Interactive search and exploration of waveform data with searchlight

Abstract

Searchlight [9] enables search and exploration of large, multidimensional data sets interactively. It allows users to explore by specifying rich constraints for the "objects" they are interested in identifying. Constraints can express a variety of properties, including a shape of the object (e.g., a wave-form interval of length 10-100ms), its aggregate properties (e.g., the average amplitude of the signal over the interval is greater than 10), and similarity to another object (e.g., the distance between the interval's waveform and the query waveform is less than 5). Searchlight allows users to specify an arbitrary number of such constraints, with mixing different types of constraints in the same query. Searchlight enhances the query execution engine of an array DBMS (currently SciDB) with the ability to perform sophisticated search using the power of Constraint Programming (CP). This allows an existing CP solver from Or-Tools (an open-source suite of operations research tools from Google) to directly access data inside the DBMS without the need to extract and transform it. This demo will illustrate the rich search and exploration capabilities of Searchlight, and its innovative technical features, by using the real-world MIMIC II data set, which contains waveform data for multi-parameter recordings of ICU patients, such as ABP (Arterial Blood Pressure) and ECG (electrocardiogram). Users will be able to search for interesting waveform intervals by specifying aggregate properties of the corresponding signals. In addition, they will be able to search for intervals similar to already found, where similarity is defined as a distance between the signal sequences.