Ahmed Soliman

Graduate Trainee, University of Connecticut

1 active project

Patient Time-Series-Based Record Linkage

In this study we are interested in validating a novel method that could efficiently link de-identified patient records to their time-series data (e.g. ergometric tests). This improved linkage method should aid medical institutions (e.g. rehabilitation centers) to assess the quality…

Scientific Questions Being Studied

In this study we are interested in validating a novel method that could efficiently link de-identified patient records to their time-series data (e.g. ergometric tests). This improved linkage method should aid medical institutions (e.g. rehabilitation centers) to assess the quality of their recorded data without using any personally identifying information. This kind of data quality assessment has been demonstrated in a previous peer-reviewed research article. Although we have verified our novel method on synthetic data, we want to run our algorithm on real time-series data that belongs to de-identified patient records. This way our assessment will be more sound as we will show that our method works efficiently on real datasets.

Project Purpose(s)

  • Methods Development

Scientific Approaches

We plan to apply our novel method for patient record linkage which is based on time-series matching. By record linkage we mean identifying records (containing time-series data) pertinent to the same individual. Our Linking methods/algorithms generally employ hierarchical clustering algorithms. We also use a fast sorting algorithm to help eliminate identical records. Then, we construct a graph that links similar records. Finally, we find the connected components within such graph. Please see http://www.rlatools.com for more information about theses linking tools.

This study will use datasets of time-series data (e.g. heart rate) collected from ergo-meters or wearable devices, such as fitbits. Please note that this research will never use any external data sources. Also, we are interested in a diverse sample in general.

Anticipated Findings

This study is anticipated to present a novel linking method that is faster and more-efficient than what is available via the currently available tools. This novel method will enable medical institutions to assess the quality of their collected data without compromising the privacy of the patients.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Research Team

Owner:

Collaborators:

  • Nalini Ravishanker - Late Career Tenured Researcher, University of Connecticut
1 - 1 of 1
<
>
Request a Review of this Research Project

You can request that the All of Us Resource Access Board (RAB) review a research purpose description if you have concerns that this research project may stigmatize All of Us participants or violate the Data User Code of Conduct in some other way. To request a review, you must fill in a form, which you can access by selecting ‘request a review’ below.