ArchiText: Interactive Hierarchical Topic Modeling.

Human-in-the-loop topic modeling allows users to explore and steer the process to produce better quality topics that align with their needs. When integrated into visual analytic systems, many existing automated topic modeling algorithms are given interactive parameters to allow users to tune or adjust them. However, this has limitations when the algorithms cannot be easily adapted to changes, and it is difficult to realize interactivity closely supported by underlying algorithms. Instead, we emphasize the concept of tight integration, which advocates for the need to co-develop interactive algorithms and interactive visual analytic systems in parallel to allow flexibility and scalability. In this article, we describe design goals for efficiently and effectively executing the concept of tight integration among computation, visualization, and interaction for hierarchical topic modeling of text data. We propose computational base operations for interactive tasks to achieve the design goals. To instantiate our concept, we present ArchiText, a prototype system for interactive hierarchical topic modeling, which offers fast, flexible, and algorithmically valid analysis via tight integration. Utilizing interactive hierarchical topic modeling, our technique lets users generate, explore, and flexibly steer hierarchical topics to discover more informed topics and their document memberships.

Lumos: a statewide linkage programme in Australia integrating general practice data to guide system redesign.

Objective: With ageing of the Australian population, more people are living longer and experiencing chronic or complex health conditions. The challenge is to have information that supports the integration of services across the continuum of settings and providers, to deliver person-centred, seamless, efficient and effective healthcare. However, in Australia, data are typically siloed within health settings, precluding a comprehensive view of patient journeys. Here, we describe the establishment of the Lumos programme-the first statewide linked data asset across primary care and other settings in Australia and evaluate its representativeness to the census population. Methods and analysis: Records extracted from general practices throughout New South Wales (NSW), Australia's most populous state, were linked to patient records from acute and other settings. Innovative privacy and security technologies were employed to facilitate ongoing and regular updates. The marginal demographic distributions of the Lumos cohort were compared with the NSW census population by calculating multiple measures of representation to evaluate its generalisability. Results: The first Lumos programme data extraction linked 1.3 million patients' general practice records to other NSW health system data. This represented 16% of the NSW population. The demographic distribution of patients in Lumos was >95% aligned to that of the NSW population in the calculated measures of representativeness. Conclusion: The Lumos programme delivers an enduring, regularly updated data resource, providing unique insights about statewide, cross-setting healthcare utilisation. General practice patients represented in the Lumos data asset are representative of the NSW population overall. Lumos data can reliably be used to identify at-risk regions and groups, to guide the planning and design of health services and to monitor their impact throughout NSW.

Using Dispensing Data to Evaluate Adherence Implementation Rates in Community Pharmacy.

Background: Medication non-adherence remains a significant problem for the health care system with clinical, humanistic and economic impact. Dispensing data is a valuable and commonly utilized measure due accessibility in electronic health data. The purpose of this study was to analyze the changes on adherence implementation rates before and after a community pharmacist intervention integrated in usual real life practice, incorporating big data analysis techniques to evaluate Proportion of Days Covered (PDC) from pharmacy dispensing data. Methods: Retrospective observational study. A de-identified database of dispensing data from 20,335 patients (n = 11,257 on rosuvastatin, n = 6,797 on irbesartan, and n = 2,281 on desvenlafaxine) was analyzed. Included patients received a pharmacist-led medication adherence intervention and had dispensing records before and after the intervention. As a measure of adherence implementation, PDC was utilized. Analysis of the database was performed using SQL and Python. Results: Three months after the pharmacist intervention there was an increase on average PDC from 50.2% (SD: 30.1) to 66.9% (SD: 29.9) for rosuvastatin, from 50.8% (SD: 30.3) to 68% (SD: 29.3) for irbesartan and from 47.3% (SD: 28.4) to 66.3% (SD: 27.3) for desvenlafaxine. These rates declined over 12 months to 62.1% (SD: 32.0) for rosuvastatin, to 62.4% (SD: 32.5) for irbesartan and to 58.1% (SD: 31.1) for desvenlafaxine. In terms of the proportion of adherent patients (PDC >= 80.0%) the trend was similar, increasing after the pharmacist intervention from overall 17.4 to 41.2% and decreasing after one year of analysis to 35.3%. Conclusion: Big database analysis techniques provided results on adherence implementation over 2 years of analysis. An increase in adherence rates was observed after the pharmacist intervention, followed by a gradual decrease over time. Enhancing the current intervention using an evidence-based approach and integrating big database analysis techniques to a real-time measurement of adherence could help community pharmacies improve and sustain medication adherence.

Pharmacist-led medication non-adherence intervention: reducing the economic burden placed on the Australian health care system.

Background: Scarcity of prospective medication non-adherence cost measurements for the Australian population with no directly measured estimates makes determining the burden medication non-adherence places on the Australian health care system difficult. This study aims to indirectly estimate the national cost of medication non-adherence in Australia comparing the cost prior to and following a community pharmacy-led intervention. Methods: Retrospective observational study. A de-identified database of dispensing data from 20,335 patients (n=11,257 on rosuvastatin, n=6,797 on irbesartan and n=2,281 on desvenlafaxine) was analyzed and average adherence rate determined through calculation of PDC. Included patients received a pharmacist-led medication adherence intervention and had twelve months dispensing records; six months before and six months after the intervention. The national cost estimate of medication non-adherence in hypertension, dyslipidemia and depression pre- and post-intervention was determined through utilization of disease prevalence and comorbidity, non-adherence rates and per patient disease-specific adherence-related costs. Results: The total national cost of medication non-adherence across three prevalent conditions, hypertension, dyslipidemia and depression was $10.4 billion equating to $517 per adult. Following enrollment in the pharmacist-led intervention medication non-adherence costs per adult decreased $95 saving the Australian health care system and patients $1.9 billion annually. Conclusion: In the absence of a directly measured national cost of medication non-adherence, this estimate demonstrates that pharmacists are ideally placed to improve patient adherence and reduce financial burden placed on the health care system due to non-adherence. Funding of medication adherence programs should be considered by policy and decision makers to ease the current burden and improve patient health outcomes moving forward.

Extracting unrecognized gene relationships from the biomedical literature via matrix factorizations.

BACKGROUND: The construction of literature-based networks of gene-gene interactions is one of the most important applications of text mining in bioinformatics. Extracting potential gene relationships from the biomedical literature may be helpful in building biological hypotheses that can be explored further experimentally. Recently, latent semantic indexing based on the singular value decomposition (LSI/SVD) has been applied to gene retrieval. However, the determination of the number of factors k used in the reduced rank matrix is still an open problem. RESULTS: In this paper, we introduce a way to incorporate a priori knowledge of gene relationships into LSI/SVD to determine the number of factors. We also explore the utility of the non-negative matrix factorization (NMF) to extract unrecognized gene relationships from the biomedical literature by taking advantage of known gene relationships. A gene retrieval method based on NMF (GR/NMF) showed comparable performance with LSI/SVD. CONCLUSION: Using known gene relationships of a given gene, we can determine the number of factors used in the reduced rank matrix and retrieve unrecognized genes related with the given gene by LSI/SVD or GR/NMF.

Hierarchical community detection via rank-2 symmetric nonnegative matrix factorization.

BACKGROUND: Community discovery is an important task for revealing structures in large networks. The massive size of contemporary social networks poses a tremendous challenge to the scalability of traditional graph clustering algorithms and the evaluation of discovered communities. METHODS: We propose a divide-and-conquer strategy to discover hierarchical community structure, nonoverlapping within each level. Our algorithm is based on the highly efficient rank-2 symmetric nonnegative matrix factorization. We solve several implementation challenges to boost its efficiency on modern computer architectures, specifically for very sparse adjacency matrices that represent a wide range of social networks. CONCLUSIONS: Empirical results have shown that our algorithm has competitive overall efficiency and leading performance in minimizing the average normalized cut, and that the nonoverlapping communities found by our algorithm recover the ground-truth communities better than state-of-the-art algorithms for overlapping community detection. In addition, we present a new dataset of the DBLP computer science bibliography network with richer meta-data and verifiable ground-truth knowledge, which can foster future research in community finding and interpretation of communities in large networks.