Among various building information model (BIM) reconstruction methods for existing building, image-based method can identify building components from scanned as-built drawings and has won great attention due to its lower cost, less professional operators and better reconstruction performance. However, this kind of method will cost a great deal of time to design and extract features. Moreover, the manually extracted features have poor robustness and contain less non-geometric information. In order to solve this problem, this paper proposes a deep learning-based method to detect building components from scanned 2D drawings. Taking structural drawings as an example, in this article, 1500 images of structural drawings were firstly collected and preprocessed to guarantee the quality of data. After that, the neural network model—You Only Look Once (YOLO) was trained, verified and tested. In addition, a series of metrics were utilized to evaluate the performance of recognition. The results of test experiments show that the components in structural drawings (e.g., grid reference, column and beam) can be successfully detected, while the average detection accuracy of the whole image is over 80% and the average detection time for each image is 0.71 s. The experimental results demonstrate that the proposed method is robust and timesaving, which provides a good basis for the reconstruction of BIM from 2D drawings.
We consider nonparametric classification with smooth regression functions, where it is well known that notions of margin in E[Y |X] determine fast or slow rates in both active and passive learning. Here we elucidate a striking distinction between the two settings. Namely, we show that some seemingly benign nuances in notions of margin-involving the uniqueness of the Bayes classifier, and which have no apparent effect on rates in passive learning-determine whether or not any active learner can outperform passive learning rates. In particular, for Audibert-Tsybakov's margin condition (allowing general situations with non-unique Bayes classifiers), no active learner can gain over passive learning in commonly studied settings where the marginal on X is near uniform. Our results thus negate the usual intuition from past literature that active rates should improve over passive rates in nonparametric settings.
BackgroundAmoxicillin is a commonly used antibiotic which has a short half-life in human. The frequent administration of amoxicillin is often required to keep the plasma drug level in an effective range. The short dosing interval of amoxicillin could also cause some side effects and drug resistance, and impair its therapeutic efficacy and patients’ compliance. Therefore, a three-pulse release tablet of amoxicillin is desired to generate sustained release in vivo, and thus to avoid the above mentioned disadvantages.MethodsThe pulsatile release tablet consists of three pulsatile components: one immediate-release granule and two delayed release pellets, all containing amoxicillin. The preparation of a pulsatile release tablet of amoxicillin mainly includes wet granulation craft, extrusion/spheronization craft, pellet coating craft, mixing craft, tablet compression craft and film coating craft. Box–Behnken design, Scanning Electron Microscope and in vitro drug release test were used to help the optimization of formulations. A crossover pharmacokinetic study was performed to compare the pharmacokinetic profile of our in-house pulsatile tablet with that of commercial immediate release tablet. The pharmacokinetic profile of this pulse formulation was simulated by physiologically based pharmacokinetic (PBPK) model with the help of Simcyp®.Results and DiscussionSingle factor experiments identify four important factors of the formulation, namely, coating weight of Eudragit L30 D-55 (X1), coating weight of AQOAT AS-HF (X2), the extrusion screen aperture (X3) and compression forces (X4). The interrelations of the four factors were uncovered by a Box–Behnken design to help to determine the optimal formulation. The immediate-release granule, two delayed release pellets, together with other excipients, namely, Avicel PH 102, colloidal silicon dioxide, polyplasdone and magnesium stearate were mixed, and compressed into tablets, which was subsequently coated with Opadry® film to produce pulsatile tablet of amoxicillin. In vitro release study firstly indicated a three-pulse release profile of the tablet. Later the pulse tablet was found to generate the sustained release of amoxicillin in beagle dogs. Furthermore, the Simcyp® software was used to simulate the in vivo concentration time curve model of the three-pulse release tablet for amoxicillin in both human and beagle dog. The prediction by PBPK model nicely fitted the observation in human and beagle dog.ConclusionsThis study has demonstrated the interrelation of factors affecting the pulsatile formulation of amoxicillin using a Box–Behnken design. The three-pulse release tablets of amoxicillin were proven to generate pulsatile release in vitro and sustained release in vivo. This formulation was also found to extend the effective plasma concentration in human compared to the tablet of immediate release based on the simulation data by PBPK modeling. This study provides an example of using PBPK to guide the development of pulsatile dosage forms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.