Using supervised and unsupervised techniques to determine groups of patients with different doctor-patient stability

Decision trees and self organising feature maps (SOFM) are frequently used to identify groups. This research aims to compare the similarities between any groupings found between supervised (Classification and Regression Trees - CART) and unsupervised classification (SOFM), and to identify insights into factors associated with doctor-patient stability. Although CART and SOFM uses different learning paradigms to produce groupings, both methods came up with many similar groupings. Both techniques showed that self perceived health and age are important indicators of stability. In addition, this study has indicated profiles of patients that are at risk which might be interesting to general practitioners.

The importance of neighbourhood size in self organising systems

In recent times, the analysis of SOM (self-organising map) performance has concentrated on optimising the gain decay, rather than the size, form and decay of the neighbourhood function. We propose that the size, form and decay of region size plays a much more significant role in the learning, and especially in the development, of topographic feature maps. In this paper, a biologically-derived SOM model is presented. This model is able to select a single winning neuron and to form Gaussian outputs about this winner, without the need for a meta-level decision-making structure to artificially select a winner and fit a Gaussian output to that winner. Using this model, some fundamental characteristics of the relationship between neighbourhood size and SOM output states are demonstrated.

Clustering massive high dimensional data with dynamic feature maps

This paper presents an algorithm based on the Growing Self Organizing Map (GSOM) called the High Dimensional Growing Self Organizing Map with Randomness (HDGSOMr) that can cluster massive high dimensional data efficiently. The original GSOM algorithm is altered to accommodate for the issues related to massive high dimensional data. These modifications are presented in detail with experimental results of a massive real-world dataset.

Recognition and response to human movement contained in motion capture data using the Self Organising Map

Neural Networks have been used successfully for recognition of human gestures in many applications including analysis of motion capture data. This paper investigates the potential for using the same methods for both recognition and synthesising responses in relation to movement contained in motion capture sequences.

Beyond the topological map : developing alternate mappings in self organisation

The self organising map is a well established unsupervised
learning technique which is able to form sophisticated representations of an input data set. However, conventional Self Organising Map (SOM) algorithms are limited to the production of topological maps — that is, maps where distance between points on the map have a direct relationship to the Euclidean distance between the training vectors corresponding to those points.

It would be desirable to be able to create maps which form clusters on primitive attributes other than Euclidean distance; for example, clusters based upon orientation or shape. Such maps could provide a novel approach to pattern recognition tasks by providing a new method to associate groups of data.

In this paper, it is shown that the type of map produced by SOM algorithms is a direct consequence of the lateral connection strategy employed. Given this knowledge, a technique is required to establish the feasability of using an alternative lateral connection strategy. Such a technique is presented. Using this technique, it is possible to rule out lateral connection strategies that will not produce output states useful to the organisation process. This technique is demonstrated using conventional Laplacian interconnection as well as a number of novel interconnection strategies.

Analysing stillbirth data using dynamic self organizing maps

Even with the presence of modern obstetric care, stillbirth rate seems to stay stagnant or has even risen slightly in countries such as England and has become a significant public health concern [1]. In the light of current medical research, maternal risk factors such as diabetes and hypertensive disease were identified as possible risk factors and are taken into consideration in antenatal care. However, medical practitioners and researchers suspect possible relationships between trends in maternal demographics, antenatal care and pregnancy information of current stillbirth in consideration [2]. Although medical data and knowledge is available appropriate computing techniques to analyze the data may lead to identification of high risk groups. In this paper we use an unsupervised clustering technique called Growing Self organizing Map (GSOM) to analyse the stillbirth data and present patterns which can be important to medical researchers.

Growing self-organizing map for online continuous clustering

The internet age has fuelled an enormous explosion in the amount of information generated by humanity. Much of this information is transient in nature, created to be immediately consumed and built upon (or discarded). The field of data mining is surprisingly scant with algorithms that are geared towards the unsupervised knowledge extraction of such dynamic data streams. This chapter describes a new neural network algorithm inspired by self-organising maps. The new algorithm is a hybrid algorithm from the growing self-organising map (GSOM) and the cellular probabilistic self-organising map (CPSOM). The result is an algorithm which generates a dynamically growing feature map for the purpose of clustering dynamic data streams and tracking clusters as they evolve in the data stream.

An empirical study of neighbourhood decay in Kohonen's self organizing map

In this paper, empirical results are presented which suggest that size and rate of decay of region size plays a much more significant role in the learning, and especially the development, of topographic feature maps. Using these results as a basis, a scheme for decaying region size during SOM training is proposed. The proposed technique provides near optimal training time. This scheme avoids the need for sophisticated learning gain decay schemes, and precludes the need for a priori knowledge of likely training times. This scheme also has some potential uses for continuous learning.

Growing Self Organizing Map with an Imposed Binary Search Tree for Discovering Temporal Input Patterns

In this paper the Binary Search Tree Imposed Growing Self Organizing Map (BSTGSOM) is presented as an extended version of the Growing Self Organizing Map (GSOM), which has proven advantages in knowledge discovery applications. A Binary Search Tree imposed on the GSOM is mainly used to investigate the dynamic perspectives of the GSOM based on the inputs and these generated temporal patterns are stored to further analyze the behavior of the GSOM based on the input sequence. Also, the performance advantages are discussed and compared with that of the original GSOM.

Can a disease-specific education program augment self-management skills and improve Health-Related Quality of Life in people with hip or knee osteoarthritis?

Background: Patient education and self-management programs are offered in many countries to people with chronic conditions such as osteoarthritis (OA). The most well-known is the disease-specific Stanford Arthritis Self-Management Program (ASMP). While Australian and international clinical guidelines promote the concept of self-management for OA, there is currently little evidence to support the use of the ASMP. Several meta-analyses have reported that arthritis self-management programs had minimal or no effect on reducing pain and disability. However, previous studies have had methodological shortcomings including the use of outcome measures which do not accurately reflect program goals. Additionally, limited cost-effectiveness analyses have been undertaken and the cost-utility of the program has not been explored.

Methods/design: This study is a randomised controlled trial to determine the efficacy (in terms of Health-Related Quality of Life and self-management skills) and cost-utility of a 6-week group-based Stanford ASMP for people with hip or knee OA.

Six hundred participants referred to an orthopaedic surgeon or rheumatologist for hip or knee OA will be recruited from outpatient clinics at 2 public hospitals and community-based private practices within 2 private hospital settings in Victoria, Australia. Participants must be 18 years or over, fluent in English and able to attend ASMP sessions. Exclusion criteria include cognitive dysfunction, previous participation in self-management programs and placement on a waiting list for joint replacement surgery or scheduled joint replacement.

Eligible, consenting participants will be randomised to an intervention group (who receive the ASMP and an arthritis self-management book) or a control group (who receive the book only). Follow-up will be at 6 weeks, 3 months and 12 months using standardised self-report measures. The primary outcome is Health-Related Quality of Life at 12 months, measured using the Assessment of Quality of Life instrument. Secondary outcome measures include the Health Education Impact Questionnaire, Western Ontario and McMaster Universities Osteoarthritis Index (pain subscale and total scores), Kessler Psychological Distress Scale and the Hip and Knee Multi-Attribute Priority Tool. Cost-utility analyses will be undertaken using administrative records and self-report data. A subgroup of 100 participants will undergo qualitative interviews to explore the broader potential impacts of the ASMP.

Discussion: Using an innovative design combining both quantitative and qualitative components, this project will provide high quality data to facilitate evidence-based recommendations regarding the ASMP.

Getting your life back on track after stroke : a phase II multi-centered, single-blind, randomized, controlled trial of the Stroke Self-Management Program vs. the Stanford Chronic Condition Self-Management Program or standard care in stroke survivors

Background Self-management is seen as a primary mechanism to support the optimization of care for people with chronic diseases such as symptomatic vascular disease. There are no established and evidence-based stroke-specific chronic disease self-management programs. Our aim is to evaluate whether a stroke-specific program is safe and feasible as part of a Phase II randomized-controlled clinical trial.
Methods Stroke survivors are recruited from a variety of sources including: hospital stroke services, local paper advertisements, Stroke South Australia newsletter (volunteer peer support organization), Divisions of General Practice, and community service providers across Adelaide, South Australia. Subjects are invited to participate in a multi-center, single-blind, randomized, controlled trial. Eligible participants are randomized to either;
• standard care,
• standard care plus a six week generic chronic condition self-management group education program, or,
• standard care plus an eight week stroke specific self-management education group program.
Interventions are conducted after discharge from hospital. Participants are assessed at baseline, immediate post intervention and six months.
Study Outcomes The primary outcome measures determine study feasibility and safety, measuring, recruitment, participation, compliance and adverse events.
Secondary outcomes include:
• positive and active engagement in life measured by the Health Education Impact Questionnaire,
• improvements in quality of life measured by the Assessment of Quality of Life instrument,
• improvements in mood measured by the Irritability, Depression and Anxiety Scale,
• health resource utilization measured by a participant held diary and safety.

Conclusion The results of this study will determine whether a definitive Phase III efficacy trial is justified.