Sequential Support Vector Machine Research Articles

Arterial Distension Monitoring Scheme Using FPGA-Based Inference Machine in Ultrasound Scanner Circuit System.

This paper presents an arterial distension monitoring scheme using a field-programmable gate array (FPGA)-based inference machine in an ultrasound scanner circuit system. An arterial distension monitoring requires a precise positioning of an ultrasound probe on an artery as a prerequisite. The proposed arterial distension monitoring scheme is based on a finite state machine that incorporates sequential support vector machines (SVMs) to assist in both coarse and fine adjustments of probe position. The SVMs sequentially perform recognitions of ultrasonic A-mode echo pattern for a human carotid artery. By employing sequential SVMs in combination with convolution and average pooling, the number of features for the inference machine is significantly reduced, resulting in less utilization of hardware resources in FPGA. The proposed arterial distension monitoring scheme was implemented in an FPGA (Artix7) with a resource utilization percentage less than 9.3%. To demonstrate the proposed scheme, we implemented a customized ultrasound scanner consisting of a single-element transducer, an FPGA, and analog interface circuits with discrete chips. In measurements, we set virtual coordinates on a human neck for 9 human subjects. The achieved accuracy of probe positioning inference is 88%, and the Pearson coefficient (r) of arterial distension estimation is 0.838.

Integrated support vector machine and pharmacophore based virtual screening driven identification of thiophene carboxamide scaffold containing compound as potential PARP1 inhibitor

Poly (ADP-ribose) polymerase-1 (PARP1) inhibition strategy for cancer treatment is gaining advantage particularly in patients having a mutation in BRCA1/BRCA2 gene. To date, four drugs have obtained FDA approval and some inhibitors are in clinical trials. To identify more potent PARP1 inhibitors extensive research is going on to enrich the library of PARP1 inhibitors with compounds belonging to different classes. We employed an integrated virtual screening approach to identify potential PARP1 inhibitors. The sequential support vector machine (SVM) and pharmacophore model based virtual screening was carried out on the Maybridge library. The obtained hits were docked in the binding site of the PARP1 catalytic domain and nine drug-like compounds showing good ADME properties and form critical molecular interactions with the binding site residues were considered for the in vitro PARP1 inhibition assay. MD simulations were performed to decipher the stability of the PARP1-ligand complexes. Hydrogen bond interactions were also probed for their stability during MD simulations. We have identified three compounds (BTB02767, GK01172, and KM09200) showing 50% inhibition of PARP1 enzyme activity at 25 μM. BTB02767 and KM09200 have phthalazinone scaffold, while GK01172 bears a thiophene carboxamide scaffold, which could be a new chemotype of PARP1 inhibitors. In conclusion, GK01172 may serve as an important compound for further development of PARP1 inhibitors containing thiophene carboxamide scaffold. Communicated by Ramaswamy H. Sarma

Fast basin stability estimation for dynamic systems under large perturbations with sequential support vector machine

The vulnerability of real-world dynamic systems or processes to large perturbations has nullified the conventional linear stability analysis. Despite the theoretical advance on computation of Lyapunov functions for dynamics analysis, they oftentimes enlist subtle construction technique and confounding process, and thus lack broad applications. As a global stability measure, basin stability utilizes the volume of the basin of attraction to quantify the stability of system dynamics, and has been thrown to sharp relief recently. Concretely, the perturbation manifests as a distribution of the initial conditions of system states. Monte Carlo simulation, or the Bernoulli trials of the time-domain simulation at the accessible realization of initial conditions, is the state of the art to quantify basin stability. However, this inevitably involves daunting computational budget. At its core, basin stability estimation is a classification problem on each sampling point in the state space at the onset of the possibly large perturbation. Hence, we propose a sequential support vector machine (SVM) framework to efficiently locate the stability boundaries and delineate the basin of attraction. The high-fidelity time-domain simulation is only called upon those sampling points that are predicted by SVM to be close to the boundaries, and the SVM as well as the boundaries are sequentially updated with the addition of those newly evaluated sampling points. This innovative sequential approach reduces over 90% of the computational cost in the conventional time-domain simulation.

Multicategory Classification via Forward-Backward Support Vector Machine.

In this paper, we propose a new algorithm to extend support vector machine (SVM) for binary classification to multicategory classification. The proposed method is based on a sequential binary classification algorithm: we first classify a target class by excluding the possibility of labeling as any other classes using a forward step of sequential SVM; we then exclude the already classified classes and repeat the same procedure for the remaining classes in a backward step. The proposed algorithm relies on SVM for each binary classification and utilizes only feasible data in each step; therefore, the method guarantees convergence and entails light computational burden. We prove Fisher consistency of the proposed forward-backward-SVM (FB-SVM) and obtain a stochastic bound for the predicted misclassification rate. We conduct extensive simulations and analyze real-world data to demonstrate the superior performance of FB-SVM, for example, FB-SVM achieves a classification accuracy much higher than the current standard for predicting conversion from mild cognitive impairment to Alzheimer's disease.

Prediction of membrane protein types by exploring local discriminative information from evolutionary profiles

Membrane protein is a pivotal constituent of a cell that exerts a crucial influence on diverse biological processes. The accurate identification of membrane protein types is deeply essential for revealing molecular mechanisms and drug development. Primarily, several traditional methods were exploited to classify these types. However, experimental methods are laborious, time-consuming, and costly due to rapid exploration of uncharacterized protein sequences generated in the postgenomic era. Hence, machine learning-based methods are more indispensable for reliable and fast identification of membrane protein types. A variety of state-of-the-art investigations have been elucidated to improve prediction performance, but predictive validity is still insufficient. Motivated by this, we designed a promising sequential support vector machine based predictor called TargetHMP to predict types of membrane proteins. We captured the local informative features by exploring evolutionary profiles through a novel method called the segmentation-based pseudo position-specific scoring matrix (Seg-PsePSSM). TargetHMP attained high accuracy of 94.99%, 93.48%, and 90.36% on the S1, S2, and S3 datasets, respectively, using a vigorous leave-one-out-cross-validation test. The results indicate that the performance of the proposed method outperformed prior predictors. We expect that the proposed approach will help research academia in general and pharmaceutical drug discovery in particular.

DiP-SVM : Distribution Preserving Kernel Support Vector Machine for Big Data

In literature, the task of learning a support vector machine for large datasets has been performed by splitting the dataset into manageable sized “partitions” and training a sequential support vector machine on each of these partitions separately to obtain local support vectors. However, this process invariably leads to the loss in classification accuracy as global support vectors may not have been chosen as local support vectors in their respective partitions. We hypothesize that retaining the original distribution of the dataset in each of the partitions can help solve this issue. Hence, we present DiP-SVM, a distribution preserving kernel support vector machine where the first and second order statistics of the entire dataset are retained in each of the partitions. This helps in obtaining local decision boundaries which are in agreement with the global decision boundary, thereby reducing the chance of missing important global support vectors. We show that DiP-SVM achieves a minimal loss in classification accuracy among other distributed support vector machine techniques on several benchmark datasets. We further demonstrate that our approach reduces communication overhead between partitions leading to faster execution on large datasets and making it suitable for implementation in cloud environments.

Sequential support vector machine classification for small-grain weed species discrimination with special regard to Cirsium arvense and Galium aparine

Site-specific weed management can reduce the amount of herbicides used in comparison to classical broadcast applications. The ability to apply herbicides on weed patches within the field requires automation. This study focuses on the automatic detection of different species with imaging sensors. Image processing algorithms determine shape features for the plants in the images. With these shape descriptions classification algorithms can be trained to identify the weed and crop species. Since weeds differ in their economic loss due to their yield effect and are controlled by different herbicides, it is necessary to correctly distinguish between the species. Image series of different measurements with plant samples at different growth stages were analysed. For the classification a sequential classification approach was chosen, involving three different support vector machine (SVM) models. In a first step groups of similar plant species were successfully identified (monocotyledons, dicotyledons and barley). Distinctions within the class of dicotyledons proved to be particularly difficult. For that purpose species in this group were subject to a second and third classification step. For each of these steps different features were found to be most important. Feature weighting was done with the RELIEF-F algorithm and SVM-Weighting. The focus was on the early identification of the two most harmful species Cirsium arvense and Galium aparine, with optimal accuracy than using a non-sequential classification approach. An overall classification accuracy of 97.7% was achieved in the first step. For the two subsequent classifiers accuracy rates of 80% and more were obtained for C. arvense and G. aparine.