Comparison of uncut Roux-en-Y anastomosis and Billroth-II with Braun anastomosis after distal gastrectomy.

Background: This study aimed to compare the clinical outcomes and patient benefits of uncut Roux-en-Y (URY) anastomosis and Billroth-II with Braun (BB) anastomosis after distal gastrectomy. Methods: We retrospectively reviewed the data of patients who underwent URY or BB anastomosis after distal gastrectomy between March 2015 and December 2017. Clinical characteristics, survival data, postoperative recovery data, and long-term outcomes were recorded and compared between the two groups. Results: A total of 231 patients were included, with 167 in the URY group and 64 in the BB group. Kaplan-Meier curves for overall survival showed no differences after propensity score matching (p = 0.488). Long-term postoperative quality of life evaluation also showed no significant differences. Compared to the BB group, patients in the URY group had a significantly shorter time to start a liquid diet after propensity score matching (67.6 h vs. 46.5 h, p = 0.003), and a lower occurrence of bile reflux on follow-up gastroscopy (p < 0.001). Conclusion: The URY anastomosis appears to be a feasible method for digestive tract reconstruction after distal gastrectomy, resulting in less bile reflux and better postoperative recovery. However, there is no significant difference between URY and BB anastomosis in terms of overall survival and long-term quality of life.

Modified Landweber algorithm for robust particle sizing by using Fraunhofer diffraction.

In this paper, a robust modified Landweber algorithm was proposed to retrieve the particle size distributions from Fraunhofer diffraction. Three typical particle size distributions, i.e., Rosin-Rammler, lognormal, and bimodal normal distributions for particles ranging from 4.8 to 96 µm, were employed to verify the performance of the algorithm. To show its merits, the proposed algorithm was compared with the Tikhonov regularization algorithm and the ℓ1-norm-based algorithm. Simulation results showed that, for noise-free data, both the modified Landweber algorithm and the ℓ1-norm-based algorithm were better than the Tikhonov regularization algorithm in terms of accuracy. When the data was noise-contaminated, the modified Landweber algorithm was superior to the other two algorithms in both accuracy and speed. An experimental setup was also established and the results validated the feasibility and effectiveness of the proposed method.