The effect of prior visual information on recognition of speech and sounds

To identify and categorize complex stimuli such as familiar objects or speech, the human brain integrates information that is abstracted at multiple levels from its sensory inputs. Using cross-modal priming for spoken words and sounds, this functional magnetic resonance imaging study identified 3 distinct classes of visuoauditory incongruency effects: visuoauditory incongruency effects were selective for 1) spoken words in the left superior temporal sulcus (STS), 2) environmental sounds in the left angular gyrus (AG), and 3) both words and sounds in the lateral and medial prefrontal cortices (IFS/mPFC). From a cognitive perspective, these incongruency effects suggest that prior visual information influences the neural processes underlying speech and sound recognition at multiple levels, with the STS being involved in phonological, AG in semantic, and mPFC/IFS in higher conceptual processing. In terms of neural mechanisms, effective connectivity analyses (dynamic causal modeling) suggest that these incongruency effects may emerge via greater bottom-up effects from early auditory regions to intermediate multisensory integration areas (i.e., STS and AG). This is consistent with a predictive coding perspective on hierarchical Bayesian inference in the cortex where the domain of the prediction error (phonological vs. semantic) determines its regional expression (middle temporal gyrus/STS vs. AG/intraparietal sulcus).

Yhh1p/Cft1p directly links poly(A) site recognition and RNA polymerase II transcription termination

RNA polymerase II (pol II) transcription termination requires co‐transcriptional recognition of a functional polyadenylation signal, but the molecular mechanisms that transduce this signal to pol II remain unclear. We show that Yhh1p/Cft1p, the yeast homologue of the mammalian AAUAAA interacting protein CPSF 160, is an RNA‐binding protein and provide evidence that it participates in poly(A) site recognition. Interestingly, RNA binding is mediated by a central domain composed of predicted β‐propeller‐forming repeats, which occurs in proteins of diverse cellular functions. We also found that Yhh1p/Cft1p bound specifically to the phosphorylated C‐terminal domain (CTD) of pol II in vitro and in a two‐hybrid test in vivo. Furthermore, transcriptional run‐on analysis demonstrated that yhh1 mutants were defective in transcription termination, suggesting that Yhh1p/Cft1p functions in the coupling of transcription and 3′‐end formation. We propose that direct interactions of Yhh1p/Cft1p with both the RNA transcript and the CTD are required to communicate poly(A) site recognition to elongating pol II to initiate transcription termination.

Improving the accuracy of mine site water balances through improved estimation of runoff volumes

A mine site water balance is important for communicating information to interested stakeholders, for reporting on water performance, and for anticipating and mitigating water-related risks through water use/demand forecasting. Gaining accuracy over the water balance is therefore crucial for sites to achieve best practice water management and to maintain their social license to operate. For sites that are located in high rainfall environments the water received to storage dams through runoff can represent a large proportion of the overall inputs to site; inaccuracies in these flows can therefore lead to inaccuracies in the overall site water balance. Hydrological models that estimate runoff flows are often incorporated into simulation models used for water use/demand forecasting. The Australian Water Balance Model (AWBM) is one example that has been widely applied in the Australian context. However, the calibration of AWBM in a mining context can be challenging. Through a detailed case study, we outline an approach that was used to calibrate and validate AWBM at a mine site. Commencing with a dataset of monitored dam levels, a mass balance approach was used to generate an observed runoff sequence. By incorporating a portion of this observed dataset into the calibration routine, we achieved a closer fit between the observed vs. simulated dataset compared with the base case. We conclude by highlighting opportunities for future research to improve the calibration fit through improving the quality of the input dataset. This will ultimately lead to better models for runoff prediction and thereby improve the accuracy of mine site water balances.

Blood transfer devices for malaria rapid diagnostic tests : evaluation of accuracy, safety and ease of use

Background: Malaria rapid diagnostic tests (RDTs) are increasingly used by remote health personnel with minimal training in laboratory techniques. RDTs must, therefore, be as simple, safe and reliable as possible. Transfer of blood from the patient to the RDT is critical to safety and accuracy, and poses a significant challenge to many users. Blood transfer devices were evaluated for accuracy and precision of volume transferred, safety and ease of use, to identify the most appropriate devices for use with RDTs in routine clinical care. Methods: Five devices, a loop, straw-pipette, calibrated pipette, glass capillary tube, and a new inverted cup device, were evaluated in Nigeria, the Philippines and Uganda. The 227 participating health workers used each device to transfer blood from a simulated finger-prick site to filter paper. For each transfer, the number of attempts required to collect and deposit blood and any spilling of blood during transfer were recorded. Perceptions of ease of use and safety of each device were recorded for each participant. Blood volume transferred was calculated from the area of blood spots deposited on filter paper. Results: The overall mean volumes transferred by devices differed significantly from the target volume of 5 microliters (p < 0.001). The inverted cup (4.6 microliters) most closely approximated the target volume. The glass capillary was excluded from volume analysis as the estimation method used is not compatible with this device. The calibrated pipette accounted for the largest proportion of blood exposures (23/225, 10%); exposures ranged from 2% to 6% for the other four devices. The inverted cup was considered easiest to use in blood collection (206/ 226, 91%); the straw-pipette and calibrated pipette were rated lowest (143/225 [64%] and 135/225 [60%] respectively). Overall, the inverted cup was the most preferred device (72%, 163/227), followed by the loop (61%, 138/227). Conclusions: The performance of blood transfer devices varied in this evaluation of accuracy, blood safety, ease of use, and user preference. The inverted cup design achieved the highest overall performance, while the loop also performed well. These findings have relevance for any point-of-care diagnostics that require blood sampling.

Overview of the ShARe/CLEF eHealth Evaluation Lab 2014

This paper reports on the 2nd ShARe/CLEFeHealth evaluation lab which continues our evaluation resource building activities for the medical domain. In this lab we focus on patients' information needs as opposed to the more common campaign focus of the specialised information needs of physicians and other healthcare workers. The usage scenario of the lab is to ease patients and next-of-kins' ease in understanding eHealth information, in particular clinical reports. The 1st ShARe/CLEFeHealth evaluation lab was held in 2013. This lab consisted of three tasks. Task 1 focused on named entity recognition and normalization of disorders; Task 2 on normalization of acronyms/abbreviations; and Task 3 on information retrieval to address questions patients may have when reading clinical reports. This year's lab introduces a new challenge in Task 1 on visual-interactive search and exploration of eHealth data. Its aim is to help patients (or their next-of-kin) in readability issues related to their hospital discharge documents and related information search on the Internet. Task 2 then continues the information extraction work of the 2013 lab, specifically focusing on disorder attribute identification and normalization from clinical text. Finally, this year's Task 3 further extends the 2013 information retrieval task, by cleaning the 2013 document collection and introducing a new query generation method and multilingual queries. De-identified clinical reports used by the three tasks were from US intensive care and originated from the MIMIC II database. Other text documents for Tasks 1 and 3 were from the Internet and originated from the Khresmoi project. Task 2 annotations originated from the ShARe annotations. For Tasks 1 and 3, new annotations, queries, and relevance assessments were created. 50, 79, and 91 people registered their interest in Tasks 1, 2, and 3, respectively. 24 unique teams participated with 1, 10, and 14 teams in Tasks 1, 2 and 3, respectively. The teams were from Africa, Asia, Canada, Europe, and North America. The Task 1 submission, reviewed by 5 expert peers, related to the task evaluation category of Effective use of interaction and targeted the needs of both expert and novice users. The best system had an Accuracy of 0.868 in Task 2a, an F1-score of 0.576 in Task 2b, and Precision at 10 (P@10) of 0.756 in Task 3. The results demonstrate the substantial community interest and capabilities of these systems in making clinical reports easier to understand for patients. The organisers have made data and tools available for future research and development.

Topic dependent language modelling for spoken term detection

This paper investigates the effect of topic dependent language models (TDLM) on phonetic spoken term detection (STD) using dynamic match lattice spotting (DMLS). Phonetic STD consists of two steps: indexing and search. The accuracy of indexing audio segments into phone sequences using phone recognition methods directly affects the accuracy of the final STD system. If the topic of a document in known, recognizing the spoken words and indexing them to an intermediate representation is an easier task and consequently, detecting a search word in it will be more accurate and robust. In this paper, we propose the use of TDLMs in the indexing stage to improve the accuracy of STD in situations where the topic of the audio document is known in advance. It is shown that using TDLMs instead of the traditional general language model (GLM) improves STD performance according to figure of merit (FOM) criteria.

Emotion recognition of static and dynamic faces in autism spectrum disorder

There is substantial evidence for facial emotion recognition (FER) deficits in autism spectrum disorder (ASD). The extent of this impairment, however, remains unclear, and there is some suggestion that clinical groups might benefit from the use of dynamic rather than static images. High-functioning individuals with ASD (n = 36) and typically developing controls (n = 36) completed a computerised FER task involving static and dynamic expressions of the six basic emotions. The ASD group showed poorer overall performance in identifying anger and disgust and were disadvantaged by dynamic (relative to static) stimuli when presented with sad expressions. Among both groups, however, dynamic stimuli appeared to improve recognition of anger. This research provides further evidence of specific impairment in the recognition of negative emotions in ASD, but argues against any broad advantages associated with the use of dynamic displays.

Human identification from video using advanced gait recognition techniques

The solutions proposed in this thesis contribute to improve gait recognition performance in practical scenarios that further enable the adoption of gait recognition into real world security and forensic applications that require identifying humans at a distance. Pioneering work has been conducted on frontal gait recognition using depth images to allow gait to be integrated with biometric walkthrough portals. The effects of gait challenging conditions including clothing, carrying goods, and viewpoint have been explored. Enhanced approaches are proposed on segmentation, feature extraction, feature optimisation and classification elements, and state-of-the-art recognition performance has been achieved. A frontal depth gait database has been developed and made available to the research community for further investigation. Solutions are explored in 2D and 3D domains using multiple images sources, and both domain-specific and independent modality gait features are proposed.

Towards pose-robust face recognition on video

This thesis investigates face recognition in video under the presence of large pose variations. It proposes a solution that performs simultaneous detection of facial landmarks and head poses across large pose variations, employs discriminative modelling of feature distributions of faces with varying poses, and applies fusion of multiple classifiers to pose-mismatch recognition. Experiments on several benchmark datasets have demonstrated that improved performance is achieved using the proposed solution.

Aspect-based opinion extraction from customer reviews

Text is the main method of communicating information in the digital age. Messages, blogs, news articles, reviews, and opinionated information abounds on the Internet. People commonly purchase products online and post their opinions about purchased items. This feedback is displayed publicly to assist others with their purchasing decisions, creating the need for a mechanism with which to extract and summarize useful information for enhancing the decision-making process. Our contribution is to improve the accuracy of extraction by combining different techniques from three major areas, named Data Mining, Natural Language Processing techniques and Ontologies. The proposed framework sequentially mines product’s aspects and users’ opinions, groups representative aspects by similarity, and generates an output summary. This paper focuses on the task of extracting product aspects and users’ opinions by extracting all possible aspects and opinions from reviews using natural language, ontology, and frequent “tag” sets. The proposed framework, when compared with an existing baseline model, yielded promising results.

Recognising emotional expressions of complaining customers : A cross-cultural study

Purpose This study aims to test service providers’ ability to recognise non-verbal emotions in complaining customers of same and different cultures. Design/methodology/approach In a laboratory study, using a between-subjects experimental design (n = 153), we tested the accuracy of service providers’ perceptions of the emotional expressions of anger, fear, shame and happiness of customers from varying cultural backgrounds. After viewing video vignettes of customers complaining (with the audio removed), participants (in the role of service providers) assessed the emotional state of the customers portrayed in the video. Findings Service providers in culturally mismatched dyads were prone to misreading anger, happiness and shame expressed by dissatisfied customers. Happiness was misread in the displayed emotions of both dyads. Anger was recognisable in the Anglo customers but not Confucian Asian, while Anglo service providers misread both shame and happiness in Confucian Asian customers. Research limitations/implications The study was conducted in the laboratory and was based solely on participant’s perceptions of actors’ non-verbal facial expressions in a single encounter. Practical implications Given the level of ethnic differences in developed nations, a culturally sensitive workplace is needed to foster effective functioning of service employee teams. Ability to understand cultural display rules and to recognise and interpret emotions is an important skill for people working in direct contact with customers. Originality/value This research addresses the lack of empirical evidence for the recognition of customer emotions by service providers and the impact of cross-cultural differences.

An accuracy assessment of ASTER stereo images-derived digital elevation model by using rational polynomial coefficient model

The along-track stereo images of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor with 15 m resolution were used to generate Digital Elevation Model (DEM) on an area with low and near Mean Sea Level (MSL) elevation in Johor, Malaysia. The absolute DEM was generated by using the Rational Polynomial Coefficient (RPC) model which was run on ENVI 4.8 software. In order to generate the absolute DEM, 60 Ground Control Pointes (GCPs) with almost vertical accuracy less than 10 meter extracted from topographic map of the study area. The assessment was carried out on uncorrected and corrected DEM by utilizing dozens of Independent Check Points (ICPs). Consequently, the uncorrected DEM showed the RMSEz of ± 26.43 meter which was decreased to the RMSEz of ± 16.49 meter for the corrected DEM after post-processing. Overall, the corrected DEM of ASTER stereo images met the expectations.