Comparison of decision tree, support vector machines, and Bayesian network approaches for classification of falls in Parkinson’s disease

Being able to accurately predict the risk of falling is crucial in patients with Parkinson’s dis- ease (PD). This is due to the unfavorable effect of falls, which can lower the quality of life as well as directly impact on survival. Three methods considered for predicting falls are decision trees (DT), Bayesian networks (BN), and support vector machines (SVM). Data on a 1-year prospective study conducted at IHBI, Australia, for 51 people with PD are used. Data processing are conducted using rpart and e1071 packages in R for DT and SVM, con- secutively; and Bayes Server 5.5 for the BN. The results show that BN and SVM produce consistently higher accuracy over the 12 months evaluation time points (average sensitivity and specificity > 92%) than DT (average sensitivity 88%, average specificity 72%). DT is prone to imbalanced data so needs to adjust for the misclassification cost. However, DT provides a straightforward, interpretable result and thus is appealing for helping to identify important items related to falls and to generate fallers’ profiles.

Automatic ICD-10 classification of cancers from free-text death certificates

Objective Death certificates provide an invaluable source for cancer mortality statistics; however, this value can only be realised if accurate, quantitative data can be extracted from certificates – an aim hampered by both the volume and variable nature of certificates written in natural language. This paper proposes an automatic classification system for identifying cancer related causes of death from death certificates. Methods Detailed features, including terms, n-grams and SNOMED CT concepts were extracted from a collection of 447,336 death certificates. These features were used to train Support Vector Machine classifiers (one classifier for each cancer type). The classifiers were deployed in a cascaded architecture: the first level identified the presence of cancer (i.e., binary cancer/nocancer) and the second level identified the type of cancer (according to the ICD-10 classification system). A held-out test set was used to evaluate the effectiveness of the classifiers according to precision, recall and F-measure. In addition, detailed feature analysis was performed to reveal the characteristics of a successful cancer classification model. Results The system was highly effective at identifying cancer as the underlying cause of death (F-measure 0.94). The system was also effective at determining the type of cancer for common cancers (F-measure 0.7). Rare cancers, for which there was little training data, were difficult to classify accurately (F-measure 0.12). Factors influencing performance were the amount of training data and certain ambiguous cancers (e.g., those in the stomach region). The feature analysis revealed a combination of features were important for cancer type classification, with SNOMED CT concept and oncology specific morphology features proving the most valuable. Conclusion The system proposed in this study provides automatic identification and characterisation of cancers from large collections of free-text death certificates. This allows organisations such as Cancer Registries to monitor and report on cancer mortality in a timely and accurate manner. In addition, the methods and findings are generally applicable beyond cancer classification and to other sources of medical text besides death certificates.

Maize kernel hardness classification by near infrared (NIR) hyperspectral imaging and multivariate data analysis.

The use of near infrared (NIR) hyperspectral imaging and hyperspectral image analysis for distinguishing between hard, intermediate and soft maize kernels from inbred lines was evaluated. NIR hyperspectral images of two sets (12 and 24 kernels) of whole maize kernels were acquired using a Spectral Dimensions MatrixNIR camera with a spectral range of 960-1662 nm and a sisuChema SWIR (short wave infrared) hyperspectral pushbroom imaging system with a spectral range of 1000-2498 nm. Exploratory principal component analysis (PCA) was used on absorbance images to remove background, bad pixels and shading. On the cleaned images. PCA could be used effectively to find histological classes including glassy (hard) and floury (soft) endosperm. PCA illustrated a distinct difference between glassy and floury endosperm along principal component (PC) three on the MatrixNIR and PC two on the sisuChema with two distinguishable clusters. Subsequently partial least squares discriminant analysis (PLS-DA) was applied to build a classification model. The PLS-DA model from the MatrixNIR image (12 kernels) resulted in root mean square error of prediction (RMSEP) value of 0.18. This was repeated on the MatrixNIR image of the 24 kernels which resulted in RMSEP of 0.18. The sisuChema image yielded RMSEP value of 0.29. The reproducible results obtained with the different data sets indicate that the method proposed in this paper has a real potential for future classification uses.

Porn, censorship, classification and free speech: Global paradoxes in the governance of media content

In this presentation, I reflect upon the global landscape surrounding the governance and classification of media content, at a time of rapid change in media platforms and services for content production and distribution, and contested cultural and social norms. I discuss the tensions and contradictions arising in the relationship between national, regional and global dimensions of media content distribution, as well as the changing relationships between state and non-state actors. These issues will be explored through consideration of issues such as: recent debates over film censorship; the review of the National Classification Scheme conducted by the Australian Law Reform Commission; online controversies such as the future of the Reddit social media site; and videos posted online by the militant group ISIS.

Direct measurement of inner prosthesis loading during activities of daily living: Proposal for classification of functional outcome

Background The purpose of this presentation is to outline the relevance of the categorization of the load regime data to assess the functional output and usage of the prosthesis of lower limb amputees. The objectives are • To highlight the need for categorisation of activities of daily living • To present a categorization of load regime applied on residuum, • To present some descriptors of the four types of activity that could be detected, • To provide an example the results for a case. Methods The load applied on the osseointegrated fixation of one transfemoral amputee was recorded using a portable kinetic system for 5 hours. The load applied on the residuum was divided in four types of activities corresponding to inactivity, stationary loading, localized locomotion and directional locomotion as detailed in previously publications. Results The periods of directional locomotion, localized locomotion, and stationary loading occurred 44%, 34%, and 22% of recording time and each accounted for 51%, 38%, and 12% of the duration of the periods of activity, respectively. The absolute maximum force during directional locomotion, localized locomotion, and stationary loading was 19%, 15%, and 8% of the body weight on the anteroposterior axis, 20%, 19%, and 12% on the mediolateral axis, and 121%, 106%, and 99% on the long axis. A total of 2,783 gait cycles were recorded. Discussion Approximately 10% more gait cycles and 50% more of the total impulse than conventional analyses were identified. The proposed categorization and apparatus have the potential to complement conventional instruments, particularly for difficult cases.

IUCN classification zones concord with, but underestimate, the population genetic structure of the zebra shark Stegostoma fasciatum in the Indo-West Pacific

The Indo-West Pacific (IWP), from South Africa in the western Indian Ocean to the western Pacific Ocean, contains some of the most biologically diverse marine habitats on earth, including the greatest biodiversity of chondrichthyan fishes. The region encompasses various densities of human habitation leading to contrasts in the levels of exploitation experienced by chondrichthyans, which are targeted for local consumption and export. The demersal chondrichthyan, the zebra shark, Stegostoma fasciatum, is endemic to the IWP and has two current regional International Union for the Conservation of Nature (IUCN) Red List classifications that reflect differing levels of exploitation: ‘Least Concern’ and ‘Vulnerable’. In this study, we employed mitochondrial ND4 sequence data and 13 microsatellite loci to investigate the population genetic structure of 180 zebra sharks from 13 locations throughout the IWP to test the concordance of IUCN zones with demographic units that have conservation value. Mitochondrial and microsatellite data sets from samples collected throughout northern Australia and Southeast Asia concord with the regional IUCN classifications. However, we found evidence of genetic subdivision within these regions, including subdivision between locations connected by habitat suitable for migration. Furthermore, parametric FST analyses and Bayesian clustering analyses indicated that the primary genetic break within the IWP is not represented by the IUCN classifications but rather is congruent with the Indonesian throughflow current. Our findings indicate that recruitment to areas of high exploitation from nearby healthy populations in zebra sharks is likely to be minimal, and that severe localized depletions are predicted to occur in zebra shark populations throughout the IWP region.

Isohydric and anisohydric characterisation of vegetable crops. The classification of vegetables by their physiological responses to water stress

Research on the physiological response of crop plants to drying soils and subsequent water stress has grouped plant behaviours as isohydric and anisohydric. Drying soil conditions, and hence declining soil and root water potentials, cause chemical signals—the most studied being abscisic acid (ABA)—and hydraulic signals to be transmitted to the leaf via xylem pathways. Researchers have attempted to allocate crops as isohydric or anisohydric. However, different cultivars within crops, and even the same cultivars grown in different environments/climates, can exhibit both response types. Nevertheless, understanding which behaviours predominate in which crops and circumstances may be beneficial. This paper describes different physiological water stress responses, attempts to classify vegetable crops according to reported water stress responses, and also discusses implications for irrigation decision-making.

Molecular classification of familial colorectal and endometrial carcinoma

Hereditary nonpolyposis colorectal cancer (HNPCC) is the most common known clearly hereditary cause of colorectal and endometrial cancer (CRC and EC). Dominantly inherited mutations in one of the known mismatch repair (MMR) genes predispose to HNPCC. Defective MMR leads to an accumulation of mutations especially in repeat tracts, presenting microsatellite instability. HNPCC is clinically a very heterogeneous disease. The age at onset varies and the target tissue may vary. In addition, families that fulfill the diagnostic criteria for HNPCC but fail to show any predisposing mutation in MMR genes exist. Our aim was to evaluate the genetic background of familial CRC and EC. We performed comprehensive molecular and DNA copy number analyses of CRCs fulfilling the diagnostic criteria for HNPCC. We studied the role of five pathways (MMR, Wnt, p53, CIN, PI3K/AKT) and divided the tumors into two groups, one with MMR gene germline mutations and the other without. We observed that MMR proficient familial CRC consist of two molecularly distinct groups that differ from MMR deficient tumors. Group A shows paucity of common molecular and chromosomal alterations characteristic of colorectal carcinogenesis. Group B shows molecular features similar to classical microsatellite stable tumors with gross chromosomal alterations. Our finding of a unique tumor profile in group A suggests the involvement of novel predisposing genes and pathways in colorectal cancer cohorts not linked to MMR gene defects. We investigated the genetic background of familial ECs. Among 22 families with clustering of EC, two (9%) were due to MMR gene germline mutations. The remaining familial site-specific ECs are largely comparable with HNPCC associated ECs, the main difference between these groups being MMR proficiency vs. deficiency. We studied the role of PI3K/AKT pathway in familial ECs as well and observed that PIK3CA amplifications are characteristic of familial site-specific EC without MMR gene germline mutations. Most of the high-level amplifications occurred in tumors with stable microsatellites, suggesting that these tumors are more likely associated with chromosomal rather than microsatellite instability and MMR defect. The existence of site-specific endometrial carcinoma as a separate entity remains equivocal until predisposing genes are identified. It is possible that no single highly penetrant gene for this proposed syndrome exists, it may, for example be due to a combination of multiple low penetrance genes. Despite advances in deciphering the molecular genetic background of HNPCC, it is poorly understood why certain organs are more susceptible than others to cancer development. We found that important determinants of the HNPCC tumor spectrum are, in addition to different predisposing germline mutations, organ specific target genes and different instability profiles, loss of heterozygosity at MLH1 locus, and MLH1 promoter methylation. This study provided more precise molecular classification of families with CRC and EC. Our observations on familial CRC and EC are likely to have broader significance that extends to sporadic CRC and EC as well.

Interval Data Classification under Partial Information: A Chance-Constraint Approach

This paper presents a Chance-constraint Programming approach for constructing maximum-margin classifiers which are robust to interval-valued uncertainty in training examples. The methodology ensures that uncertain examples are classified correctly with high probability by employing chance-constraints. The main contribution of the paper is to pose the resultant optimization problem as a Second Order Cone Program by using large deviation inequalities, due to Bernstein. Apart from support and mean of the uncertain examples these Bernstein based relaxations make no further assumptions on the underlying uncertainty. Classifiers built using the proposed approach are less conservative, yield higher margins and hence are expected to generalize better than existing methods. Experimental results on synthetic and real-world datasets show that the proposed classifiers are better equipped to handle interval-valued uncertainty than state-of-the-art.

Bovine cysticercosis-Development of a real-time PCR to enhance classification of suspect cysts identified at meat inspection

Laboratory confirmation methods are important in bovine cysticerosis diagnosis as other pathologies can result in morphologically similar lesions resulting in false identifications. We developed a probe-based real-time PCR assay to identify Taenia saginata in suspect cysts encountered at meat inspection and compared its use with the traditional method of identification, histology, as well as a published nested PCR. The assay simultaneously detects T. saginata DNA and a bovine internal control using the cytochrome c oxidase subunit 1 gene of each species and shows specificity against parasites causing lesions morphologically similar to those of T. saginata. The assay was sufficiently sensitive to detect 1 fg (Ct 35.09 +/- 0.95) of target DNA using serially-diluted plasmid DNA in reactions spiked with bovine DNA as well as in all viable and caseated positive control cysts. A loss in PCR sensitivity was observed with increasing cyst degeneration as seen in other molecular methods. In comparison to histology, the assay offered greater sensitivity and accuracy with 10/19 (53%) T. saginata positives detected by real-time PCR and none by histology. When the results were compared with the reference PCR, the assay was less sensitive but offered advantages of faster turnaround times and reduced contamination risk. Estimates of the assay's repeatability and reproducibility showed the assay is highly reliable with reliability coefficients greater than 0.94. Crown Copyright (C) 2013 Published by Elsevier B.V. All rights reserved.

Adaptive frequency scaled wavelet packet decomposition for frog call classification

Environmental changes have put great pressure on biological systems leading to the rapid decline of biodiversity. To monitor this change and protect biodiversity, animal vocalizations have been widely explored by the aid of deploying acoustic sensors in the field. Consequently, large volumes of acoustic data are collected. However, traditional manual methods that require ecologists to physically visit sites to collect biodiversity data are both costly and time consuming. Therefore it is essential to develop new semi-automated and automated methods to identify species in automated audio recordings. In this study, a novel feature extraction method based on wavelet packet decomposition is proposed for frog call classification. After syllable segmentation, the advertisement call of each frog syllable is represented by a spectral peak track, from which track duration, dominant frequency and oscillation rate are calculated. Then, a k-means clustering algorithm is applied to the dominant frequency, and the centroids of clustering results are used to generate the frequency scale for wavelet packet decomposition (WPD). Next, a new feature set named adaptive frequency scaled wavelet packet decomposition sub-band cepstral coefficients is extracted by performing WPD on the windowed frog calls. Furthermore, the statistics of all feature vectors over each windowed signal are calculated for producing the final feature set. Finally, two well-known classifiers, a k-nearest neighbour classifier and a support vector machine classifier, are used for classification. In our experiments, we use two different datasets from Queensland, Australia (18 frog species from commercial recordings and field recordings of 8 frog species from James Cook University recordings). The weighted classification accuracy with our proposed method is 99.5% and 97.4% for 18 frog species and 8 frog species respectively, which outperforms all other comparable methods.

Host Rock Classification (HRC) system for nuclear waste disposal in crystalline bedrock

A new rock mass classification scheme, the Host Rock Classification system (HRC-system) has been developed for evaluating the suitability of volumes of rock mass for the disposal of high-level nuclear waste in Precambrian crystalline bedrock. To support the development of the system, the requirements of host rock to be used for disposal have been studied in detail and the significance of the various rock mass properties have been examined. The HRC-system considers both the long-term safety of the repository and the constructability in the rock mass. The system is specific to the KBS-3V disposal concept and can be used only at sites that have been evaluated to be suitable at the site scale. By using the HRC-system, it is possible to identify potentially suitable volumes within the site at several different scales (repository, tunnel and canister scales). The selection of the classification parameters to be included in the HRC-system is based on an extensive study on the rock mass properties and their various influences on the long-term safety, the constructability and the layout and location of the repository. The parameters proposed for the classification at the repository scale include fracture zones, strength/stress ratio, hydraulic conductivity and the Groundwater Chemistry Index. The parameters proposed for the classification at the tunnel scale include hydraulic conductivity, Q´ and fracture zones and the parameters proposed for the classification at the canister scale include hydraulic conductivity, Q´, fracture zones, fracture width (aperture + filling) and fracture trace length. The parameter values will be used to determine the suitability classes for the volumes of rock to be classified. The HRC-system includes four suitability classes at the repository and tunnel scales and three suitability classes at the canister scale and the classification process is linked to several important decisions regarding the location and acceptability of many components of the repository at all three scales. The HRC-system is, thereby, one possible design tool that aids in locating the different repository components into volumes of host rock that are more suitable than others and that are considered to fulfil the fundamental requirements set for the repository host rock. The generic HRC-system, which is the main result of this work, is also adjusted to the site-specific properties of the Olkiluoto site in Finland and the classification procedure is demonstrated by a test classification using data from Olkiluoto. Keywords: host rock, classification, HRC-system, nuclear waste disposal, long-term safety, constructability, KBS-3V, crystalline bedrock, Olkiluoto

Modeling Efficient Classification as a Process of Confidence Assessment and Delegation

In visual object detection and recognition, classifiers have two interesting characteristics: accuracy and speed. Accuracy depends on the complexity of the image features and classifier decision surfaces. Speed depends on the hardware and the computational effort required to use the features and decision surfaces. When attempts to increase accuracy lead to increases in complexity and effort, it is necessary to ask how much are we willing to pay for increased accuracy. For example, if increased computational effort implies quickly diminishing returns in accuracy, then those designing inexpensive surveillance applications cannot aim for maximum accuracy at any cost. It becomes necessary to find trade-offs between accuracy and effort. We study efficient classification of images depicting real-world objects and scenes. Classification is efficient when a classifier can be controlled so that the desired trade-off between accuracy and effort (speed) is achieved and unnecessary computations are avoided on a per input basis. A framework is proposed for understanding and modeling efficient classification of images. Classification is modeled as a tree-like process. In designing the framework, it is important to recognize what is essential and to avoid structures that are narrow in applicability. Earlier frameworks are lacking in this regard. The overall contribution is two-fold. First, the framework is presented, subjected to experiments, and shown to be satisfactory. Second, certain unconventional approaches are experimented with. This allows the separation of the essential from the conventional. To determine if the framework is satisfactory, three categories of questions are identified: trade-off optimization, classifier tree organization, and rules for delegation and confidence modeling. Questions and problems related to each category are addressed and empirical results are presented. For example, related to trade-off optimization, we address the problem of computational bottlenecks that limit the range of trade-offs. We also ask if accuracy versus effort trade-offs can be controlled after training. For another example, regarding classifier tree organization, we first consider the task of organizing a tree in a problem-specific manner. We then ask if problem-specific organization is necessary.

Combat sports analytics: Boxing punch classification using overhead depth imagery

In competitive combat sporting environments like boxing, the statistics on a boxer's performance, including the amount and type of punches thrown, provide a valuable source of data and feedback which is routinely used for coaching and performance improvement purposes. This paper presents a robust framework for the automatic classification of a boxer's punches. Overhead depth imagery is employed to alleviate challenges associated with occlusions, and robust body-part tracking is developed for the noisy time-of-flight sensors. Punch recognition is addressed through both a multi-class SVM and Random Forest classifiers. A coarse-to-fine hierarchical SVM classifier is presented based on prior knowledge of boxing punches. This framework has been applied to shadow boxing image sequences taken at the Australian Institute of Sport with 8 elite boxers. Results demonstrate the effectiveness of the proposed approach, with the hierarchical SVM classifier yielding a 96% accuracy, signifying its suitability for analysing athletes punches in boxing bouts.