Detection of Typical Compensatory Movements during Autonomously Performed Exercises Preventing Low Back Pain (LBP)

With the growing number of people seeking medical advice due to low back pain (LBP), individualised physiotherapeutic rehabilitation is becoming increasingly relevant. Thirty volunteers were asked to perform three typical LBP rehabilitation exercises (Prone-Rocking, Bird-Dog and Rowing) in two categories: clinically prescribed exercise (CPE) and typical compensatory movement (TCM). Three inertial sensors were used to detect the movement of the back during exercise performance and thus generate a dataset that is used to develop an algorithm that detects typical compensatory movements in autonomously performed LBP exercises. The best feature combinations out of 50 derived features displaying the highest capacity to differentiate between CPE and TCM in each exercise were determined. For classifying exercise movements as CPE or TCM, a binary decision tree was trained with the best performing features. The results showed that the trained classifier is able to distinguish CPE from TCM in Bird-Dog, Prone-Rocking and Rowing with up to 97.7% (Head Sensor, one feature), 98.9% (Upper back Sensor, one feature) and 80.5% (Upper back Sensor, two features) using only one sensor. Thus, as a proof-of-concept, the introduced classification models can be used to detect typical compensatory movements in autonomously performed LBP exercises.

Multiparametric Ultrasound Approach Using a Tree-Based Decision Classifier for Inconclusive Focal Liver Lesions Evaluated by Contrast Enhanced Ultrasound

Background: Multiparametric ultrasound (MPUS) is a concept whereby the examiner is encouraged to use the latest features of an ultrasound machine. The aim of this study was to reanalyze inconclusive focal liver lesions (FLLs) that had been analyzed via contrast enhanced ultrasound (CEUS) using the MPUS approach with the help of a tree-based decision classifier. Materials and methods: We retrospectively analyzed FLLs that were inconclusive upon CEUS examination in our department, focusing our attention on samples taken over a period of two years (2017−2018). MPUS reanalysis followed a three-step algorithm, taking into account the liver stiffness measurement (LSM), time–intensity curve analysis (TIC), and parametric imaging (PI). After processing all steps of the algorithm, a binary decision tree classifier (BDTC) was used to achieve a software-assisted decision. Results: Area was the only TIC-CEUS parameter that showed a significant difference between malign and benign lesions with a cutoff of >−19.3 dB for washout phenomena (AUROC = 0.58, Se = 74.0%, Sp = 45.7%). Using the binary decision tree classifier (BDTC) algorithm, we correctly classified 71 out of 91 lesions according to their malignant or benignant status, with an accuracy of 78.0% (sensitivity = 62%, specificity = 45%, and precision = 80%). Conclusions: By reevaluating inconclusive FLLs that had been analyzed via CEUS using MPUS, we managed to determine that 78% of the lesions were malignant and, in 28% of them, we established the lesion type.

An Algorithm for Notifiable Disease Modeling and Prediction Using Artificial Intelligence Techniques: A Case of Kenya.

BackgroundThe disease outbreak management operations of most countries (notably Kenya) present numerous novel ideas of how to best make use of notifiable disease data to effect proactive interventions. Notifiable disease data is reported, aggregated and variously consumed. Over the years, there has been a deluge of notifiable disease data and the challenge for notifiable disease data management entities has been how to objectively and dynamically aggregate such data in a manner such as to enable the efficient consumption to inform relevant mitigation measures. Various models have been explored, tried and tested with varying results; some purely mathematical and statistical, others quasi-mathematical cum software model-driven.MethodsOne of the tools that has been explored is Artificial Intelligence (AI). AI is a technique that enables computers to intelligently perform and mimic actions and tasks usually reserved for human experts. AI presents a great opportunity for redefining how the data is more meaningfully processed and packaged. This research explores AI’s Machine Learning (ML) theory as a differentiator in the crunching of notifiable disease data and adding perspective. An algorithm has been designed to test different notifiable disease outbreak data cases, a shift to managing disease outbreaks via the symptoms they generally manifest. Each notifiable disease is broken down into a set of symptoms, dubbed symptom burden variables, and consequently categorized into eight clusters: Bodily, Gastro-Intestinal, Muscular, Nasal, Pain, Respiratory, Skin, and finally, Other Symptom Clusters. ML’s decision tree theory has been utilized in the determination of the entropies and information gains of each symptom cluster based on select test data sets.ResultsOnce the entropies and information gains have been determined, the information gain variables are then ranked in descending order; from the variables with the highest information gains to those with the lowest, thereby giving a clear-cut criteria of how the variables are ordered. The ranked variables are then utilized in the construction of a binary decision tree, which graphically and structurally represents the variables. Should any variables have a tie in the information gain rankings, such are given equal importance in the construction of the binary decision-tree. From the presented data, the computed information gains are ordered as; Gastro-Intestinal, Bodily, Pain, Skin, Respiratory, Others. Muscular, and finally Nasal Symptoms respectively. The corresponding binary decision tree is then constructed.ConclusionsThe algorithm successfully singles out the disease burden variable(s) that are most critical as the point of diagnostic focus to enable the relevant authorities take the necessary, informed interventions. This algorithm provides a good basis for a country’s localized diagnostic activities driven by data from the reported notifiable disease cases. The algorithm presents a dynamic mechanism that can be used to analyze and aggregate any notifiable disease data set, meaning that the algorithm is not fixated or locked on any particular data set.

A Hierarchical Analysis On Ecosystem Classification of Land

Background: The ecosystem classification of land (ECL) has been studied for a couple of decades, from the beginning of the perfect organism system “top-down” approach to a reversed “bottom-up” approach, defining micro-ecological unit. After reviewing two study cases of the ecosystem classification of land, the ecosystem classification framework implements in different ecoregions were examined and analyzed. Results: Theoretically, Bailey’s upper levels ECL (1995) was applied to the United States, and world continents. China's Eco-geographic classification was most likely fitted into Bailey’s Ecosystem Classification regime. With a binary decision tree analysis, it demonstrated that the top-level, Domain has an empty entity between the US and China ecoregion framework. Based on the biogeoclimate condition, vegetation distribution, landform, and plant species feature, classified HIIC1 into two subsections ( labeled as i, ii ), and delineated iia of QiLian Mountain East Alpine Shrub and Alpine Tundra ecozone into iia-1 and iia-2 zone.Conclusions: 1) The Plateau Domain 500 should be added into the top-level Bailey’s ecoregion framework, coordinately it includes HI and HII Divisions, and humid, dub-humid, semiarid, and arid provinces. 2) Two case comparisons recommend using a practical approach, objectively defined ecosystem classification for the lower-level ECLs in matter of time and project cost.

MiR-30a and miR-200c differentiate cholangiocarcinomas from gastrointestinal cancer liver metastases

Prior studies have demonstrated the utility of microRNA assays for predicting some cancer tissue origins, but these assays need to be further optimized for predicting the tissue origins of adenocarcinomas of the liver. We performed microRNA profiling on 195 frozen primary tumor samples using 14 types of tumors that were either adenocarcinomas or differentiated from adenocarcinomas. The 1-nearest neighbor method predicted tissue-of-origin in 33 samples of a test set, with an accuracy of 93.9% at feature selection p values ranging from 10−4 to 10−10. According to binary decision tree analyses, the overexpression of miR-30a and the underexpression of miR-200 family members (miR-200c and miR-141) differentiated intrahepatic cholangiocarcinomas from extrahepatic adenocarcinomas. When binary decision tree analyses were performed using the test set, the prediction accuracy was 84.8%. The overexpression of miR-30a and the reduced expressions of miR-200c, miR-141, and miR-425 could distinguish intrahepatic cholangiocarcinomas from liver metastases from the gastrointestinal tract.

Research on Multiple Classification Based on Improved SVM Algorithm for Balanced Binary Decision Tree

Support vector machines (SVMs) are designed to solve the binary classification problems at the beginning, but in the real world, there are a lot of multiclassification cases. The multiclassification methods based on SVM are mainly divided into the direct methods and the indirect methods, in which the indirect methods, which consist of multiple binary classifiers integrated in accordance with certain rules to form the multiclassification model, are the most commonly used multiclassification methods at present. In this paper, an improved multiclassification algorithm based on the balanced binary decision tree is proposed, which is called the IBDT-SVM algorithm. In this algorithm, it considers not only the influence of “between-classes distance” and “class variance” in traditional measures of between-classes separability but also takes “between-classes variance” into consideration and proposes a new improved “between-classes separability measure.” Based on the new “between-classes separability measure,” it finds out the two classes with the largest between-classes separability measure and uses them as the positive and negative samples to train and learn the classifier. After that, according to the principle of the class-grouping-by-majority, the remaining classes are close to these two classes and merged into the positive samples and the negative samples to train SVM classifier again. For the samples with uneven distribution or sparse distribution, this method can avoid the error caused by the shortest canter distance classification method and overcome the “error accumulation” problem existing in traditional binary decision tree to the greatest extent so as to obtain a better classifier. According to the above algorithm, each layer node of the decision tree is traversed until the output classification result is a single-class label. The experimental results show that the IBDT-SVM algorithm proposed in this paper can achieve better classification accuracy and effectiveness for multiple classification problems.

Comparative study of Image Classification Algorithms

This investigation analyzed five common machine learning techniques for performing image classification included Support Vector Machines (SVM), K-Nearest Neighbor (KNN), Naïve Bayes (NB), Binary Decision Tree (BDT) and Discriminant Analysis (DA). AlexNet deep learning model was utilized to fabricate these machine learning classifiers. The structure classifiers were executed and assessed by standard execution models of Accuracy (ACC), Precision (P), Sensitivity (S), Specificity (Spe) and Area Under the ROC Curve (AUC). The five strategies were assessed utilizing 2608 histopathological pictures for head and neck cancer. The examination was directed utilizing multiple times 10-overlay cross validation. For every strategy, the pre-trained AlexNet network was utilized to separate highlights from the activation layer. The outcomes outlined that, there was no contrast between the consequences of SVM and KNN. Both have the equivalent and the higher accuracy than others were 99.98 %, though 99.81%, 97.32% and 93.68% for DA, BDT and NB, separately. The current examination shows that the SVM, KNN and DA are the best techniques for classifying our dataset images.

Fast HEVC Inter-Prediction Algorithm Based on Matching Block Features

A fast inter-prediction algorithm based on matching block features is proposed in this article. The position of the matching block of the current CU in the previous frame is found by the motion vector estimated by the corresponding located CU in the previous frame. Then, a weighted motion vector computation method is presented to compute the motion vector of the matching block of the current CU according to the motions of the PUs the matching block covers. A binary decision tree is built to decide the CU depths and PU mode for the current CU. Four training features are drawn from the characteristics of the CUs and PUs the matching block covers. Simulation results show that the proposed algorithm achieves average 1.1% BD-rate saving, 14.5% coding time saving and 0.01-0.03 dB improvement in peak signal-to-noise ratio (PSNR), compared to the present fast inter-prediction algorithm in HEVC.