Enhanced segmentation of gastrointestinal polyps from capsule endoscopy images with artifacts using ensemble learning.

BACKGROUND: Endoscopy artifacts are widespread in real capsule endoscopy (CE) images but not in high-quality standard datasets. AIM: To improve the segmentation performance of polyps from CE images with artifacts based on ensemble learning. METHODS: We collected 277 polyp images with CE artifacts from 5760 h of videos from 480 patients at Guangzhou First People's Hospital from January 2016 to December 2019. Two public high-quality standard external datasets were retrieved and used for the comparison experiments. For each dataset, we randomly segmented the data into training, validation, and testing sets for model training, selection, and testing. We compared the performance of the base models and the ensemble model in segmenting polyps from images with artifacts. RESULTS: The performance of the semantic segmentation model was affected by artifacts in the sample images, which also affected the results of polyp detection by CE using a single model. The evaluation based on real datasets with artifacts and standard datasets showed that the ensemble model of all state-of-the-art models performed better than the best corresponding base learner on the real dataset with artifacts. Compared with the corresponding optimal base learners, the intersection over union (IoU) and dice of the ensemble learning model increased to different degrees, ranging from 0.08% to 7.01% and 0.61% to 4.93%, respectively. Moreover, in the standard datasets without artifacts, most of the ensemble models were slightly better than the base learner, as demonstrated by the IoU and dice increases ranging from -0.28% to 1.20% and -0.61% to 0.76%, respectively. CONCLUSION: Ensemble learning can improve the segmentation accuracy of polyps from CE images with artifacts. Our results demonstrated an improvement in the detection rate of polyps with interference from artifacts.

Improvements in Brazed-Joint Properties of Silicon Nitride and Titanium Alloys Using Laser-Induced Microscale Rice Leaf Structures.

Si3N4 ceramics with a microscale rice leaf structure (MRLS) and titanium alloy were connected via brazing, and the influence of the surface microstructure on the ceramic connection was analyzed. MRLS fabrication is an efficient and high-degree-of-freedom method that can be used to change a material's surface morphology and wettability. The MRLS was obtained at a laser power of 110 W, with line spacings of 100 and 50 µm. The laser-treated surface included nanoparticles and micro particles, exhibiting a coral-like structure after agglomeration. When the MRLS was used to braze the titanium alloy, no defects were observed at the brazing interface, and the formation was excellent. Throughout the brazed joint, the MRLS remained intact and formed a strong metallurgical bond with the brazing filler metal. A finite element analysis was performed to study the cross-sectional morphology after joint fracture; from the load-time curve, it was found that the MRLS on the surface not only helped improve the mechanical occlusion and brazing area at the interface, but also helped generate compressive stress on the Si3N4 side. Crack propagation was hindered, thereby increasing the joint strength.

Fabrication of Metallic Superhydrophobic Surfaces with Tunable Condensate Self-Removal Capability and Excellent Anti-Frosting Performance.

Laser fabrication of metallic superhydrophobic surfaces (SHSs) for anti-frosting has recently attracted considerable attention. Effective anti-frosting SHSs require the efficient removal of condensed microdroplets through self-propelled droplet jumping, which is strongly influenced by the surface morphology. However, detailed analyses of the condensate self-removal capability of laser-structured surfaces are limited, and guidelines for laser processing parameter control for fabricating rationally structured SHSs for anti-frosting have not yet been established. Herein, a series of nanostructured copper-zinc alloy SHSs are facilely constructed through ultrafast laser processing. The surface morphology can be properly tuned by adjusting the laser processing parameters. The relationship between the surface morphologies and condensate self-removal capability is investigated, and a guideline for laser processing parameterization for fabricating optimal anti-frosting SHSs is established. After 120 min of the frosting test, the optimized surface exhibits less than 70% frost coverage because the remarkably enhanced condensate self-removal capability reduces the water accumulation amount and frost propagation speed (<1 µm/s). Additionally, the material adaptability of the proposed technique is validated by extending this methodology to other metals and metal alloys. This study provides valuable and instructive insights into the design and optimization of metallic anti-frosting SHSs by ultrafast laser processing.

[The effect of proton pump inhibitor on intragastric acidity and it relation to S-mephenytoin hydroxylase genetic polymorphism].

OBJECTIVE: To study the effect of proton pump inhibitor rabeprazole (Rab) and omeprazole (Ome) on intragastric acidity and it relationship to S-mephenytoin hydroxylase (CYP2C19) genetic polymorphism. METHODS: Thirty duodenal ulcer patients were randomly divided into two groups. They were treated by the following strategies: Rab 10mg or Ome 20mg twice daily orally for two days. Intragastric pH over 24 hours was recorded from morning of the second day. Blood CYP2C19 genotyping of each patient was detected by PCR-restriction enzyme analysis. RESULTS: In Rab group, the mean intragastric pH values of the PM, hetEM and homEM were 6.3 +/- 0.1, 6.1 +/- 0.2, 6.0 +/- 0.1, no significant difference was observed among the three different genotype groups. In Ome group, they were 6.3 +/- 0.1, 5.9 +/- 0.2, 5.6 +/- 0.3. The differences were significant (P < 0.05). After treated with Rab for 24 hours, the mean and median 24 h intragastric pH values were significant higher than that of Ome (6.1 +/- 0.2, 6.1 +/- 0.3 vs 5.8 +/- 0.4, 5.7 +/- 0.4, P < 0.05). The reason was that big difference existed in homEM type between Rab (6.0 +/- 0.1) than Ome (5.6 +/- 0.3) group. NAB occurred in 3 patients of each group. The sustaining time of NAB in Rab group was shorter than that of Ome group [(2.6 +/- 0.2) h vs (3.4 +/- 0.6) h]. CONCLUSIONS: The effect of Rab on control of intragastric pH was less affected by an individual's CYP2C19 status. Both Rab and Ome could not overcome NAB, but Rab could shorten the sustaining time of nocturnal acid breakthrough.

[Long-term efficacy of lactulose in patients with subclinical hepatic encephalopathy].

OBJECTIVE: To observe the long-term clinical efficacy of lactulose in patients with subclinical hepatic encephalopathy (SHE) and the natural history of this disease. METHODS: Patients with SHE diagnosed with psychometric tests (number connection test and digit symbol test) were randomly allocated to control group (Cg), short-term (Stg) and long-term treatment group (Ltg), respectively. Lactulose (15 - 90 ml/day) was orally given to 21 patients in Stg for 8 weeks, and 22 in Ltg for 24 weeks.The drug was not given to 21 patients in Cg. Psychometric test, somatosensory evoked potentials (SEP) and venous blood ammonia concentration (BA) were investigated before and after treatment at 8 week intervals for 6 months of follow-up. RESULTS: In Ltg, psychometric tests were significantly improved, BA was lowered and N(50) peak latency in SEP was not prolonged as compared with those in Cg, in which psychometric tests and N(50) peak latency gradually worsened during the follow-up. After 8 weeks of treatment, psychometric tests, BA and N(50) peak latency were significantly improved in Stg, but at week 24 they were deteriorated after cessation of treatment. The development of clinical hepatic encephalopathy was more frequent in Cg (40%) and Stg (30%) than in Ltg (5%). One patient each was not followed up after week 24 in Cg and Stg, respectively and 2 patients in Ltg stopped lactulose because of severe diarrhea. CONCLUSIONS: In patients with SHE, long-term treatment with lactulose could lower BA, improve psychometric tests, prevent SEP from getting worse, and eventually cut down the occurrence rate of clinical hepatic encephalopathy.