SOLICITATION NOTICE
R -- Uterine Cervix Images and Spine X-ray Classification, and Medical Validation
- Notice Date
- 7/7/2005
- Notice Type
- Solicitation Notice
- NAICS
- 519190
— All Other Information Services
- Contracting Office
- Department of Health and Human Services, National Institutes of Health, National Library of Medicine, 8600 Rockville Pike, Bethesda, MD, 20894
- ZIP Code
- 20894
- Solicitation Number
- 05-142-CYC
- Response Due
- 7/22/2005
- Archive Date
- 8/6/2005
- Description
- It is the intent of the National Library of Medicine (NLM) to procure professional services from The Curators of the University of Missouri, Rolla, Missouri (Dr. R. Joe Stanley) on a sole source basis. Period of Performance: September 1, 2005 - August 31, 2006. Professional services are required over a 12-month period for the analysis, design and development of techniques for (1) illumination normalization of uterine cervix images, (2) unifying and extending spine x-ray classification work to make it independent of shape resolution, and (3) medical validation of segmented vertebral shapes. The Communications Engineering Branch (CEB) of the National Library of Medicine conducts research and development into the archiving and retrieval of biomedical multimedia information using technology such as Content-Based Image Retrieval (CBIR), which allows the retrieval of images by features such as shape, color, and texture of regions of interest within the images. Normalization of illumination across images is a prerequisite for effective retrieval for images with color and texture features. Digitized uterine cervix images are available in pairs for each visit of the survey participant. Many images exhibit uneven or high dynamic range in illumination across the image. In some cases pertinent pathology is in shadow. It is necessary to normalize the images for these conditions as well as automatically lighten the shadows where possible. NLM also maintains digitized spine x-ray images from second National Health and Nutrition Examination Survey (NHANES II). Prior work on classifying segmented vertebral boundaries for pathology needs to be unified and extended to allow shapes of different resolutions. Varying degree of detail is available at these resolutions and can help improve image retrieval experience. Modified code should also be compatible with current network oriented CBIR system being developed at NLM. The work will also build an extended reference set of segmented vertebrae from the digitized images by collecting from at least two medical experts' segmentations of vertebrae for 5000 shapes from approximately 1000 images of the digitized spine, according to specifications and methodology negotiated with the government, including both cervical and lumbar spine images from the NHANES II data set. Task description: Subtask 1: Implement an extended capability of illumination and color normalization on uterine cervix images.The investigator shall continue the illumination normalization work previously done by (a) extending the algorithm capability to normalize color and luminance across image pairs as well as across the collection to improve retrieval quality; and (b) lighten shadow regions to (i) improve feature computation and (ii) assist in visual analysis. It may be necessary to share results from this work with other Government contractors. If such a need arises, the Government shall specify the appropriate interface formats. In addition, for the core task, the Government shall provide images, information about computed features, and any other guidance as necessary. Subtask 2: Extend capability of classification algorithm for digitized spine x-ray images. The investigator shall update earlier work on classification of segmented vertebral shapes for presence / absence of anterior osteophytes, disc space narrowing, and subluxation and spondylolisthesis to include shapes of different resolutions. NLM currently uses a variety of resolutions for vertebral shape data ranging from 9 point coarse boundary outlines to more than 400 boundary points. Classification algorithms developed earlier do not support this variety and are not suitable for use in a network oriented CBIR system. Proposed task will include updating the algorithms and making it usable in a system setting. The Government shall provide necessary information to make it suitable to interface with the system. Subtask 3: Collect spine x-ray reference segmentations. The investigator shall, through the involvement of board certified medical experts, build an extended reference set of segmented vertebrae from the digitized images. The Government shall provide the investigator with segmented boundary shapes. The investigator shall collect from at least two board certified medical experts, reviews on segmentations of vertebrae in approximately 1000 images of the digitized spine, according to specifications and methodology negotiated with the government. The reviews will include validation with corrections, if necessary, of existing segmented boundaries, 9-point landmark markup validation, and pathology detail (anterior osteophytes, disc space narrowing, subluxation and/or spondylolisthesis, and other observations, if any) for 5000 vertebrae from images be provided by the Government. Both cervical and lumbar spine images and segmentations will be included. The protocol for segmentation review and the software will be provided by the Government. It is expected that the medical experts will have available to them the necessary hardware (512MB RAM, late model PC running Windows XP SP2) for the task. Evaluation Criteria: (1) The investigator shall have at least 5 years experience as a researcher in advanced technical applications for biomedical processing, and shall have demonstrable expertise (by conference and technical journal publications) in the areas of advanced image processing, including biomedical image segmentation; artificial neural networks and classification methods, and the application of such methods to the classification of biomedical pathology in human subjects; and hands-on software skills in MATLAB and C/C++. (2) The investigator shall have demonstrable expertise in the application of advanced image segmentation and classification techniques to the NHANES II digitized x-rays. (3) The investigator shall have access to at least two board certified medical experts in the field related to the topic of research.. The University of Missouri-Rolla is uniquely qualified for this work because of the expertise and experience of the principal investigator below (Dr. Joe Stanley) in the specific area of feature classification of digitized spine x-rays [1-4] and prior published work in color normalization for medical images [5-9], combined with his extensive educational background and experience in advanced image processing. Principal Investigator: Dr. R. Joe Stanley is on the faculty of the Department of Electrical and Computer Engineering at the University of Missouri-Rolla. Dr. Stanley is uniquely qualified for this work, based on his professional training and over 6 years experience in both academia and the private sector. Dr. Stanley was principal investigator for the image recognition program at Systems & Electronics, Inc. (SEI) in St. Louis, Missouri from May 1999 - July 2000. As principal investigator, he oversaw research and software development for postal and medical imaging-based systems, including systems to perform semi-automated human white blood cell differential counting and automated pill recognition. Dr. Stanley's work resulted in two patent applications. He completed his doctoral study at the University of Missouri-Columbia (UMC) under fellowships from the National Library of Medicine and the National Cancer Institute in a medical informatics training program. While at UMC, he received first and second place awards in student paper competitions at national conferences in 1998 and 1996, respectively, for his research in automated human chromosome image analysis. He also received the Outstanding Graduate Student Researcher Award from the Department of Computer Engineering and Computer Science at UMC in 1997. Dr. Stanley has demonstrated experience in successfully collaborating with medical experts. Specifically, in prior research projects (human chromosome image analysis and white blood cell image analysis), collaboration with a medical expert, i.e. pathologist, was used for data acquisition, data interpretation and technical expertise in algorithm development. Presently, Dr. Stanley is investigating image processing and pattern recognition techniques in three research projects: (1) Image Processing and High-Throughput Microarray Data. Dr. Stanley serves as the principal investigator for this research project. His role is to develop algorithms for image registration and feature extraction to evaluate paired grayscale microarray images; (2) Multi-Modal NDE Development for Corrosion Detection and Analysis. This project builds on a number of recent SBIR programs as well as the Enhanced Equipment for Material Thinning (TCORR) and Structural Repair of Aging Aircraft (SRAA) programs to develop a multi-modal approach for corrosion detection. Dr. Stanley serves as principal investigator for the data fusion component for this project. His role is to develop algorithms for fusing multiple sensor information sources, including microwave, eddy current, and ultrasound to improve the detection and characterization of corrosion in sample structures; (3) Dermatology Image Analysis. Dr. Stanley collaborates with Dr. Randy Moss and Dr. William V. Stoecker at the University of Missouri-Rolla in the development of image processing and pattern recognition techniques to detect malignant melanoma in dermatology images. Dr. Stanley's contributions include the development of techniques to locate and remove hair from dermatology images and to locate and segment key dermatology image features for the detection of melanoma. The proposed acquisition will be procured under FAR Part 13 - Acquisitions of Non- Commercial Items. This is not a Request for Quotations (RFQ), nor is an RFQ available. However, all responsive sources may submit a capabilities statement in a timely manner that will be considered by the Government. Sources interested in responding to this notice must be able to provide convincing evidence that they possess the requisite expertise and experience to successfully perform the services as specified above. Responses must be in writing and must be received in the office within fifteen (15) business days from the publication date of this notice. Proposals must include pricing information. NLM Synopsis No. NLM 05-142/CYC. Inquires regarding this procurement may be made to Cara Y. Calimano, Contract Specialist, NLM on (301) 496-6127.
- Place of Performance
- Address: 8600 Rockville Pike, Bethesda, Maryland
- Zip Code: 20894
- Country: USA
- Zip Code: 20894
- Record
- SN00842797-W 20050709/050707211829 (fbodaily.com)
- Source
-
FedBizOpps.gov Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |