Clinician-driven automated classification of limb fractures from free-text radiology reports

The aim of this research is to report initial experimental results and evaluation of a clinician-driven automated method that can address the issue of misdiagnosis from unstructured radiology reports. Timely diagnosis and reporting of patient symptoms in hospital emergency departments (ED) is a critical component of health services delivery. However, due to disperse information resources and vast amounts of manual processing of unstructured information, a point-of-care accurate diagnosis is often difficult. A rule-based method that considers the occurrence of clinician specified keywords related to radiological findings was developed to identify limb abnormalities, such as fractures. A dataset containing 99 narrative reports of radiological findings was sourced from a tertiary hospital. The rule-based method achieved an F-measure of 0.80 and an accuracy of 0.80. While our method achieves promising performance, a number of avenues for improvement were identified using advanced natural language processing (NLP) techniques.

Automatic de-identification of electronic health records : an Australian perspective

We present an approach to automatically de-identify health records. In our approach, personal health information is identified using a Conditional Random Fields machine learning classifier, a large set of linguistic and lexical features, and pattern matching techniques. Identified personal information is then removed from the reports. The de-identification of personal health information is fundamental for the sharing and secondary use of electronic health records, for example for data mining and disease monitoring. The effectiveness of our approach is first evaluated on the 2007 i2b2 Shared Task dataset, a widely adopted dataset for evaluating de-identification techniques. Subsequently, we investigate the robustness of the approach to limited training data; we study its effectiveness on different type and quality of data by evaluating the approach on scanned pathology reports from an Australian institution. This data contains optical character recognition errors, as well as linguistic conventions that differ from those contained in the i2b2 dataset, for example different date formats. The findings suggest that our approach compares to the best approach from the 2007 i2b2 Shared Task; in addition, the approach is found to be robust to variations of training size, data type and quality in presence of sufficient training data.

Towards a smart cancer registry

Aims Pathology notification for a Cancer Registry is regarded as the most valid information for the confirmation of a diagnosis of cancer. In view of the importance of pathology data, an automatic medical text analysis system (Medtex) is being developed to perform electronic Cancer Registry data extraction and coding of important clinical information embedded within pathology reports. Methods The system automatically scans HL7 messages received from a Queensland pathology information system and analyses the reports for terms and concepts relevant to a cancer notification. A multitude of data items for cancer notification such as primary site, histological type, stage, and other synoptic data are classified by the system. The underlying extraction and classification technology is based on SNOMED CT1 2. The Queensland Cancer Registry business rules3 and International Classification of Diseases – Oncology – Version 34 have been incorporated. Results The cancer notification services show that the classification of notifiable reports can be achieved with sensitivities of 98% and specificities of 96%5, while the coding of cancer notification items such as basis of diagnosis, histological type and grade, primary site and laterality can be extracted with an overall accuracy of 80%6. In the case of lung cancer staging, the automated stages produced were accurate enough for the purposes of population level research and indicative staging prior to multi-disciplinary team meetings2 7. Medtex also allows for detailed tumour stream synoptic reporting8. Conclusions Medtex demonstrates how medical free-text processing could enable the automation of some Cancer Registry processes. Over 70% of Cancer Registry coding resources are devoted to information acquisition. The development of a clinical decision support system to unlock information from medical free-text could significantly reduce costs arising from duplicated processes and enable improved decision support, enhancing efficiency and timeliness of cancer information for Cancer Registries.

Automatic classification of free-text radiology reports to identify limb fractures using machine learning and the SNOMED CT ontology

Objective To develop and evaluate machine learning techniques that identify limb fractures and other abnormalities (e.g. dislocations) from radiology reports. Materials and Methods 99 free-text reports of limb radiology examinations were acquired from an Australian public hospital. Two clinicians were employed to identify fractures and abnormalities from the reports; a third senior clinician resolved disagreements. These assessors found that, of the 99 reports, 48 referred to fractures or abnormalities of limb structures. Automated methods were then used to extract features from these reports that could be useful for their automatic classification. The Naive Bayes classification algorithm and two implementations of the support vector machine algorithm were formally evaluated using cross-fold validation over the 99 reports. Result Results show that the Naive Bayes classifier accurately identifies fractures and other abnormalities from the radiology reports. These results were achieved when extracting stemmed token bigram and negation features, as well as using these features in combination with SNOMED CT concepts related to abnormalities and disorders. The latter feature has not been used in previous works that attempted classifying free-text radiology reports. Discussion Automated classification methods have proven effective at identifying fractures and other abnormalities from radiology reports (F-Measure up to 92.31%). Key to the success of these techniques are features such as stemmed token bigrams, negations, and SNOMED CT concepts associated with morphologic abnormalities and disorders. Conclusion This investigation shows early promising results and future work will further validate and strengthen the proposed approaches.

De-Identification of health records using Anonym: Effectiveness and robustness across datasets

Objective Evaluate the effectiveness and robustness of Anonym, a tool for de-identifying free-text health records based on conditional random fields classifiers informed by linguistic and lexical features, as well as features extracted by pattern matching techniques. De-identification of personal health information in electronic health records is essential for the sharing and secondary usage of clinical data. De-identification tools that adapt to different sources of clinical data are attractive as they would require minimal intervention to guarantee high effectiveness. Methods and Materials The effectiveness and robustness of Anonym are evaluated across multiple datasets, including the widely adopted Integrating Biology and the Bedside (i2b2) dataset, used for evaluation in a de-identification challenge. The datasets used here vary in type of health records, source of data, and their quality, with one of the datasets containing optical character recognition errors. Results Anonym identifies and removes up to 96.6% of personal health identifiers (recall) with a precision of up to 98.2% on the i2b2 dataset, outperforming the best system proposed in the i2b2 challenge. The effectiveness of Anonym across datasets is found to depend on the amount of information available for training. Conclusion Findings show that Anonym compares to the best approach from the 2006 i2b2 shared task. It is easy to retrain Anonym with new datasets; if retrained, the system is robust to variations of training size, data type and quality in presence of sufficient training data.

Automated classification of limb fractures from free-text radiology reports using a clinician-informed gazetteer methodology

Background Timely diagnosis and reporting of patient symptoms in hospital emergency departments (ED) is a critical component of health services delivery. However, due to dispersed information resources and a vast amount of manual processing of unstructured information, accurate point-of-care diagnosis is often difficult. Aims The aim of this research is to report initial experimental evaluation of a clinician-informed automated method for the issue of initial misdiagnoses associated with delayed receipt of unstructured radiology reports. Method A method was developed that resembles clinical reasoning for identifying limb abnormalities. The method consists of a gazetteer of keywords related to radiological findings; the method classifies an X-ray report as abnormal if it contains evidence contained in the gazetteer. A set of 99 narrative reports of radiological findings was sourced from a tertiary hospital. Reports were manually assessed by two clinicians and discrepancies were validated by a third expert ED clinician; the final manual classification generated by the expert ED clinician was used as ground truth to empirically evaluate the approach. Results The automated method that attempts to individuate limb abnormalities by searching for keywords expressed by clinicians achieved an F-measure of 0.80 and an accuracy of 0.80. Conclusion While the automated clinician-driven method achieved promising performances, a number of avenues for improvement were identified using advanced natural language processing (NLP) and machine learning techniques.

Long time, no see?

Long Time, No See? is a crowd-sourced project that asks people to reflect upon what kind of long term future they would each like to promote. It is an evolving experiment in the social practice of ‘everyday futuring’. To participate download the Long Time, No See? IPhone APP that gently guides you during a short walk, encouraging you to experience new places, sensations and thoughts in your locality. At nine stages along that journey you donate ‘field notes’ as images, texts, sounds and ‘themes’, offering a unique opportunity to reveal possible pathways towards more sustaining futures. The APP records the shape of your walk on the ground and draws an island on the ‘map’ shown here, populated by your nine sets of responses. The themes you have chosen then connect your island into an evolving ‘world’ map of connections and possibilities, which you can then explore at your leisure. In these ways, Long Time, No See? doesn’t ask you for lofty visions or ask you to lay out a program of action, but instead asks you to consider what is around you today, steering your eyes, ears and embodied experiences towards new futures that demonstrate your ‘care’ for what comes after you. Please use the contribute tab below to learn how to add your voice! PARTICIPATE To contribute 1: Download the APP {bit.do/ltns}, iPhone/iPad is supported right now. 2: Register a ‘walker name’. 3: Take a leisurely walk (30 -60mins) and contribute image, text, sound and themes when asked. 4: Wait while we verify and upload your walk (allow about 24 hours) 5: View your contributions via your ‘walker name’ and discover how it relates to others, here at the Cube and at www.long-time-no-see.org. NB You can undertake each walk over more than one day if that suits. You may even drive, cycle or move by other modes. DOWNLOAD THE APP: bit.do/ltns (insert QI Code) FIND OUT MORE www.long-time-no-see.org

The re-introduction project

The Re-introduction Project began with an art-science research residency in 2012, funded through the Australian 'Synapse' art-science residency program. It was developed in partnership with the Australian Wildlife Conservancy, Australia's largest private conservation agency and their South-East regional scientist Matt Hayward and conducted through a series of seven high intensity field-trips to AWC’s remote properties in VIC, NSW and SA. These trips coincided with key times at which the AWC’s mobile scientific teams were undertaking intensive scientific activities. The program coincided with specific events that senior scientist collaborator Dr Matt Hayward led in 2012 at Mallee Regions (Yookamurra, Scotia and Buckaringa), Lake Eyre Basin (Kalamurina) and Sydney (North Head). The initial outcome of the project was the work Pitfall (An Opportunistic Survey) - a new media installation created in light, media, object, text and sound presented near the AWC headquarters at Mildura in far NW Victoria. Pitfall built upon ideas and cross disciplinary processes developed during this residency/collaboration with Australian Wildlife Conservancy inspired by working with their ecological scientists during pitfall-trap survey events used to survey small mammals and invertebrates. ‘Pitfall’ was designed in response to a playful survey that I asked the AWC scientists to engage with around ideas of avoiding ecological pitfalls into the future. This continually-evolving artwork was built from multiple screens, a tabletop landscape mapped with projections, fibre optics, 3D spatial sound and infrared night imagery.

Using the internet to assist court processes : delivery of justice in an electronic age

This article presents an overview of two aspects of the role the internet now plays in the court system - first, the extent to which judges, administrators and court officials at the different levels in the court hierarchy are using the internet to deliver enhanced access to the Australian justice system for the community as a whole, and second, how they have embraced that same technology as an aid for accessing information for better judgment delivery and administration.

Adherence to physical activity and electronic media guidelines in Australian pre-school children

The aim of this study was to determine compliance with the National Association for Sport and Physical Education (NASPE) guideline for physical activity and American Academy of Pediatrics (AAP) recommendation for electronic media use among urban pre-school children in two large cities on the East Coast of Australia. Cross-sectional data were collected from 266 parents. Time spent using electronic media (watching television, DVDs or on the computer) and in physical activity were parent reported. The proportion who met each guideline was calculated. 56 per cent and 79% of children met the NASPE guideline on weekdays and weekends, respectively, while 73% and 70% met the AAP recommendation on weekdays and weekends, with no difference between boys and girls. A substantial minority do not meet physical activity and electronic media use recommendations, highlighting the need to better understand what factors contribute to physical activity and electronic media use among this group of pre-schoolers.

The role of perceived usefulness and attitude on electronic health record acceptance

Information and communications technologies are a significant component of the healthcare domain, and electronic health records play a major role in it. Therefore, it is important that they are accepted en masse by healthcare professionals. How healthcare professionals perceive the usefulness of electronic health records and their attitudes towards them have been shown to have significant effects on the overall acceptance in many healthcare systems around the world. This paper investigates the role of perceived usefulness and attitude on the intention to use electronic health records by future healthcare professionals using polynomial regression with response surface analysis. Results show that the relationships between these variables are more complex than predicted in prior research. The paper concludes that the properties of the above determinants must be further investigated to clearly understand: (i) their role in predicting the intention to use electronic health records; and (ii) in designing systems that are better adopted by healthcare professionals of the future.

Progress and challenges in the implementation of Electronic Medical Records in Saudi Arabia : a systematic review

Electronic Medical Record (EMR) systems are being implemented increasingly worldwide. Saudi Arabia is one of the developing countries that commenced implementing such systems in 1988. Whilst EMR uptake has been low in Saudi Arabia until now, a number of hospitals have implemented EMR systems successfully. This paper analyses available studies (n = 28) in the literature regarding EMR implementation in Saudi Arabia to identify the progress of EMR implementation to date and to identify the facilitators and barriers to implementation.

Privacy enhanced electronic cheque system

With the introduction of Check 21 law and the development of FSTC's echeck system, there has been an increasing usage of e-cheque conversions and acceptance among retailers, banks, and consumers. However, the current e-cheque system does not address issues concerning privacy, confidentiality, and traceability. We highlight the issues concerning the current electronic cheque system and provide a solution to overcome those drawbacks.