Jay A. Baker, MD

Related Content

Services

• Breast Cancer
• Radiology

Chief

Department / Division:
Radiology / Breast Imaging

Address:
DUMC 3808
Durham, NC 27710

Appointment Telephone:
(919) 684-7641

Office Telephone:
(919) 684-7645

Fax Telephone:
(919) 684-7125

• M.D., Duke University School of Medicine, North Carolina, 1992

• Diagnostic Radiology, Duke University Medical Center, North Carolina, 1992-97

• Breast Imaging, Duke University Medical Center, North Carolina, 1997-98

Clinical Interests:
All aspects of breast imaging including mammography, breast ultrasound, breast MRI, and needle biopsies; breast imaging research including digital breast tomosynthesis and computer aided diagnosis

Research Interests:
As a radiologist in the Division of Breast Imaging, I am interested in studying techniques to better detect and assess breast lesions that may represent breast cancer.  The major focus of my research activity includes both a basic science and clinical approach to developing computer aided diagnosis (CAD) systems to assist radiologists in detecting and classifying breast lesions.  

Breast cancer is the most common malignancy occurring in women and the second most frequent cause of non-skin cancer deaths among women.  Screening mammography programs have repeatedly shown a reduction in the mortality from breast cancer by 30 to 60%.  However, breast imaging suffers from a lack of specificity. The result is that 60 to 80% of breast biopsies performed in this country are for benign lesions and are therefore - in retrospect - unnecessary.  Because of the overlap in imaging features of benign and malignant lesions, however, these lesions cannot be differentiated without tissue sampling, and the extraordinary number of breast biopsies performed markedly increases the cost of breast cancer prevention programs and is an impediment to breast screening for some women.  Our work has focused on building computer aided classification systems to assist the radiologist in differentiating benign from malignant breast lesions without the use of invasive biopsies.  In our systems, imaging features of breast lesions are combined using artificial intelligence techniques with information such as the patient's age, family history, and change from prior imaging studies to determine the likelihood that a particular lesion is malignant.  This information can guide the radiologist to offer follow-up imaging rather than biopsy for those women with lesions that are very unlikely to be breast cancer.

My clinical research efforts include detailed evaluations of the strengths and weaknesses of computer aided detection systems that are now commercially available and in increasingly widespread use in radiology departments.  These systems analyze a digital version of the mammogram image to detect suspicious regions that may represent malignant masses or clusters of malignant microcalcifications.  While these systems have been shown to detect cancers initially overlooked by the interpreting radiologist, we are evaluating the clinical impact and limitations of the systems presently available.

A new focus of research in our lab is the development of full field digital mammography (FFDM) systems that acquire mammographic information using a digital detector rather than film.  The advantage of this technique is the possibility of developing advanced applications such as tomosynthesis, contrast enhanced mammography, and dual energy imaging, as well as clinical advantages such as the ability to manipulate the appearance of the image after acquisition and improve film storage and transport.  We are collaborating with a major imaging equipment manufacturer to develop both their commercial FFDM system and to develop tomosynthesis using that system.  Tomosynthesis is a technique in which several low dose X-ray images of the breast are obtained at various angles and thin tomographic slices of the breast are reconstructed.  This technique removes the problem of overlapping breast tissue, making detection of breast lesions easier, and, in theory, improving the sensitivity of mammography.

Representative Publications:
Baker JA, Kornguth PJ, Lo JY, Floyd CE Jr. Artificial neural network: improving the quality of breast biopsy recommendations. Radiology. 1996 Jan;198(1):131-5. (1996) Abstract

Baker JA, Soo MS, Rosen EL. Artifacts and pitfalls in sonographic imaging of the breast. AJR Am J Roentgenol. 2001 May;176(5):1261-6. (2001) Abstract

Baker JA, Kornguth PJ, Floyd CE Jr. Breast imaging reporting and data system standardized mammography lexicon: observer variability in lesion description. AJR Am J Roentgenol. 1996 Apr;166(4):773-8. (1996) Abstract

Baker JA, Kornguth PJ, Soo MS, Walsh R, Mengoni P. Sonography of solid breast lesions: observer variability of lesion description and assessment. AJR Am J Roentgenol. 1999 Jun;172(6):1621-5. (1999) Abstract

Baker JA, Soo MS. The evolving role of sonography in evaluating solid breast masses. Semin Ultrasound CT MR. 2000 Aug;21(4):286-96. (2000) Abstract

Baker JA, Soo MS. Breast US: assessment of technical quality and image interpretation. Radiology. 2002 Apr;223(1):229-38. (2002) Abstract

Lo JY, Markey MK, Baker JA, Floyd CE Jr. Cross-institutional evaluation of BI-RADS predictive model for mammographic diagnosis of breast cancer. AJR Am J Roentgenol. 2002 Feb;178(2):457-63. (2002) Abstract

Baker JA, Kornguth PJ, Lo JY, Williford ME, Floyd CE Jr. Breast cancer: prediction with artificial neural network based on BI-RADS standardized lexicon. Radiology. 1995 Sep;196(3):817-22. (1995) Abstract