reduction of the VAS score on coughing at 8 h after T1 was greater in the flurbiprofen axetil group. The incidence of adverse reactions was significantly lower in the flurbiprofen axetil group, with only one adverse reaction recorded. In contrast, 18 adverse reactions were reported in the tramadol group. CONCLUSION: Flurbiprofen axetil showed superior efficacy to tramadol in early postoperative analgesia after upper abdominal surgery. Flurbiprofen axetil was associated with a significantly lower incidence of adverse reactions in comparison to tramadol.

Perioperative analgesia with parecoxib sodium improves postoperative pain and immune function in patients undergoing hepatectomy for hepatocellular carcinoma.

RATIONALE, AIMS, AND OBJECTIVES: Acute postoperative pain can result in immune dysfunction, which can be partly mitigated by efficient pain management. Opioids that have been widely applied to analgesia have been shown to suppress immune function, which has a negative impact on the treatment of patients with cancer. This study investigated the effects of perioperative fentanyl analgesia alone or in combination with parecoxib sodium on postoperative pain, immune function, and prognosis in patients undergoing hepatectomy of hepatocellular carcinoma (HCC). METHODS: A total of 80 patients scheduled for hepatectomy between October 2013 and August 2014 were included. Patients were randomly divided into two groups (n = 40) and allocated to receive parecoxibsodium 40 mg (group P) or placebo (group C) 30 minutes before induction of anaesthesia, followed by 40 mg every 12 hours for 48 hours after the operation. All patients had access to patient-controlled analgesia with intravenous fentanylpostoperatively. Venous blood samples were collected at the following time points: 30 minutes before induction of anaesthesia (T0), the end of the surgery (T1), 24 hours after surgery (T2), and 72 hours after surgery (T3). The percentages of CD3+, CD4+, CD8+, CD4+/CD8+ T cells, and CD3+CD16+CD56+ (NK) cells at these time points were quantified by flow cytometry (FCM).Visual analogue scale (VAS) scores, total fentanyl consumption, and adverse effects were recorded. The prognostic differences in overall survival (OS) and disease-free survival (DFS) between the two groups was also investigated. RESULTS: For both groups, the percentages of CD3+, CD4+ T cells, and the ratio of CD4+/CD8+ significantly decreased at T1 and T2 (P < .05). The percentages of CD3+ T cells were significantly lower in group C than that in group P at T2 (P < .05). In group C, the amount of CD3+ T cells was lower at T3 compared with T0 (P < .05). The percentages of NK cells significantly decreased at T1 in both groups (P < .05). The percentages of NK in group P were recovered nearly to baseline (T0) at T2, which was higher than that of group C (P < .05). In group C, the percentages of NK cells have not recovered nearly to baseline at T3 compared with T0 (P < .05). VAS scores at rest and on cough in group P were significantly lower than those in group C at 2, 6, 12, and 24 hours after operation (P < .05), and there were no significant differences in VAS scores between the two groups at 48 hours after surgery (P > .05). There were no significant differences regarding the incidence of adverse effects between the two groups (P > .05). Kaplan-Meier analysis indicated that the DFS time in group P was significantly longer than in group C (19.0 months, 95% confidence interval [CI], 9.8-28.2 vs 14.0 months, 95% CI, 8.1-19.9; P < .05). There was no significant difference in OS time (36.0 months, 95% CI, 13.4-58.9 vs 14.0 months, 95% CI, 10.6-25.4; P > .05) between two groups. CONCLUSIONS: The present study indicated that perioperative analgesia of parecoxib sodium combined with patient-controlled analgesic fentanyl resulted in better preserved immune function with enhancement of the analgesic efficacy to fentanyl alone of HCC patients undergoing hepatectomy and helped postpone postoperative tumour recurrence.

Low levels of tumor suppressor candidate 3 predict poor prognosis of patients with hepatocellular carcinoma.

PURPOSE: The tumor suppressor candidate 3 (TUSC3) has been considered to be closely associated with the occurrence, development and invasion of various malignant tumors. However, the expression of TUSC3 in hepatocellular carcinoma (HCC) tissues remains ambiguous. The purpose of this research was to investigate the expression of TUSC3 in HCC tissues and analyze the relationship between TUSC3 levels and clinicopathological characteristics and prognosis of HCC patients. MATERIALS AND METHODS: Immunohistochemistry was used to detect the expression of TUSC3 in HCC and the corresponding para-cancerous tissues from 92 samples of HCC patients. mRNA and protein expression levels of TUSC3 were evaluated by quantitative real-time polymerase chain reaction (qRT-PCR) and Western blot assays in 25 paired HCC and corresponding adjacent nontumor tissues. Furthermore, statistical analysis was applied to evaluate the correlation between TUSC3 level and the clinicopathological features and prognosis of HCC patients. RESULTS: Immunohistochemical assay indicated that the expression of TUSC3 was significantly lower in HCC tissues when compared with the corresponding para-cancerous tissues (χ2=11.512, P=0.001). The analysis of clinicopathological characteristics showed that low expression of TUSC3 in HCC tissues was significantly associated with Edmondson grade, Barcelona Clinic Liver Cancer stage and tumor size (P=0.008, 0.009 and 0.020, respectively). Univariate analysis showed that the expression of TUSC3 was strongly correlated with overall survival (OS) and disease-free survival (DFS) after radical surgery in HCC patients (P<0.001, P<0.001, respectively). Multivariate analysis revealed that the TUSC3 level was an independent risk factor for OS and DFS in HCC patients (P=0.001, P<0.001, respectively). Results of qRT-PCR and Western blot assays indicated that the level of TUSC3 in HCC tissues was significantly lower than that in the corresponding adjacent noncancerous tissues (P<0.01, P<0.001). CONCLUSION: The expression of TUSC3 in HCC was significantly downregulated and was correlated with tumor progression and prognosis, which could be used as an independent predictor of prognosis in HCC patients.