SBIR-STTR Award

This SBIR application seeks to commercialize the innovative Adaptive Multiple Feature Method (AMFM) tissue characterization technique that has been developed as part of NIH sponsored lung imaging research at the University of Iowa. Two-dimensional AMFM has shown great promise in lowering inter and intra-observer variance in the characterization, localization and staging of parenchymal lung disease. We expect three-dimensional AMFM, enabled by the growth of workstation processing power and the proliferation of MDCT, to be of tremendous importance to both the clinical treatment of COPD patients as well as the development and testing of new COPD therapies. Specific Aims: Aim 1: Develop 3D texture features to describe pulmonary parenchyma patterns in MDCT image data and develop a 3D adaptive multiple feature method (AMFM) for pulmonary disease pattern classification. Aim 2: Construct an expert-defined independent standard for assessment of pulmonary parenchyma classification performance. Aim 3: Show feasibility of a 3D AMFM parenchyma classification approach and demonstrate its improved performance compared with the existing 2D AMFM parenchyma classification approach. The long-term strategy for developing this technology is to spur the development of quantitative CT benchmarks for characterizing and staging degenerative lung disease such as emphysema, pulmonary fibrosis and sarcoidosis. Developing AMFM into a screening tool to complement lung cancer screening for at-risk patients is another option we are exploring for the long-term.

Public Health Relevance:
This Public Health Relevance is not available.

Thesaurus Terms:
Computed Axial Tomography, Computer Assisted Diagnosis, Diagnosis Quality /Standard, Lung Imaging /Visualization /Scanning, Respiratory Disorder Diagnosis, Technology /Technique Development, Three Dimensional Imaging /Topography Computer Program /Software, Disease /Disorder Classification, Emphysema, Image Processing, Interstitial Lung Disease, Pulmonary Fibrosis /Granuloma, Sarcoidosis Bioimaging /Biomedical Imaging, Human Data

This SBIR application seeks to commercialize the innovative Adaptive Multiple Feature Method (AMFM) tissue characterization technique that has been developed as part of NIH sponsored lung imaging research at the University of Iowa. Two-dimensional AMFM has shown great promise in lowering inter and intra-observer variance in the characterization, localization and staging of parenchymal lung disease. We expect three-dimensional AMFM, enabled by the growth of workstation processing power and the proliferation of MDCT, to be of tremendous importance to both the clinical treatment of COPD patients as well as the development and testing of new COPD therapies. Specific Aims: Aim 1: Develop 3D texture features to describe pulmonary parenchyma patterns in MDCT image data and develop a 3D adaptive multiple feature method (AMFM) for pulmonary disease pattern classification. Aim 2: Construct an expert-defined independent standard for assessment of pulmonary parenchyma classification performance. Aim 3: Show feasibility of a 3D AMFM parenchyma classification approach and demonstrate its improved performance compared with the existing 2D AMFM parenchyma classification approach. The long-term strategy for developing this technology is to spur the development of quantitative CT benchmarks for characterizing and staging degenerative lung disease such as emphysema, pulmonary fibrosis and sarcoidosis. Developing AMFM into a screening tool to complement lung cancer screening for at-risk patients is another option we are exploring for the long-term.

Public Health Relevance:
This Public Health Relevance is not available.

Thesaurus Terms:
Computed Axial Tomography, Computer Assisted Diagnosis, Diagnosis Quality /Standard, Lung Imaging /Visualization /Scanning, Respiratory Disorder Diagnosis, Technology /Technique Development, Three Dimensional Imaging /Topography Computer Program /Software, Disease /Disorder Classification, Emphysema, Image Processing, Interstitial Lung Disease, Pulmonary Fibrosis /Granuloma, Sarcoidosis Bioimaging /Biomedical Imaging, Human Data