Nicolas Jay

Analysis of social communities with iceberg and stability-based concept lattices

In this paper, we presents a research work based on formal concept analysis and interest measures associated with formal concepts. This work focuses on the ability of concept lattices to discover and represent special groups of individuals, called social communities. Concept lattices are very useful for the task of knowledge discovery in databases, but they are hard to analyze when their size become too large. We rely on concept stability and support measures to reduce the size of large concept lattices. We propose an example from real medical use cases and we discuss the meaning and the interest of concept stability for extracting and explaining social communities within a healthcare network.

A data mining approach for grouping and analyzing trajectories of care using claim data: the example of breast cancer.

BackgroundWith the increasing burden of chronic diseases, analyzing and understanding trajectories of care is essential for efficient planning and fair allocation of resources. We propose an approach based on mining claim data to support the exploration of trajectories of care.MethodsA clustering of trajectories of care for breast cancer was performed with Formal Concept Analysis. We exported Data from the French national casemix system, covering all inpatient admissions in the country. Patients admitted for breast cancer surgery in 2009 were selected and their trajectory of care was recomposed with all hospitalizations occuring within one year after surgery. The main diagnoses of hospitalizations were used to produce morbidity profiles. Cumulative hospital costs were computed for each profile.Results57,552 patients were automatically grouped into 19 classes. The resulting profiles were clinically meaningful and economically relevant. The mean cost per trajectory was 9,600€. Severe conditions were generally associated with higher costs. The lowest costs (6,957€) were observed for patients with in situ carcinoma of the breast, the highest for patients hospitalized for palliative care (26,139€).ConclusionsFormal Concept Analysis can be applied on claim data to produce an automatic classification of care trajectories. This flexible approach takes advantages of routinely collected data and can be used to setup cost-of-illness studies.

Using formal concept analysis for mining and interpreting patient flows within a healthcare network

This paper presents an original experiment based on frequent itemset search and lattice based classification. This work focuses on the ability of iceberg-lattices to discover and represent flows of patient within a healthcare network. We give examples of analysis of real medical data showing how Formal Concept Analysis techniques can be helpful in the interpretation step of the knowledge discovery in databases process. This combined approach has been successfully used to assist public health managers in designing healthcare networks and planning medical resources.

On mining complex sequential data by means of FCA and pattern structures

Nowadays data-sets are available in very complex and heterogeneous ways. Mining of such data collections is essential to support many real-world applications ranging from healthcare to marketing. In this work, we focus on the analysis of “complex” sequential data by means of interesting sequential patterns. We approach the problem using the elegant mathematical framework of formal concept analysis and its extension based on “pattern structures”. Pattern structures are used for mining complex data (such as sequences or graphs) and are based on a subsumption operation, which in our case is defined with respect to the partial order on sequences. We show how pattern structures along with projections (i.e. a data reduction of sequential structures) are able to enumerate more meaningful patterns and increase the computing efficiency of the approach. Finally, we show the applicability of the presented method for discovering and analysing interesting patient patterns from a French healthcare data-set on cancer. The quantitative and qualitative results (with annotations and analysis from a physician) are reported in this use-case which is the main motivation for this work.

On measuring similarity for sequences of itemsets

Computing the similarity between sequences is a very important challenge for many different data mining tasks. There is a plethora of similarity measures for sequences in the literature, most of them being designed for sequences of items. In this work, we study the problem of measuring the similarity between sequences of itemsets. We focus on the notion of common subsequences as a way to measure similarity between a pair of sequences composed of a list of itemsets. We present new combinatorial results for efficiently counting distinct and common subsequences. These theoretical results are the cornerstone of an effective dynamic programming approach to deal with this problem. In addition, we propose an approximate method to speed up the computation process for long sequences. We have applied our method to various data sets: healthcare trajectories, online handwritten characters and synthetic data. Our results confirm that our measure of similarity produces competitive scores and indicate that our method is relevant for large scale sequential data analysis.

An Approach for Mining Care Trajectories for Chronic Diseases

With the increasing burden of chronic illnesses, administrative health care databases hold valuable information that could be used to monitor and assess the processes shaping the trajectory of care of chronic patients. In this context, temporal data mining methods are promising tools, though lacking flexibility in addressing the complex nature of medical events. Here, we present a new algorithm able to extract patient trajectory patterns with different levels of granularity by relying on external taxonomies. We show the interest of our approach with the analysis of trajectories of care for colorectal cancer using data from the French casemix information system.

Healthcare trajectory mining by combining multidimensional component and itemsets

Sequential pattern mining is aimed at extracting correlations among temporal data. Many different methods were proposed to either enumerate sequences of set valued data (i.e., itemsets) or sequences containing multidimensional items. However, in real-world scenarios, data sequences are described as events of both multidimensional items and set valued information. These rich heterogeneous descriptions cannot be exploited by traditional approaches. For example, in healthcare domain, hospitalizations are defined as sequences of multi-dimensional attributes (e.g. Hospital or Diagnosis) associated with two sets, set of medical procedures (e.g.

Non-redundant association rules between diseases and medications: an automated method for knowledge base construction

\lbrace

Mining heterogeneous multidimensional sequential patterns

Radiography, Appendectomy

A contribution to the discovery of multidimensional patterns in healthcare trajectories

\rbrace