Geisinger Medical Center

We propose to harness the collective knowledge of healthcare providers by developing software which can crowd-source diagnostic image validation for machine learning

Shareable draft

Check out Geisinger Medical Center on @MedStartr Today We propose to harness the collective knowledge of healthcare providers by developing software which can crowd-source diagnostic image validation for machine learning. #MedMo #MedStartrVentures

https://www.medstartr.ai/s/avB18ePNW6E

Get Involved in Geisinger Medical Center

How We Help

Company Info

Clinical & IP

🔬

Clinical Validation

Numerous studies have documented the interpersonal as well as intrapersonal variation that occurs within diagnostic imaging.  For Radiology, variation exists within every type of imaging modality. A Pubmed search for “inter-observer+ variation + radiology” reveals 208 review articles. Of these, the following highlight key issues in variation: <ol> <li>Interpretive Error in Radiology is a good summary of the inherent levels of variation in radiologic images <ol> <li><< <a href="http://www.ajronline.org/doi/full/10.2214/AJR.16.16963">http://www.ajronline.org/doi/full/10.2214/AJR.16.16963</a>>></li> </ol> </li> <li>Evaluation of the effect of double reporting on test accruacy in screening and diagnostic imaging studies: A review of the evidence  demonstrates the importance of double reporting and how it can improve accuracy <ol> <li><< <a href="http://onlinelibrary.wiley.com/doi/10.1111/1754-9485.12450/full">http://onlinelibrary.wiley.com/doi/10.1111/1754-9485.12450/full</a>>></li> </ol> </li> <li>Current artefacts in cardiac and chest magnetic resonance imaging: tips and tricks demonstrates how artefacts can lead to perceptual errors and variation <ol> <li><< <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5258161/">https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5258161/</a>>></li> </ol> </li> </ol> Digital Pathology is a newer, emerging imaging modality, and thus there are less studies of observer variation. However the following contain important notes on perceptual error and how it can lead to observation variation <ol> <li>Measuring Errors in Surgical Pathology in Real Life Practice: Defining what does and doesn’t matter helps provide a framework on how to quantiate errors and quality in surgical pathology. By better understanding how errors in interpretation can occur, it can also help refine sources of variation <ol> <li><< <a href="https://academic.oup.com/ajcp/article-lookup/doi/10.1309/5KF89P63F4F6EUHB">https://academic.oup.com/ajcp/article-lookup/doi/10.1309/5KF89P63F4F6EUHB</a>>></li> </ol> </li> <li>Judging Mistakes in Pathology- Res Ipse Non Loquitur  proposes to classify errors as a subtype of interobserver variability, in that an individual pathologist is going against group consensus. Thus error can be used as a marker for variability. <ol> <li>< <a href="https://www.ncbi.nlm.nih.gov/pubmed/17090182">https://www.ncbi.nlm.nih.gov/pubmed/17090182</a>></li> </ol> </li> <li>The following are studies which demonstrate how interobserver variability can affect diagnosis<a href="#_msocom_1" id="_anchor_1" name="_msoanchor_1">[LB1]</a>  <ol> <li><a href="https://www.ncbi.nlm.nih.gov/pubmed/28664936">Mod Pathol.</a> 2017 Oct;30(10):1411-1421. doi: 10.1038/modpathol.2017.59. Epub 2017 Jun 30 PMID:28664936</li> <li><a href="https://www.ncbi.nlm.nih.gov/pubmed/27522597">Tech Coloproctol.</a> 2016 Sep;20(9):647-52. doi: 10.1007/s10151-016-1513-8. Epub 2016 Aug 13. PMID: 27522597</li> <li>Mod Pathol. 2016 Aug;29(8):879-92. doi: 10.1038/modpathol.2016.86. Epub 2016 May 13. PMID: 27174588</li> </ol> </li> </ol> Computer Aided Diagnosis (CAD) has the potential to act as an “second set of eyes”  for clinicians. It can offer a double check of the image to ensure nothing is missed. Under many current CAD workflows, the clinicians review the image first. The CAD then offers its interpretation, allowing the clinician to review regions of interest they may have missed.  The benefit of CAD can be seen here in the following:  <ol> <li>Computer-aided diagnosis in medical imaging: Historical review, current status and future potential provides an overview of how computer aided diagnosis can improve clinical workflow for radiology and pathology <ol> <li><< <a href="http://www.sciencedirect.com/science/article/pii/S0895611107000262">http://www.sciencedirect.com/science/article/pii/S0895611107000262</a>>></li> </ol> </li> <li>Both Improvement in radiologists’ detection of clustered microcalcifications on mammograms: The potential of computer aided diagnosis  and Improving breast cancer diagnosis with computer-aided diagnosis  provide specific examples of how CAD and improve diagnostic accuracy. <ol> <li><a href="http://www.sciencedirect.com/science/article/pii/S1076633299800580">http://www.sciencedirect.com/science/article/pii/S1076633299800580</a></li> <li><a href="http://journals.lww.com/investigativeradiology/Abstract/1990/10000/Improvement_in_Radiologists__Detection_of.6.aspx">http://journals.lww.com/investigativeradiology/Abstract/1990/10000/Improvement_in_Radiologists__Detection_of.6.aspx</a></li> </ol> </li> <li>The following summarize the growing importance of CAD for Pathology <ol> <li><< <a href="https://digitalpathologyassociation.org/blog/computer-aided-diagnosis-the-tipping-point-for-digital-pathology/">https://digitalpathologyassociation.org/blog/computer-aided-diagnosis-the-tipping-point-for-digital-pathology/</a>>></li> <li><<a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3132993/">https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3132993/</a>>></li> <li><a href="https://www.ncbi.nlm.nih.gov/pubmed/27283833">Breast Cancer Res Treat.</a> 2016 Jul;158(1):11-9. doi: 10.1007/s10549-016-3852-6. Epub 2016 Jun 9. PMID:27283833</li> <li><a href="https://www.ncbi.nlm.nih.gov/pubmed/26685085">Am J Surg Pathol.</a> 2016 Apr;40(4):569-76. doi: 10.1097/PAS.0000000000000574. PMID: 26685085</li> <li><a href="https://www.ncbi.nlm.nih.gov/pubmed/24843821">J Pathol Inform.</a> 2014 Mar 28;5(1):9. doi: 10.4103/2153-3539.129442. eCollection 2014 PMID: 24843821</li> <li>Arch Pathol Lab Med. 2011 Aug;135(8):1010-6. doi: 10.5858/2010-0462-OAR. PMID: 21809992</li> </ol> </li> </ol> The problem is that in order to develop robust and accurate CAD algorithms, the machine learning algorithms need well annotated training sets to learn from. These are often created by small groups of SMEs, and in some cases single dedicated researchers. Given the well documented variations within diagnostic observation and the large number of  cases needed to create a robust training set, it becomes obvious that the validity of the training set and thus the entire CAD algorithm is entirely dependent on the skill and expertise of individual SMEs using current methods.

MedStartr Engagement Score

The MedStartr Index is a reflection of how the crowds that come to MedStartr everyday to find healthcare innovations reacted. Calculated by our AI agent Fyggy based on algorithms we have been building since 2012 and over 100 different factors, the math is proprietary and complex. Suffice to say, it has been found to be highly correlative with early stage company growth. What you see here is the public version. To learn more or to get deeper access to our predictive data on startup success for your fund or company, click here.

Comments

Comments from new members are shown once the company's team has approved them. Please follow the community rules.

Be the first to comment.

What do you think of Geisinger Medical Center?

A HUGE part of what early stage companies, and even late stage ones, need and get on MedStartr is honest feedback from actual patients, clinicians, hospital leaders and industry insiders. It is very easy to get tunnel vision as a founder. Please tell the team that runs this company what you think.

Clinical Benefit Potential

The potential for a project to have a large and significant impact on medical care, treatment, or prevention.

Feasibility

The likelihood that this will provide benefit for patients / care givers and that the founders can execute on their vision and plan.

Innovativeness

The distinguishing features, concept, or technology that differentiates the entry compared to others that already exist and are in widespread use.

Fundability

The reviewers' beliefs that the idea is fundable or can make it to sustainability without outside investment.

Scalability

Some great ideas are destined to failure because they are too complex or too expensive. The reviewer should consider the likelihood that a project as explained, or perhaps with some tweaks, could possibly become a worldwide solution within a reasonable time frame.

Presentation Quality

How well the team presented the idea.

Submit anonymously — my name won't be shown to the team or on this profile

Your scores and comments are shown with your review on this company's profile.

Get Involved in Geisinger Medical Center