Temporal bone radiology report classification using open source machine learning and natural langue processing libraries.

BACKGROUND: Radiology reports are a rich resource for biomedical research. Prior to utilization, trained experts must manually review reports to identify discrete outcomes. The Audiological and Genetic Database (AudGenDB) is a public, de-identified research database that contains over 16,000 radiology reports. Because the reports are unlabeled, it is difficult to select those with specific abnormalities. We implemented a classification pipeline using a human-in-the-loop machine learning approach and open source libraries to label the reports with one or more of four abnormality region labels: inner, middle, outer, and mastoid, indicating the presence of an abnormality in the specified ear region. METHODS: Trained abstractors labeled radiology reports taken from AudGenDB to form a gold standard. These were split into training (80 %) and test (20 %) sets. We applied open source libraries to normalize and convert every report to an n-gram feature vector. We trained logistic regression, support vector machine (linear and Gaussian), decision tree, random forest, and naïve Bayes models for each ear region. The models were evaluated on the hold-out test set. RESULTS: Our gold-standard data set contained 726 reports. The best classifiers were linear support vector machine for inner and outer ear, logistic regression for middle ear, and decision tree for mastoid. Classifier test set accuracy was 90 %, 90 %, 93 %, and 82 % for the inner, middle, outer and mastoid regions, respectively. The logistic regression method was very consistent, achieving accuracy scores within 2.75 % of the best classifier across regions and a receiver operator characteristic area under the curve of 0.92 or greater across all regions. CONCLUSIONS: Our results indicate that the applied methods achieve accuracy scores sufficient to support our objective of extracting discrete features from radiology reports to enhance cohort identification in AudGenDB. The models described here are available in several free, open source libraries that make them more accessible and simplify their utilization as demonstrated in this work. We additionally implemented the models as a web service that accepts radiology report text in an HTTP request and provides the predicted region labels. This service has been used to label the reports in AudGenDB and is freely available.

Listening effort and fatigue: what exactly are we measuring? A British Society of Audiology Cognition in Hearing Special Interest Group 'white paper'.

OBJECTIVE: There is growing interest in the concepts of listening effort and fatigue associated with hearing loss. However, the theoretical underpinnings and clinical meaning of these concepts are unclear. This lack of clarity reflects both the relative immaturity of the field and the fact that research studies investigating listening effort and fatigue have used a variety of methodologies including self-report, behavioural, and physiological measures. DESIGN: This discussion paper provides working definitions for listening effort and listening-related fatigue. Using these definitions as a framework, methodologies to assess these constructs are reviewed. RESULTS: Although each technique attempts to characterize the same construct (i.e. the clinical presentation of listening effort and fatigue), different assumptions are often made about the nature of these phenomena and their behavioural and physiological manifestations. CONCLUSION: We suggest that researchers consider these assumptions when interpreting their data and, where possible, make predictions based on current theoretical knowledge to add to our understanding of the underlying mechanisms of listening effort and listening-related fatigue. FOREWORD: Following recent interest in the cognitive involvement in hearing, the British Society of Audiology (BSA) established a Special Interest Group on Cognition in Hearing in May 2013. In an exploratory group meeting, the ambiguity surrounding listening effort and fatigue was discussed. To address this problem, the group decided to develop a 'white paper' on listening effort and fatigue. This is a discussion document followed by an international set of commentaries from leading researchers in the field. An approach was made to the editor of the International Journal of Audiology who agreed to this suggestion. This paper, and the associated commentaries that follow, are the result.

A audiometria tonal liminar e/ou avaliação instrumental confirmam posteriormente a audiometria de tronco cerebral obtida precocemente em crianças deficientes atuditivas? / Does Pure Tone Threshold Audiometry and/or Behavorial Auditory Assessment Confirm Subsequent Response Obtained Precociously in Children with Auditive Deficiency?

Esta pesquisa foi realizada a partir do levantamento dos diagnósticos audiológicos obtidos através da audiometria de tronco cerebral (ATC), realizada precocemente, e da audiometria tonal liminar (ATO e/ou avaliação instrumental (AI), realizadas anos depois nas mesmas 30 crianças com deficiência auditiva em atendimento no setor de Audiologia Educacional do Serviço de Atendimento Fonoaudiológico da Universidade Federal de Santa Maria, com o objetivo de compara-loa e verificar a concordância existente entre eles. Resultados: A comparação, levando em conta os graus de perda auditiva obtidos através dos três procedimentos de avaliação considerados, mostrou uma concordância de 97,30% dos diagnósticos nas perdas auditivas de grau profundo, 77,78% no grau severo, 57,15% no grau moderado e de 33,33% no grau moderadamente-severo. No geral, do ponto de vista clínico, verificou-se concordância diagnóstica em 57 (95,00%) das 60 orelhas pesquisadas.

This research was made after survey of the audiologics diagnostics obtained through of the auditory brainstem response (ABR), pure tone threshold audiometry (PTA) and/or behavioral auditory assessment (BAA) made in 30 children with auditive deficiency with the purpose of to compare and Check the agreement between them. Results: In the comparison of the diagnostics of the grades of hearing loss obtained watched a agreement of 97,30% of the diagnostics in the hearing losses of profound grade, 77,78% in the severe grade, 57,15% in the moderate grade and 33,33% in the moderately-severe; in general, of the clinical point of view, watched diagnostic agreement in 57 (95,00%) of the 60 ears researched.