Identification of New Zealand bats in flight from analysis of echolocation calls by artificial neural networks.

Time-expanded and heterodyned echolocation calls of the New Zealand long-tailed Chalinolobus tuberculatus and lesser short-tailed bat Mystacina tuberculata were recorded and digitally analysed. Temporal and spectral parameters were measured from time-expanded calls and power spectra generated for both time-expanded and heterodyned calls. Artificial neural networks were trained to classify the calls of both species using temporal and spectral parameters and power spectra as input data. Networks were then tested using data not previously seen. Calls could be unambiguously identified using parameters and power spectra from time-expanded calls. A neural network, trained and tested using power spectra of calls from both species recorded using a heterodyne detector set to 40 kHz (the frequency with the most energy of the fundamental of C. tuberculatus call), could identify 99% and 84% of calls of C. tuberculatus and M. tuberculata, respectively. A second network, trained and tested using power spectra of calls from both species recorded using a heterodyne detector set to 27 kHz (the frequency with the most energy of the fundamental of M. tuberculata call), could identify 34% and 100% of calls of C. tuberculatus and M. tuberculata, respectively. This study represents the first use of neural networks for the identification of bats from their echolocation calls. It is also the first study to use power spectra of time-expanded and heterodyned calls for identification of chiropteran species. The ability of neural networks to identify bats from their echolocation calls is discussed, as is the ecology of both species in relation to the design of their echolocation calls.

Acoustic identification of 12 species of echolocating bat by discriminant function analysis and artificial neural networks

We recorded echolocation calls from 14 sympatric species of bat in Britain. Once digitised, one temporal and four spectral features were measured from each call. The frequency-time course of each call was approximated by fitting eight mathematical functions, and the goodness of fit, represented by the mean-squared error, was calculated. Measurements were taken using an automated process that extracted a single call from background noise and measured all variables without intervention. Two species of Rhinolophus were easily identified from call duration and spectral measurements. For the remaining 12 species, discriminant function analysis and multilayer back-propagation perceptrons were used to classify calls to species level. Analyses were carried out with and without the inclusion of curve-fitting data to evaluate its usefulness in distinguishing among species. Discriminant function analysis achieved an overall correct classification rate of 79% with curve-fitting data included, while an artificial neural network achieved 87%. The removal of curve-fitting data improved the performance of the discriminant function analysis by 2 %, while the performance of a perceptron decreased by 2 %. However, an increase in correct identification rates when curve-fitting information was included was not found for all species. The use of a hierarchical classification system, whereby calls were first classified to genus level and then to species level, had little effect on correct classification rates by discriminant function analysis but did improve rates achieved by perceptrons. This is the first published study to use artificial neural networks to classify the echolocation calls of bats to species level. Our findings are discussed in terms of recent advances in recording and analysis technologies, and are related to factors causing convergence and divergence of echolocation call design in bats.

Paternal postnatal depressive symptoms, infant sleeping and feeding behaviors, and rigid parental regulation : a correlational study

Paternal postnatal depression (PND) is now recognized as a serious and prevalent problem, associated with poorer well-being and functioning of all family members. Aspects of infant temperament, sleeping and feeding perceived by parents as problematic are associated with maternal PND, however, less is known about paternal PND. This study investigated depressive symptoms (Edinburgh postnatal depression scale (EPDS)) in 219 fathers of infants aged from 1 to 24 weeks (median 7.0 weeks). Infant predictor variables were sleeping problems, feeding problems and both mother and father reported temperament. Control variables were partner’s support, other support and life events. Rigidity of parenting beliefs regarding infant regulation was also measured as a potential moderating factor. Infant feeding difficulties were associated with paternal depressive symptoms, subsuming the variance associated with both sleep problems and temperament. This relationship was not moderated by regulation beliefs. It was concluded that infant feeding is important to fathers. Fathers of infants with feeding difficulties may not be able to fulfill their idealized construction of involved fatherhood. Role incongruence may have an etiological role in paternal PND.

Parent feeding interactions and practices during childhood cancer treatment: A qualitative investigation

In the general population it is evident that parent feeding practices can directly shape a child’s life long dietary intake. Young children undergoing childhood cancer treatment may experience feeding difficulties and limited food intake, due to the inherent side effects of their anti-cancer treatment. What is not clear is how these treatment side effects are influencing the parent–child feeding relationship during anti-cancer treatment. This retrospective qualitative study collected telephone based interview data from 38 parents of childhood cancer patients who had recently completed cancer treatment (child’s mean age: 6.98 years). Parents described a range of treatment side effects that impacted on their child’s ability to eat, often resulting in weight loss. Sixty-one percent of parents (n = 23) reported high levels of stress in regard to their child’s eating and weight loss during treatment. Parents reported stress, feelings of helplessness, and conflict and/or tension between parent and the child during feeding/eating interactions. Parents described using both positive and negative feeding practices, such as: pressuring their child to eat, threatening the insertion of a nasogastric feeding tube, encouraging the child to eat and providing home cooked meals in hospital. Results indicated that parent stress may lead to the use of coping strategies such as positive or negative feeding practices to entice their child to eat during cancer treatment. Future research is recommended to determine the implication of parent feeding practice on the long term diet quality and food preferences of childhood cancer survivors.

Acoustic identification of twelve species of echolocating bat by discriminant function analysis and artificial neural networks

We recorded echolocation calls from 14 sympatric species of bat in Britain. Once digitised, one temporal and four spectral features were measured from each call. The frequency-time course of each call was approximated by fitting eight mathematical functions, and the goodness of fit, represented by the mean-squared error, was calculated. Measurements were taken using an automated process that extracted a single call from background noise and measured all variables without intervention. Two species of Rhinolophus were easily identified from call duration and spectral measurements. For the remaining 12 species, discriminant function analysis and multilayer back-propagation perceptrons were used to classify calls to species level. Analyses were carried out with and without the inclusion of curve-fitting data to evaluate its usefulness in distinguishing among species. Discriminant function analysis achieved an overall correct classification rate of 79% with curve-fitting data included, while an artificial neural network achieved 87%. The removal of curve-fitting data improved the performance of the discriminant function analysis by 2 %, while the performance of a perceptron decreased by 2 %. However, an increase in correct identification rates when curve-fitting information was included was not found for all species. The use of a hierarchical classification system, whereby calls were first classified to genus level and then to species level, had little effect on correct classification rates by discriminant function analysis but did improve rates achieved by perceptrons. This is the first published study to use artificial neural networks to classify the echolocation calls of bats to species level. Our findings are discussed in terms of recent advances in recording and analysis technologies, and are related to factors causing convergence and divergence of echolocation call design in bats.

Relationships between maternal overweight prior to pregnancy, feeding mode and infant feeding beliefs and practices

Aim To examine whether pre-pregnancy weight status was associated with maternal feeding beliefs and practices in the early post-partum period. Methods Secondary analysis of longitudinal data from Australian mothers. Participants (N=486) were divided into two weight status groups based on self-reported pre-pregnancy weight and measured height: healthy weight (BMI <25kg/m2; n=321) and overweight (BMI>25kg/m2; n=165). Feeding beliefs and practices were self-reported via an established questionnaire that assessed concerns about infant overeating and undereating, awareness of infant cues, feeding to a schedule, and using food to calm. Results Infants of overweight mothers were more likely to have been given solid foods in the previous 24hrs (29% vs 20%) and fewer were fully breastfed (50% vs 64%). Multivariable regression analyses (adjusted for maternal education, parity, average infant weekly weight gain, feeding mode and introduction of solids) revealed pre-pregnancy weight status was not associated with using food to calm, concern about undereating, awareness of infant cues or feeding to a schedule. However feeding mode was associated with feeding beliefs and practices. Conclusions Although no evidence for a relationship between maternal weight status and early maternal feeding beliefs and practices was observed, differences in feeding mode and early introduction of solids was observed. The emergence of a relationship between feeding practices and maternal weight status may occur when the children are older, solid feeding is established and they become more independent in feeding.

The relationship between controlling feeding practices and boys’ and girls’ eating in the absence of hunger

Parental controlling feeding practices have been directly associated with maladaptive child eating behaviours, such as Eating in the Absence of Hunger (EAH). The aim of this study was to examine EAH in very young children (3-4 years old) and to investigate the association between maternal controlling feeding practices and energy intake from a standardised selection of snacks consumed ‘in the absence of hunger’. Thirty-seven mother-child dyads enrolled in the NOURISH RCT participated in a modified EAH protocol conducted in the child’s home. All children displayed EAH, despite 80% reporting to be full or very full following completion of lunch 15 minutes earlier. The relationship between maternal and child covariates and controlling feeding practices and EAH were examined using non-parametric tests, and were stratified by child gender. For boys only, pressure to eat was positively associated with EAH. Neither restriction nor monitoring practices were associated with EAH in either boys or girls. Overall, the present findings suggest gender differences in the relationship between maternal feeding practices and children’s eating behaviours emerge early and should be considered in future research and intervention design.

Prediction of GMA welding characteristic parameter by artificial neural network system

Nowadays, demand for automated Gas metal arc welding (GMAW) is growing and consequently need for intelligent systems is increased to ensure the accuracy of the procedure. To date, welding pool geometry has been the most used factor in quality assessment of intelligent welding systems. But, it has recently been found that Mahalanobis Distance (MD) not only can be used for this purpose but also is more efficient. In the present paper, Artificial Neural Networks (ANN) has been used for prediction of MD parameter. However, advantages and disadvantages of other methods have been discussed. The Levenberg–Marquardt algorithm was found to be the most effective algorithm for GMAW process. It is known that the number of neurons plays an important role in optimal network design. In this work, using trial and error method, it has been found that 30 is the optimal number of neurons. The model has been investigated with different number of layers in Multilayer Perceptron (MLP) architecture and has been shown that for the aim of this work the optimal result is obtained when using MLP with one layer. Robustness of the system has been evaluated by adding noise into the input data and studying the effect of the noise in prediction capability of the network. The experiments for this study were conducted in an automated GMAW setup that was integrated with data acquisition system and prepared in a laboratory for welding of steel plate with 12 mm in thickness. The accuracy of the network was evaluated by Root Mean Squared (RMS) error between the measured and the estimated values. The low error value (about 0.008) reflects the good accuracy of the model. Also the comparison of the predicted results by ANN and the test data set showed very good agreement that reveals the predictive power of the model. Therefore, the ANN model offered in here for GMA welding process can be used effectively for prediction goals.

An early feeding practices intervention for obesity prevention

Objective - Report long term outcomes of the NOURISH randomized controlled trial (RCT) that evaluated a universal intervention commencing in infancy to provide anticipatory guidance to first-time mothers on ‘protective’ complementary feeding practices which were hypothesized to reduce childhood obesity risk. Subjects and Methods - The NOURISH RCT enrolled 698 mothers (mean age 30.1 years, SD=5.3) with healthy term infants (51% female). Mothers were randomly allocated to usual care or to attend two 6-session, 12-week group education modules. Outcomes were assessed five times: baseline (infants 4.3 months); 6 months after module 1 (infants 14 months); 6 months after module 2 (infants 2 years) and at 3.5 and 5 years of age. Maternal feeding practices were self-reported using validated questionnaires. BMI Z-score was calculated from measured child height and weight. Linear Mixed Models evaluated intervention (group) effect across time. Results - Retention at 5 years of age was 61%. Across ages 2-5 years, intervention mothers reported less frequent use of non-responsive feeding practices on 6/9 scales. At 5 years they also reported more appropriate responses to food refusal on 7/12 items (Ps ≤.05). No statistically significant group effect was noted for anthropometric outcomes (BMI Z-score: P=.06), or the prevalence of overweight/obesity (control 13.3% vs. intervention 11.4%, P=.66). Conclusions - Anticipatory guidance on complementary feeding resulted in first-time mothers reporting increased use of protective feeding practices. These intervention effects were sustained up to five years of age and were paralleled by a non-significant trend for lower child BMI Z-scores at all post-intervention assessment points.

Autonomous helicopter hover using an artificial neural network

Details the developments to date of an unmanned air vehicle (UAV) based on a standard size 60 model helicopter. The design goal is to have the helicopter achieve stable hover with the aid of an INS and stereo vision. The focus of the paper is on the development of an artificial neural network (ANN) that makes use of only the INS data to generate hover commands, which are used to directly manipulate the flight servos. Current results show that networks incorporating some form of recurrency (state history) offer little advantage over those without. At this stage, the ANN has partially maintained periods of hover even with misaligned sensors.

Extending the validity of the Feeding Practices and Structure Questionnaire

Background Feeding practices are commonly examined as potentially modifiable determinants of children’s eating behaviours and weight status. Although a variety of questionnaires exist to assess different feeding aspects, many lack thorough reliability and validity testing. The Feeding Practices and Structure Questionnaire (FPSQ) is a tool designed to measure early feeding practices related to non-responsive feeding and structure of the meal environment. Face validity, factorial validity, internal reliability and cross-sectional correlations with children’s eating behaviours have been established in mothers with 2-year-old children. The aim of the present study was to further extend the validity of the FPSQ by examining factorial, construct and predictive validity, and stability. Methods Participants were from the NOURISH randomised controlled trial which evaluated an intervention with first-time mothers designed to promote protective feeding practices. Maternal feeding practices (FP) and child eating behaviours were assessed when children were aged 2 years and 3.7 years (n=388). Confirmatory Factor analysis, group differences, predictive relationships, and stability were tested. Results The original 9-factor structure was confirmed when children were aged 3.7±0.3 years. Cronbach’s alpha was above the recommended 0.70 cut-off for all factors except Structured Meal Timing, Over Restriction and Distrust in Appetite which were 0.58, 0.67 and 0.66 respectively. Allocated group differences reflected behaviour consistent with intervention content and all feeding practices were stable across both time points (range of r= 0.45-0.70). There was some evidence for the predictive validity of factors with 2 FP showing expected relationships, 2 FP showing expected and unexpected relationships and 5 FP showing no relationship. Conclusions Reliability and validity was demonstrated for most subscales of the FPSQ. Future validation is warranted with culturally diverse samples and with fathers and other caregivers. The use of additional outcomes to further explore predictive validity is recommended as well as testing construct validity and test-retest reliability of the questionnaire.

Evaluation of protein bait laced with various insecticides on the Queensland fruit fly (Diptera: Tephritidae): Attraction, feeding, mortality and bait persistence

The use of malathion in fruit fly protein bait sprays has raised serious concerns due to its adverse effects on non-target organisms. This has necessitated the evaluation of novel reduced-risk compounds. This study evaluated the effects of spinosad, fipronil, malathion and chlorpyrifos mixed with fruit fly protein bait (Mauri Pinnacle protein®) on attraction, feeding and mortality of the Queensland fruit fly, Bactrocera tryoni (Froggatt). The effects of outdoor weathering of these mixtures on fly mortality were also determined. In field-cage experiment, protein-starved flies showed the same level of attraction to baits containing spinosad, fipronil, malathion, chlorpyrifos and protein alone used as control. Female protein-starved flies were deterred from feeding on baits containing malathion and chlorpyrifos compared to baits containing spinosad, fipronil and protein alone. Baits containing malathion and chlorpyrifos caused higher fly mortality and rapid fly knock down than spinosad and fipronil. However, spinosad acted slowly and caused an increase in fly mortality over time, causing up to 90% fly mortality after 72-h. Baits containing malathion and chlorpyrifos, applied on citrus leaves and weathered outdoors, had longer residual effectiveness in killing flies than spinosad and fipronil. Residual effectiveness of the spinosad bait mixture waned significantly after 3 days of outdoor weathering. Results suggest that spinosad and fipronil can be potential alternatives for malathion in protein bait sprays.

Dietary intake of children aged 12-16 months and associations with maternal feeding beliefs and practices

This thesis provides the first detailed data describing the dietary intake of first-born Australian children aged 12-16 months. Overall, quality of intake could improve, with toddlers being exposed to energy-dense, nutrient-poor foods which may adversely affect the development of long-term healthy food preferences and growth trajectory. The leaner, but healthy weight toddler who exhibited more frequent food refusal was described a fussy eater or prompted higher maternal concern. However these behaviours are consistent with typical child development during the second year of life. Mothers can be supported to understand food refusal as manifestation of children's ability to self-regulate energy intake.

Current management of transitional feeding issues in preterm neonates born in Queensland, Australia

Background: Many preterm neonates display difficulty establishing suck-feeding competence in the weeks following birth. Ineffective management of transitional feeding issues may cause patient complications, and can contribute to increased length of stay. Aims: Given that many neonatal nurseries appear to vary in their neonatal feeding management practices, the aim of this study was to investigate and document the routine level of support and intervention currently provided for preterm neonates with transitional feeding issues across the various level II (special care) nurseries (SCNs) in Queensland, Australia. Methods: A questionnaire was mailed to all Queensland SCNs in 2005 (n = 36). The questionnaire contained a series of closed-choice and short-answer questions designed to obtain information from each SCN regarding their current practices for managing transitional feeding issues in preterm neonates. Results were confirmed during a follow-up phone call. Results: Responses were obtained from 29 SCNs (80.6%). None of these nurseries reported having any formal, written policies regarding the management of transitional feeding issues in preterm neonates. Wide variations were reported in relation to the suck-feeding assessments and interventions used by staff within the various SCNs. Of the 29 nurseries, 4 (13.8%) reported using checklists or assessments to judge readiness for suck-feeds, and 5 (17.2%) reported using pulse oximetry to judge tolerance of suck-feeding attempts. Eighteen SCNs (62.1%) reported offering some form of active intervention to assist neonates with transitional feeding issues, with the most common intervention techniques reported being non-nutritive sucking during tube feeds, pre-feeding oral stimulation, and actively pacing suck-feeds. Twenty-two SCNs (75.4%) reported having access to a lactation consultant to assist mothers with breastfeeding issues. Conclusions: Differences were reported in the routine management of transitional feeding issues in preterm neonates across the various SCNs in Queensland. It is suggested that evidence based guidelines need to be developed, and that, in order to do this, further research studies are required to determine current best practice, as well as to answer remaining questions. © 2008 Elsevier Ireland Ltd. All rights reserved.

Attainment of early feeding milestones in preterm neonates

Objective: This study aimed to document the ages at which preterm neonates commence suckle-feeds and attain exclusive suckle-feeding, as well as the time taken to transition from commencement of suckle-feeds to exclusive suckle-feeding. It was hypothesized that gestational age (GA) at birth and degree of neonatal morbidity would influence the timing of these early feeding milestones. Study Design: A chart review was conducted for all neonates born <37;0 weeks GA admitted to a tertiary level perinatal facility over a 12-month period (n=735). Complete data relating to attainment of feeding milestones were available on 472 neonates. Results: Correlation analysis indicated that both a low GA at birth and a high neonatal morbidity rating were statistically significantly correlated with an increased transition time from commencement of suckle-feeds to exclusive suckle-feeding. Cox regression indicated that both of these variables were statistically significant risk factors for a delayed GA at attainment of exclusive suckle-feeding. Conclusion: Preterm neonates who were less mature at birth and/or who displayed a greater degree of neonatal morbidity took longer to transition from starting suckle-feeds to achieving independent suckle-feeding, and were more mature at attainment of independent suckle-feeding.