Search

Simple Summary: In radiotherapy treatment, organs at risk and target volumes are contoured by the clinicians to prepare a dosimetry plan. In retrospective data, these structures are not often standardised to universal names across the patients plans, which is required to enable data mining and analysis. In this paper, a new method was proposed and evaluated to automatically standardise radiotherapy structures names using machine learning algorithms. The proposed approach was deployed over a dataset with 1613 patients collected from Liverpool & Macarthur Cancer Therapy Centres, New South Wales, Australia. It was concluded that machine learning techniques can standardise the dosimetry plan structures, taking into consideration the integration of multiple modalities representing each structure during the training process. In progressing the use of big data in health systems, standardised nomenclature is required to enable data pooling and analyses. In many radiotherapy planning systems and their data archives, target volumes (TV) and organ-at-risk (OAR) structure nomenclature has not been standardised. Machine learning (ML) has been utilised to standardise volumes nomenclature in retrospective datasets. However, only subsets of the structures have been targeted. Within this paper, we proposed a new approach for standardising all the structures nomenclature by using multi-modal artificial neural networks. A cohort consisting of 1613 breast cancer patients treated with radiotherapy was identified from Liverpool & Macarthur Cancer Therapy Centres, NSW, Australia. Four types of volume characteristics were generated to represent each target and OAR volume: textual features, geometric features, dosimetry features, and imaging data. Five datasets were created from the original cohort, the first four represented different subsets of volumes and the last one represented the whole list of volumes. For each dataset, 15 sets of combinations of features were generated to investigate the effect of using different characteristics on the standardisation performance. The best model reported 99.416% classification accuracy over the hold-out sample when used to standardise all the nomenclatures in a breast cancer radiotherapy plan into 21 classes. Our results showed that ML based automation methods can be used for standardising naming conventions in a radiotherapy plan taking into consideration the inclusion of multiple modalities to better represent each volume. [ABSTRACT FROM AUTHOR]

Background One of the strategies proven most successful in curbing rising rates of childhood obesity involves targeting parents as agents of change. Prior studies have focused on what messages to communicate, but few have investigated how they should be communicated. Objective To identify the channels most effective for communicating with parents of overweight and obese children and understand whether their use of parenting information sources differs from others in the community. Design/setting This study utilizes data from the Longitudinal Study of Australian Children ( LSAC). Families were included if weight and height information was available for parents and children at three data collection points: Waves 1, 2 and 4 (collected 2004, 2006 and 2010, respectively, n = 5107). Analysis A priori and a posteriori segmentation methods identified groups of parents that were similar in the sources used to obtain information about parenting, and examined whether some segments were more likely to have obese children. Results Four segments were identified that differed in their information source use: the 'personal networks', 'books', 'official sources' and 'mixed approach' segments. The 'official sources' and 'mixed approach' segments were most likely to have obese children, and they used doctors, government/community organizations and friends to obtain information on parenting. These segments were also less educated and had lower employment. Conclusions Messages are most likely to reach families with obese children if communicated through doctors, government publications and community organizations. Further, messages targeting social groupings of parents will leverage the power of advice from friends, which is another valuable information source for this group. [ABSTRACT FROM AUTHOR]

Study design:Cross-section design.Objectives:The development of reliable screen technology for predicting those at risk of depression in the long-term remains a challenge. The objective of this research was to determine factors that classify correctly adults with spinal cord injury (SCI) with depressed mood and to develop a diagnostic algorithm that could be applied for prediction of depressed mood in the long-term.Setting:SCI rehabilitation unit, rehabilitation outpatient clinic and Australian community.Methods:Participants included 107 adults with SCI. The assessment regimen included demographic and injury variables, negative mood states, pain intensity, health-related quality of life and self-efficacy. Participants were divided into those with 'normal' mood versus those with elevated depressed mood. Discriminant function analysis (DFA) was then used to isolate factors that in combination, best classify the presence or absence of depressed mood.Results:At the time of assessment, 24 participants (22.4%) had elevated depressed mood. DFA identified six factors that discriminated between those with depressed mood (P<0.01) and those with normal mood, explaining 61% of the variance. Factors consisted of pain intensity, mental health, emotional and social functioning, self-efficacy and fatigue. DFA correctly classified 91.7% (n=22 of 24) of those with depressed mood and 95.2% (n=79 of 83) of those without. Demographic, injury and physical health function variables were not found to discriminate depressed mood.Conclusion:Clinical implications of applying a diagnostic algorithm for detecting depression in adults with SCI are discussed. Prospective research is needed to test the predictive efficacy of the algorithm. [ABSTRACT FROM AUTHOR]