Adaptive recognition of machining features in sheet metal parts based on a graph class-incremental learning strategy

The integration of computer-aided design (CAD), computer-aided process planning (CAPP), and computer-aided manufacturing (CAM) systems is significantly enhanced by employing deep learning-based automatic feature recognition (AFR) methods. These methods outperform traditional, rule-based approaches, particularly in handling the complexities of intersecting features. However, existing deep learning-based AFR methods face two major challenges. The initial challenge stems from the frequent utilization of voxelized or point-cloud representations of CAD models, resulting in the unfortunate loss of valuable geometric and topological information inherent in original Boundary representation (B-Rep) models. The second challenge involves the limitation of supervised deep learning methods in identifying machining features that are not present in the predefined dataset. This constraint renders them suboptimal for the continually evolving datasets of real industrial scenarios. To address the first challenge, this study introduces a graph-structured language, Multidimensional Attributed Face-Edge Graph (maFEG), crafted to encapsulate the intricate geometric and topological details of CAD models. Furthermore, a graph neural network, Sheet-metalNet, is proposed for the efficient learning and interpretation of maFEGs. To tackle the second challenge, a three-component incremental learning strategy is proposed: an initial phase of pre-training and fine-tuning, a prototype sampling-based replay, and a stage employing knowledge distillation for parameter regularization. The effectiveness of Sheet-metalNet and its complementary incremental learning strategy is evaluated using the open-source MFCAD++ dataset and the newly created SMCAD dataset. Experimental results show that Sheet-metalNet surpasses state-of-the-art AFR methods in machining feature recognition accuracy. Moreover, Sheet-metalNet demonstrates adaptability to dynamic dataset changes, maintaining high performance when encountering newly introduced features, thanks to its innovative incremental learning strategy.

The prognostic value of pretreatment albumin-to-fibrinogen ratio in small cell lung cancer patients receiving first-line platinum-based chemotherapy

This study examined the role of pretreatment albumin-to-fibrinogen ratio (AFR) in the prognosis of small-cell lung cancer (SCLC) patients receiving first-line platinum-based chemotherapy. A total of 131 SCLC patients were enrolled. The predictive value of the AFR for progression free survival (PFS) and overall survival (OS) were evaluated by receiver operating characteristic (ROC) curve analysis. The predictive factor of survival was assessed by univariate and multivariate Cox proportional regression analysis. The correlation between OS, PFS and AFR was determined by the log-rank test using the Kaplan-Meier method. AFR was an effective predictor of OS in SCLC patients with a cut-off value of 7.78. AFR was independent risk factors for OS and PFS. Kaplan Meier analysis showed that PFS and OS in patients with high AFR levels were significantly higher than those with low AFR levels. These results suggest that AFR could be an effective predictor of survival in patients with SCLC.

Staged Acclimatization in a Failing Fontan by AFR in AFR: Ever-Decreasing Circles

We describe serial implantation of atrial flow regulator (AFR) devices in the Fontan fenestration of a 4-year-old patient. Initially, the fenestration size was decreased using a 6/5 AFR, resulting in improved saturations and hemodynamics. One year later, further improvement was achieved by placing a 4/10 AFR inside the original device. (Level of Difficulty: Advanced.).

Versatility of Atrial Flow Regulator Device in Congenital Heart Disease: A Case Series

The Atrial Flow Regulator (AFR) is a self-expandable double-disk device with a central fenestration, intended to maintain a calibrated communication across the interatrial septum. Only case reports and small case series have been published about its use in the pediatric and congenital heart disease (CHD) population. We described AFR implantation in three congenital patients with different anatomies and indications. In the first case, the AFR was deployed to create a stable fenestration in a Fontan conduit, while in the second, it was used to reduce a Fontan fenestration. In the third case, we implanted an AFR to decompress the left atrium of an adolescent with complex CHD in natural history, with complete mixing, ductal-dependent systemic circulation and combined pulmonary hypertension. This case series demonstrates the great potential of the AFR device in the CHD field, showing versatility, efficacy, and safety in establishing a calibrated and stable shunt, with promising hemodynamic and symptomatic benefits.

The diagnostic role of circulating inflammation-based biomarker in gallbladder carcinoma

AIM

To investigate the diagnostic roles of circulating inflammatory biomarkers in gallbladder carcinoma (GBC).

PATIENTS & METHODS

Circulating inflammatory cell count, fibrinogen, albumin, carcinoembryonic antigen (CEA) and CA199 were measured, neutrophil-to-lymphocyte ratio (NLR), dNLR, PLR, LMR and Alb-to-fib (AFR) were calculated in 306 GBC patients, 306 healthy and 305 benign controls. The reciever operating characteristic curve was used to determine diagnostic accuracy of them.

RESULTS

The area under curves of combined AFR, dNLR and lymphocyte were 0.943 and 0.985 for diagnosis of GBC from healthy and polyp controls, area under curve of combined AFR, CEA and CA199 was 0.90 for diagnosis of GBC from the cholecystitis patients.

CONCLUSION

Circulating AFR combined with lymphocyte and dNLR or CEA and CA199 could effectively distinguish GBC from the healthy and benign controls.

DNA and Protein Analyses to Confirm the Absence of Cross-Contamination and Support the Clinical Reliability of Extensively Hydrolysed Diets for Adverse Food Reaction-Pets

Adverse food reactions (AFR) are a common cause of skin diseases in cats and dogs. The correct diagnosis and management of AFR relies upon clinical nutrition. The reliability of commercial hypoallergenic diets commonly used in AFR has been questioned because studies have shown the presence of proteins not declared on the label ingredients. It is proposed that extensively hydrolysed protein-based diets constitute a reliable nutritional solution. Royal Canin Anallergenic™ Canine and Feline diets are formulated with very low molecular weight feather protein and purified corn starch. Protein gel electrophoresis and thin layer paper chromatography were used to characterize protein hydrolysis in these diets and their hydrolysed raw materials; protein species were identified by mass spectrometry. To detect cross-contaminating protein, species-specific DNA was measured and correlated with ancillary protein content using calibration curves. The only protein components detected in the extensively hydrolysed feather protein raw material were amino acids and small oligopeptides. GBSS-I (Granule-bound starch synthase 1) was detected in the finished diets; this has not been reported as a clinically apparent allergen in dogs or cats. The DNA threshold corresponding to the maximum acceptable level of ancillary protein was not exceeded in 99.9% of more than 2150 product batches tested and no products were released to the market with cross-contaminating proteins. These results demonstrate the extensive level of protein hydrolysis in Royal Canin Anallergenic™ Canine and Feline diets and the absence of cross-contaminating protein, both key requirements for a diet to be used during diagnosis and for management of pets with AFR.

Proteomic perspective of Quercus suber somatic embryogenesis

Quercus suber L. is a forest tree with remarkable ecological, social and economic value in the southern Europe ecosystems. To circumvent the difficulties of breeding such long-lived species like Q. suber in a conventional fashion, clonal propagation of Q. suber elite trees can be carried out, although this process is sometimes unsuccessful. To help decipher the complex program underlying the development of Q. suber somatic embryos from the first early stage until maturity, a proteomic approach based on DIGE and MALDI-MS has been envisaged. Results highlighted several key processes involved in the three developmental stages (proliferative, cotyledonary and mature) of Q. suber somatic embryogenesis studied. Results show that the proliferation stage is characterized by fermentation as an alternative energy source at the first steps of somatic embryo development, as well as by up-regulation of proteins involved in cell division. In this stage reactive oxygen species play a role in proliferation, while other proteins like CAD and PR5 seem to be implied in embryonic competence. In the transition to the cotyledonary stage diverse ROS detoxification enzymes are activated and reserve products (mainly carbohydrates and proteins) are accumulated, whereas energy production is increased probably to participate in the synthesis of primary metabolites such as amino acids and fatty acids. Finally, in the mature stage ethylene accumulation regulates embryo development.

BIOLOGICAL SIGNIFICANCE

Quercus suber L. is a forest tree with remarkable ecological, social and economic value in the southern Europe ecosystems. To circumvent the difficulties of breeding such long-lived species like Q. suber in a conventional fashion, clonal propagation of Q. suber elite trees can be carried out, although this process is sometimes unsuccessful. To help decipher the complex program underlying the development of Q. suber somatic embryos from the first early stage until maturity, in deep studies become necessary. This article is part of a Special Issue entitled: Translational Plant Proteomics.