Enhanced Data Sharing

Cancer Research Ideas - Enhanced Data Sharing (Archived)

Sharing data among investigators and institutions is essential for building on the cancer research progress made so far, but many barriers to effective sharing exist. Barriers can arise from technological or infrastructural limitations (data are stored on different systems or are generated by different methods, making them incompatible). Poor access or participation can also be a barrier (some data may not be shared, and some researchers may not be able to access or analyze even data that are publicly available).

The submission period for Enhanced Data Sharing ideas ended on July 1. However, we encourage you to sign up for regular e-mail updates about the National Cancer Moonshot Initiative at the Cancer Moonshot Milestones web page.

Enhanced Data Sharing

Web-based Oncologic Data and Literature Review Resource

Oncologic literature should be organized, reviewed, and rated through a web-based application, making important emerging data more widely available and impactful. A rating system would help clinicians prioritize data consumption and implementation.

Submitted by (@jsnider)

What is the research problem :

Oncologic literature and research are particularly poorly organized/disseminated to clinicians on the front lines of cancer care. There currently exists little guidance to direct oncologists towards the most practice changing developments.

How will your solution make a difference :

The breadth of established and emerging oncologic data remains overwhelming to many clinicians. A central, approachable resource with reliable reviews of oncologic practice/innovations could aid clinicians in vastly improving cancer care worldwide.

What is your proposed solution :

A nonprofit website/database and associated mobile application will be developed to raise awareness of and rank (in multiple domains) emerging manuscripts/data as to their import to the field of oncology. At least three separate rankings will be maintained – one from invited experts/editors; another from registered users/verified oncologists; another from patients. Accessibility, interactivity, and ease of use will be prioritized.

Voting

0 votes
Ideate

Enhanced Data Sharing

Use precision diagnostics to cross refer patients to best trials

Diagnostics tools are effective ways to identify patients for particular therapies. Unfortunately patients who are screened for a particular trial should be provided information on all trials which they may be eligable for.

Submitted by (@ekowack)

What is the research problem :

Companies may not be motivated to share patient data for competitive and/or ethics reasons. In addition, the administrative support for clinical trials may not respond quickly enough to process cross referrals.

How will your solution make a difference :

Patients will be matched with a trial which is best for them more quickly thereby possibly avoiding unnecessary treatments such as chemo or wasting valuable time on treatments not likely to work. It will help them make informed treatment decisions..

What is your proposed solution :

Learn from biotech companies such as Ignyta who are successfully using their Diagnostic platform to identify patients with certain oncodrivers (rare) to enroll patients on their trials. If the patient can be referred to other trials, they do it through the physician. Convene an expert panel to identify mechanisms for rapid data sharing with the patient (upon consent) so they may be cross referred to a trial which may be best for them.

Voting

0 votes
Ideate

Enhanced Data Sharing

Making User Friendly Data Mining Tools and Promoting Use

Improving tools to store and mine data is imperative to continue progress.

There is an unmet need for a universal database to access data and query tools.

Submitted by

What is the research problem :

There are many examples of unique and useful data query tools that present patient and experiment data. There is no universal website that hosts or promotes use of all of these technologies.

How will your solution make a difference :

A user-friendly resource that combines and directs the user to the data and tools would speed up progress. There are many storage and query tools available that I am not aware of. More to come! We need a grand resource in one place!

What is your proposed solution :

Advanced data mining tools are needed.

Many tools are hosted independently such as Oncomine, cBioPortal, transposable element tools etc. and do different things.

We need a resource that catalogs and directs all of the storage and query tools - ex: cancer-scientist.org, LINCS canvas browser, Genomic Data Commons etc..

Voting

0 votes
Ideate

Enhanced Data Sharing

Funding and Cultivating a Data Sharing Ethos

Funding for data sharing is small when compared to the overall funding budget and considering its necessity and impact to cancer care.The little funding that is rewarded has not led to extensibility for real world use and expertise gaps still remain.

Submitted by

What is the research problem :

Funding and cultivating talent has been hard.Building teams is not as simple as putting the best medical knowledge expert with the best "silicon valley" data analyst together.Often because of gaps in understanding and communication,such efforts fail.

How will your solution make a difference :

The lapses in interoperability is a reflection of gaps in funding and lack of "hybrid" level expertise who understand and roadmap solutions.Envisioning and architecting a pipeline of human capital with incentives with cultivate a data sharing ethos

What is your proposed solution :

The data sharing ethos should be composed of teams with "hybrids" who span a broad depth of expertise (medical and data science), and most important have communicative and open pipelines for dialogue.A culture of bringing up the next generation of folks with this medical and data science mindset will help drive future development.The share of funding should be proportionally increased to democratize great minds to be discovered rather than sought

Voting

0 votes
Ideate

Enhanced Data Sharing

Advancing patient-valued care through data sharing

To ensure that patients and clinicians are able to make fully informed decisions about pharmaceutical agents, biological products, and medical devices tested through clinical trial research, an era of open science through data sharing is necessary.

Submitted by

What is the research problem :

As advances in science and technology continue to bring disruptive changes, effective implementation of data sharing will prove far more difficult than implementation of the requirement to register clinical trials.

How will your solution make a difference :

As study populations increase to millions, the implementation of observational studies with novel statistical methods will allow for the discovery of useful, reproducible patterns and relationships from these data.

What is your proposed solution :

Scientists, physicians, and the public will mandate that all the primary data be made public, along with the analytical tools necessary to reanalyze, test, refine, and build on them. Data security will have to evolve and thereby win the public’s trust with new techniques that will do what now seems impossible: guarantee privacy protection while providing detailed information about each person.

Voting

0 votes
Ideate

Enhanced Data Sharing

Cloud based imaging for oncology

We propose usage of our cloud-based deep learning platform to aggregate images from around the globe and then automate and standardize segmentation and measurement of these images

Submitted by (@carla0)

What is the research problem :

Today, clinical data is managed on-site with limited computation and is not leveraged to benefit all physicians and patients. We need to amass and analyze data from medical imaging applications to generate better clinical outcomes at a lower cost.

How will your solution make a difference :

We aggregate imaging data and learn from it, then automate accurate and repeatable segmentation and measurement. Advanced imaging will be more widely available as we provide access to powerful computation and automated, accurate measurements

What is your proposed solution :

Combining deep learning with cloud computing allows for larger images and more advanced analysis from data collected from sources worldwide. We leverage AI to automatically segment, categorize and measure anatomy and physiology. We’ve proven this with our 4D flow application, and we’re building modules to identify and track cancerous lesions, thus reducing tedious reconstruction, post-processing and contour creation.

Voting

0 votes
Ideate

Enhanced Data Sharing

Make data management plans versioned, machine actionable, public

Before we know what data to share and how, we have to know what has already been generated or is in the process of being so. Data management plans help with that, but they would be more effective if they were versioned, machine actionable and public.

Submitted by (@daniel.mietchen)

What is the research problem :

We do not know who is researching what at a given moment:

- what data have already been gathered, and where they are;

- what is in the pipelines;

- who is addressing the same issues or using the same methods.

 

This is inefficient on a systemic level.

How will your solution make a difference :

Research would become more collaborative, reproducible and sustainable:

- data reusers could start working while data are still being gathered;

- some data are never published; with their DMPs public, we would at least know they exist and who to ask…

What is your proposed solution :

Share data management plans (DMPs), which are documents that lay out planned research.

Make them

- updatable, so as to more closely reflect ongoing research;

- machine actionable, so as to provide a basis for integrating them into automated tools and workflows, e.g. to notify data repositories or potential reusers;

- public, so as to help people find out about each other's work and to initiate reuse or collaboration earlier in the research cycle.

Voting

0 votes
Ideate

Enhanced Data Sharing

Presentation of patient status for learning health in oncology

Modern presentation of patient condition including physician impression and patient reported measures will both quantify and communicate patient status between care providers supporting records for data driven decision support.

Submitted by (@tmcnutt1)

What is the research problem :

Big data in medicine suffers from the lack of quantifiable data from clinical visits and longitudinal outcomes. Decision support systems must balance the need for structured information with physician impression and patient influences.

How will your solution make a difference :

Physicians will be more efficient at documenting patient condition and communicating through graphical means while also growing big data for inclusion in learning health systems for improved clinical decision making and future data science research.

What is your proposed solution :

Modern human computer interface technologies tailored to oncology bring unobtrusive measures of patient condition and quality of life with high longitudinal resolution in and out of the clinical setting. Graphical presentation of the patient status within existing imaging and treatment related views can communicate, quantify and record the information that physician’s actually use when making treatment and symptom management decisions.

Voting

0 votes
Ideate

Enhanced Data Sharing

Allow patients to own their own specimens

Acknowledge patients are owners of their own bodily specimens, allow them to access those specimens as needed for testing and diagnosis, and let them decide with whom to share their specimens

Submitted by (@janetfreemandaily)

What is the research problem :

Some academic research centers and other research facilities act as if they own tissues collected from patients. Sometime patients are unable to access their own tissues to enable testing or clinical trial participation at other facilities.

How will your solution make a difference :

More patients will be willing to participate in trials because they will have access to their tissues when needed to testing to pursue an FDA-approved treatment or clinical trial, or change clinical trials when a treatment is no longer working.

What is your proposed solution :

Revise the common rule to acknowledge that patients own their own specimens. Allow patients to decide who will have access to their specimens (especially rare disease tissues) for research purposes. Acknowledge that patients are entitled to learn results of genomic or other testing conducted on their tissues -- the results could have present or future health implications for them or their families.

Voting

0 votes
Ideate

Enhanced Data Sharing

Restructure research system to emphasize open source data

To accelerate cancer research, we need to break down data silos and permit open access to data. Instead of rewarding researchers for being first to gather AND report on data, reward them for generating high-quality data, as well as analyzing data

Submitted by (@janetfreemandaily)

What is the research problem :

Data is kept in research silos because research career advancement depends on publishing unique, positive research results.

How will your solution make a difference :

Researchers can still pursue studies of interest, but valuable data will be shared with all, not locked away in silos. Shared data accelerates research and enables innovation.

What is your proposed solution :

Restructure approach to research advancement. Funders require all research data be placed in open sourced databases. Researchers achieve recognition by (1) creating high-quality data, (2) analyzing data accurately, (3) verifying analysis done by others. All reports are open sourced online for critique. Journals could make money like media does now--by building reputation for accuracy and competing to report on most useful data or results.

Voting

0 votes
Ideate

Enhanced Data Sharing

Symptom/ICD-9 Code software for early detection of Gyn cancers

The CDC's Inside Knowledge Campaign is a very detailed awareness program helping women, physicians and caregivers to be more aware of the most common symptoms of gynecological cancers. An early detection software program will tag and link symptoms.

Submitted by (@servinghope)

What is the research problem :

Without the actual integration of symptom tracking into a larger patient management and billing database, there is no real way for a multidisciplinary approach to early detection.

How will your solution make a difference :

The software would search for this data and create the alert to the physicians working with the patient. Because gyn cancer symptoms are recognized cross varying scopes of care, the software will find what is missed otherwise, linking the dots.

What is your proposed solution :

Implementing the use of a software program that searches for ICD-9 codes reflecting the common symptoms related to all gynecological cancer will increase communication amongst PCP, OBGyn and gastroenterologists to better assess a possible risk for gynecological cancers.

 

This software would generate an alert once the billing system/patient record sees that any physician notes in a patient chart that the patient is experiencing a common symptom.

Voting

0 votes
Ideate