---------------------------------------------------
TREC 2018 Precision Medicine Track
---------------------------------------------------
Most work on precision medicine focuses on developing new treatments based on an individual's genetic, environmental, and lifestyle profile. The result is a data-driven approach investigating the best treatment for an individual patient. This promising approach has led to significant advances, including an explosion of scientific research, as embodied by the Precision Medicine Initiative (PMI). This presents an information problem for clinicians, however, as the vast literature available for precision medicine can make it difficult to find the most appropriate treatment for the clinician's current patient. The ability to quickly locate relevant information for a current patient using information retrieval (IR) has the potential to be an important tool for helping clinicians find the most up-to-date evidence-based treatment for their patients.
The 2018 TREC Precision Medicine track continues the prior 2017 Precision Medicine track, which was a specialization of the previous TREC Clinical Decision Support track. Specifically, the 2018 Precision Medicine track focuses on the case of providing clinical decision support to cancer patients with genetic variations that might impact the choice of treatment. The track uses synthetic patients developed by precision oncologists at the world-famous MD Anderson Cancer Center in Houston, TX. For each patient, participants are challenged with retrieving relevant scientific literature articles discussing potential treatments, as well as potential clinical trials for which the patient may be eligible.
--- Task Description ---
* Topics/Queries: Synthetic patients with cancer and one or more genetic variants
* Data Collection: (1) scientific abstracts, (2) clinical trials
We will be using synthetic cases created by precision oncologists at the University of Texas MD Anderson Cancer Center. Each case will describe the patient's disease (type of cancer), the relevant genetic variants (which genes), basic demographic information (age, sex), and other potential factors that may be relevant. The cases are semi-structured and require minimal natural language processing.
Participants of the track will be challenged with retrieving (1) biomedical articles, in the form of article abstracts (largely from MEDLINE/PubMed), addressing relevant treatments for the given patient, and (2) clinical trials (from ClinicalTrials.gov), addressing relevant clinical trials for which the patient is eligible. The first set of results represents the retrieval of existing knowledge in the scientific literature, while the second represents the potential for connecting patients with experimental treatments if existing treatments have been ineffective.
--- Important Dates ---
April 2018: Document collection available for download
May 2018: Topics available for download
August 2018: Results submission deadline
October 2018: Relevance judgements and individual evaluation scores released
Late October 2018: Initial system description papers due
November 14-16 2018: TREC 2018 conference at NIST
--- Organizing Committee ---
Kirk Roberts (UTHealth)
William Hersh (OHSU)
Dina Demner-Fushman (NLM)
Ellen Voorhees (NIST)
Alexandar Lazar (UT MD Anderson Cancer Center)
Shubham Pant (UT MD Anderson Cancer Center)
--- Contact ---