Scholars Central

Conference abstractPublished 27 Apr 2015

Development and evaluation of new quantitative imaging biomarkers for improving cancer screening efficacy

Authors: Bin Zheng

Ask about this article

Abstract

Select any words in the title or abstract, or press this and choose a sentence.

Although cancer screening is considered an effective approach to detect early cancers and reduce cancer mortality rates, the high false-positive recall rates substantially reduce the efficacy of current population-based cancer screening. Using conventional computer-aided detection (CAD) schemes also substantially increases false-positive recalls. In order to better address this difficult issue, we recently investigated several new quantitative image features or biomarkers aiming to assist improving efficacy of cancer screening. In this presentation the author will review and summarize our new approaches and recently experimental results when applying this new concept to breast cancer screening using either FFDM or DCE-MRI examination modalities. Unlike the previous lesion-based CAD schemes that primarily focused on detecting more positive lesions that may be missed or overlooked by radiologists with higher false-positives, we used lesion-free and multi-image based CAD approaches that only focus on the detection and analysis of global image features to predict near-term cancer risk and/or classify between the true- and false-positive recalls (or cases). For example, (1) using new CAD to analyze bilateral mammographic tissue density asymmetry, the computed odds ratios showed an increasing risk trend from 1.00 to 9.07 between positive and negative case groups. (2) Analyzing the global MR kinetic image features, a CAD scheme achieved an area under ROC curve of AUC=0.839. (3) Combing the global image features from four-view mammograms, CAD yielded AUC=0.793 in classifying 1052 recalled cases. These preliminary studies demonstrated feasibility of applying a new quantitative image analysis approach to potentially help establish more effective personalized cancer screening paradigm.

Ask about this article

Answers come only from this article's title, abstract and keywords, each sentence pointing to where it comes from. Your question is sent to OpenAI to write the answer. Scholars Central does not store it.

Try:

Conferences in this field

Upcoming conferences in this field.

Follow

New articles from what you follow go into your email digest, once you agree to alert emails.

  • JournalJournal of Bioengineering & Biomedical Science
  • SubjectBiochemistry & Biotechnology
  • BylineArticles whose byline prints “Bin Zheng”