Development and validation of a mobile application based on a machine learning model to aid in predicting dosage of vitamin K antagonists among Indian patients post mechanical heart valve replacement

Patients who undergo heart valve replacements with mechanical valves need to take Vitamin K Antagonists (VKA) drugs (Warfarin, Nicoumalone) which has got a very narrow therapeutic range and needs very close monitoring using PT-INR. Accessibility to physicians to titrate drugs doses is a major problem in low-middle income countries (LMIC) like India. Our work was aimed at predicting the maintenance dosage of these drugs, using the de-identified medical data collected from patients attending an INR Clinic in South India. We used artificial intelligence (AI) - machine learning to develop the algorithm. A Support Vector Machine (SVM) regression model was built to predict the maintenance dosage of warfarin, who have stable INR values between 2.0 and 4.0. We developed a simple user friendly android mobile application for patients to use the algorithm to predict the doses. The algorithm generated drug doses in 1100 patients were compared to cardiologist prescribed doses and found to have an excellent correlation.

Gastrointestinal polyp detection in endoscopic images using an improved feature extraction method

Gastrointestinal polyps are treated as the precursors of cancer development. So, possibility of cancers can be reduced at a great extent by early detection and removal of polyps. The most used diagnostic modality for gastrointestinal polyps is video endoscopy. But, as an operator dependant procedure, several human factors can lead to miss detection of polyps. In this peper, an improved computer aided polyp detection method has been proposed. Proposed improved method can reduce polyp miss detection rate and assists doctors in finding the most important regions to pay attention. Color wavelet features and convolutional neural network features are extracted from endoscopic images, which are used for training a support vector machine. Then a target endoscopic image will be given to the classifier as input in order to find whether it contains any polyp or not. If polyp is found, it will be marked automatically. Experiment shows that, color wavelet features and convolutional neural network features together construct a highly representative of endoscopic polyp images. Evaluations on standard public databases show that, proposed system outperforms state-of-the-art methods, gaining accuracy of 98.34%, sensitivity of 98.67% and specificity of 98.23%. In this paper, the strength of color wavelet features and power of convolutional neural network features are combined. Fusion of these two methodology and use of support vector machine results in an improved method for gastrointestinal polyp detection. An analysis of ROC reveals that, proposed method can be used for polyp detection purposes with greater accuracy than state-of-the-art methods.

Classification of Diabetes Diet Problems Oriented by Deep Automatic Question Answering / 医学信息学杂志

Taking the diet problem of diabetic patients as an example,the paper puts forward the problems classification system based on functions in the view of users,classifies the problems put forward by patients through the Support Vector Machine (SVM) algorithm,and provides important support for the construction of the deep automatic Question Answering (QA) system.

Multiple mental tasks classification based on nonlinear parameter of mean period using support vector machines / 西安交通大学学报·英文版

Mental task classification is one of the most important problems in Brain-computer interface. This paper studies the classification of five-class mental tasks. The nonlinear parameter of mean period obtained from frequency domain information was used as features for classification implemented by using the method of SVM (support vector machines). The averaged classification accuracy of 85. 6% over 7 subjects was achieved for 2-second EEG segments. And the results for EEG segments of 0. 5s and 5. 0s compared favorably to those of Garrett's. The results indicate that the parameter of mean period represents mental tasks well for classification. Furthermore, the method of mean period is less computationally demanding, which indicates its potential use for online BCI systems.

Multiple mental tasks classification based on nonlinear parameter of mean period using support vector machines / 药物分析学报

Mental task classification is one of the most important problems in Brain-computer interface. This paper studies the classification of five-class mental tasks. The nonlinear parameter of mean period obtained from frequency domain information was used as features for classification implemented by using the method of SVM (support vector machines). The averaged classification accuracy of 85.6% over 7 subjects was achieved for 2-second EEG segments. And the results for EEG segments of 0.5s and 5.0s compared favorably to those of Garrett's. The results indicate that the parameter of mean period represents mental tasks well for classification. Furthermore, the method of mean period is less computationally demanding, which indicates its potential use for online BCI systems.

Support Vector Machine for Prediction of Horizontal Gene Transfers in Bacteria Genomes / 生物化学与生物物理进展

Horizontal gene transfer (HGT), also Lateral gene transfer (LGT), is any process in which an organism transfers genetic material to another species that is not its offspring. With the increase of available genomic data, it has become more convenient to study the way to detect the genes, which are products of horizontal transfers among a given genome. There are few data about known horizontal gene transfers in three bacterium genomes under consideration, so the experiments, which simulated gene transfer by artificially inserting phage genes, were carried out. Combining the feature analysis methods of gene sequences with support vector machine (SVM), a novel method was developed for identifying horizontal gene transfers (HGT) in 3 fully sequenced bacterium genomes (Escherichia coli K12, Borrelia burgdorferi, Bacillus cereus ZK). According to our previous work, codon use frequency (FCU) was selected as the sequence feature, in respect that it is inherently the fusion of both codon usage bias and amino acid composition signals. In addition, another computational method was proposed considering strand asymmetry and predicting horizontal gene transfers of leading strand and lagging strand of genomes under consideration, respectively. To avoid the occasionality of simulating gene transfer through artificially inserting phage genes, 100 times of the transfer-and-recover experiment were repeated and arithmetic average of measurement for each genome being considered were reported to evaluate algorithm's performance. Ten-fold cross-validation was used for both parameter and accuracy estimation. The best results were obtained for C-Support Vector Classification (C-SVC) type by using the radial basis function kernel with ?=100, while for one-class SVM type the best performance was obtained using the polynomial kernel of three degree. The performance of the approach was compared with that of Tsirigos' method ,which is one of the best predictive approachs to date in detecting of horizontal transfer genes. Firstly, for the original method that did not consider the strand asymmetry, the C-SVC type has a high relative improvement(RI) of 31.47% on hit ratio for Escherichia coli K12, while the one-class SVM type has RI of 11.61% for Borrelia burgdorferi. Moreover, as theoretically expected, the method considering the strand asymmetry resulted in higher RI than the original method. In order to examine the approach's performance in detecting factual gene transfer events, the approach was applied in genome of Enterococcus faecalis V583. It is not only succeed in recovering all the seven factual horizontally transferred genes, also found that the whole segment from 7 kb upstream of gene EF2293 to 38 kb downstream of gene EF2299 was probably transferred into E. faecalis V583 genome simultaneously with the above seven genes.

Research on classification method of burn scar hypertrophy by using support vector machine / 医疗卫生装备

This paper introduces a new method which can judge the degree of burn scar hypertrophy by analyzing chroma of the burn scar. Its technical schedule is as follows: Firstly, the image of the burn scar is captured by using a digital camera. Then the chroma emendation is performed by using an Artificial Neural Network(ANN). At last, the chroma of burn scar is analyzed and the classification of burn scar hypertrophy is given by using a Support Vector Machine(SVM). Compared with clinical evaluation, the result deduced from this method is proved to be effective.