Enhanced Peak Detection Algorithm

ABSTRACT

The Seeker is seeking an algorithm for accurate, reproducible, and automatic identification and characterization of peaks in two-dimensional time series data. The data may be simple and contain a small number of fully resolved peaks or may be complex and contain hundreds (to thousands) of overlapping peaks that may or may not be fully resolved with baselines that may or may not fluctuate across the timeline. Algorithms must identify the start, end, and apex of each peak.

This is a Reduction-to-Practice Challenge that requires written documentation, output from the proposed algorithm, and submission of source code for validation upon request.

OVERVIEW

Time-series data is ubiquitous in many technical fields. Typically, this data exhibits peaks and/or valleys that convey information via their position in time, height, width, total area, etc., and in-depth analysis of the data depends on identifying and quantifying such parameters. In the most straight-forward circumstances the peak or valley features are well-defined and separated from each other so identification and quantification of the relevant parameters for each feature is relatively simple. In more complex circumstances the features overlap to varying degrees and determination of the key parameters of the features becomes much more complicated. In addition, other complicating factors may include a varying baseline and noisy data. The Seeker is interested in an automated algorithm to accurately and reproducibly identify peaks and quantify relevant parameters of the peaks in time-series data with a varying baseline containing sharp peaks that may overlap. Solvers will be provided extensive training and test data sets for development of algorithms to address this Challenge.

A submission to the Challenge should include the following:

1. A detailed description of the proposed solution addressing specific Technical Requirements presented in the Detailed Description of the Challenge. This should also include a thorough description of the algorithm used in the solution accompanied by a well-articulated rationale for the method employed.
2. Output from the proposed algorithm applied to a test data set. Output must be in the form described in the Detailed Description of the Challenge.  Submissions will be ranked by the Seeker based on accuracy against the test data set.
3. For the top ranked submissions, the Seeker may request source code and/or an executable with sufficient documentation to enable the Seeker to compile, execute the algorithm, and validate the method using additional validation data sets.

The Challenge award is contingent upon theoretical evaluation of the method/algorithm by the Seeker, and validation by the Seeker of the submitted reduction to practice Solution.

To receive an award, the Solvers will not have to transfer their exclusive IP rights to the Seeker. Instead, Solvers will grant to the Seeker a non-exclusive license to practice their solutions. 

Submissions to this Challenge must be received by 11:59 PM (US Eastern Time) on 07-Mar-2022.
Late submissions will not be considered.