Alec Holt

Medical applications in case-based reasoning

This commentary summarizes case-based reasoning research applied in the medical domain. In this commentary the term ‘medical’ is used in an all-encompassing manner. It comprises all aspects of health, for example, from diagnosis to nutrition planning. This article provides references to researchers in the field, systems, workshops, and landmark publications.

Using accelerometers and GPS units to identify the proportion of daily physical activity located in parks with playgrounds in New Zealand children

OBJECTIVE To identify the proportion of childrens physical activity occurring in public parks with playgrounds. METHODS Children (n=184) aged 5 to 10 years were recruited from schools located in two low socio-economic status communities in Dunedin, New Zealand. Accelerometers (Actigraph GT1M) and global positioning system units (Globalsat DG-100) were used to quantify and identify where physical activity had occurred over a 7-day period. Cross-sectional data were collected October to December 2007 (southern hemisphere spring) and the childs height and weight were measured at school. RESULTS At least 84% of participants had accelerometer and global positioning system data available for five or more hourly periods per day, for at least 4 days. Overall, 1.9% of recorded activity took place at city parks (95% confidence interval: 1.4, 2.4), although this was 2.7% (95% confidence intervals: 0.7, 4.6) among obese children. CONCLUSIONS Accelerometers and GPS data was able to be successfully recorded and matched among this age group. The proportion of childrens activity taking place in parks with playgrounds was low, although this may still be important for some subgroups.

Applying case-based reasoning techniques in GIS

This paper outlines an approach in applying artificial intelligence techniques to solve spatial problems. The approach consists of combining casebased reasoning with geographical information systems to form a hybrid system to solve spatial problems. In this paper, the hybrid is applied to the problem of soil classification which is solved by searching the case base for other spatial cases similar to the problem case. The knowledge from the retrieved case is used to formulate an answer to the problem case. The paper compares the results from this new approach with a traditional method of soil classification.

Image processing in case-based reasoning

This commentary summarizes case-based reasoning (CBR) research applied to image processing. It includes references to the systems, workshops, and landmark publications. Image processing includes a variety of image formats, from computer tomography images to remote sensing and spatial data sets.

The Disparity Information and Communication Technology for Developing Countries has in the Delivery of Healthcare Information.

Information and Communication Technologies (ICT) have merged into the world of healthcare slowly but surely. However, the marriage between the use of technology and its full impact in the health sector has not been fully realised. The focus of this paper is to highlight the impact of ICT on revolutionising access to healthcare information and thus quality of health for populations of the developing world. This paper highlights on the importance of being able to access health information and how traditional media methods have been utilised to allow this within a developing country setting, highlighting the clear digital divide. The paper then addresses the impact of convergent communication technologies and mobile technologies in providing a means of addressing existing healthcare problems within a developing country setting.

Case-based reasoning with spatial data

This paper outlines the adaption of artificial intelligence techniques in spatial information systems. This adaption improves the analytical strength of spatial information systems. This paper includes a discussion of previously coupled techniques such as expert systems, fuzzy logic, hybrid connection systems and neural networks. It proposes a spatial reasoning prototype which is a based upon the integration of case-based reasoning and spatial information systems techniques.

Development of a 15-Item Scale to Measure Parental Perceptions of Their Neighborhood

Socioecological theory suggests that there are a range of influences that affect the physical activity levels of children, including parents’ perceptions of the neighborhood. A questionnaire instrument to quantify parental neighborhood perceptions was developed for the Location of Children’s Activity in Their Environment study as a potential predictor of children’s physical activity. A literature review revealed a lack of appropriate instruments, many containing highly localized items. Following initial instrument development including pretesting, where items showed face and content validity, a two-phase pilot study was undertaken. Pilot testing investigated test–retest reliability (overall intraclass correlation coefficient = .75) and internal consistency (α = .62 and α = .73 for the two phases, respectively), finding that the instrument was acceptable overall although some changes in wording were made to specific items with low reliability and/or internal consistency. Using data from the first year of the Location of Children’s Activity in Their Environment study, the instrument showed acceptable internal consistency (overall α = .73) and predictive validity (associated with time living in residence and walking to school) to be used in future studies. The scale was not found to be associated with child physical activity, which may be due to activity displacement. Proposed subscales for the neighborhood infrastructure and social aspects were marginal for internal consistency and a safety subscale was not acceptable in this regard. Given the impact of parental perceptions of the neighborhood, and its potential modifiable nature, a measure of it should be included in studies looking at the intensity and location of children’s physical activity.

Family history assessment for colorectal cancer (CRC) risk analysis - comparison of diagram- and questionnaire-based web interfaces

BackgroundColorectal cancer (CRC) has a high incidence, especially in New Zealand. The reasons for this are unknown. While most cancers develop sporadically, a positive family history, determined by the number and age at diagnosis of affected first and second degree relatives with CRC is one of the major factors, which may increase an individual’s lifetime risk. Before a patient can be enrolled in a surveillance program a detailed assessment and documentation of the family history is important but time consuming and often inaccurate. The documentation is usually paper-based. Our aim was therefore to develop and validate the usability and efficacy of a web-based family history assessment tool for CRC suitable for the general population. The tool was also to calculate the risk and make a recommendation for surveillance.MethodsTwo versions of an electronic assessment tool, diagram-based and questionnaire-based, were developed with the risk analysis and recommendations for surveillance based on the New Zealand Guidelines Group recommendations. Accuracy of our tool was tested prior to the study by comparing risk calculations based on family history by experienced gastroenterologists with the electronic assessment. The general public, visiting a local science fair were asked to use and comment on the usability of the two interfaces.ResultsNinety people assessed and commented on the two interfaces. Both interfaces were effective in assessing the risk to develop CRC through their familial history for CRC. However, the questionnaire-based interface performed with significantly better satisfaction (p = 0.001) than the diagram-based interface. There was no difference in efficacy though.ConclusionWe conclude that a web-based questionnaire tool can assist in the accurate documentation and analysis of the family history relevant to determine the individual risk of CRC based on local guidelines. The calculator is now implemented and assessable through the web-page of a local charity for colorectal cancer awareness and integral part of the local general practitioners’ e-referral system for colonic imaging.