Interstitial Optical Diagnosis and Treatment of Breast Cancer
Abstract
Aim: To develop optical techniques for diagnosis and treatment of breast cancer We developed the hardware for collection and analysis of white light reflected from tissue (elastic scattering spectroscopy, ESS) as a diagnostic technique. Paired optical and conventional histologic measurements were obtained from 1250 sites in breast tissue and axillary nodes. Using artificial intelligence techniques (neural networks, hierarchical cluster analysis) coupled with innovative spectral processing, we looked for spectral features to identify cancer by model based analysis (MBA). We achieved a sensitivity and specificity for detecting cancer in breast tissue of 94% and 92% respectively (84% and 87% in excised axillary nodes). ESS has the potential for instant low cost detection of cancer without removing tissue from the patient in many tissues, justifying further study. Therapy aimed for complete ablation of small cancers using percutaneous Interstitial laser Photocoagulation (ILP). Safety studies treating fibroadenomas confirmed that laser necrosed tissue is resorbed and healing is without a scar or residual fibrous lump (ILP) is now used routinely for fibroadenomas). Studies on cancers were limited by recruitment problems, but ILP could ablate small cancers and contrast enhanced MR could detect untreated areas of cancer as small as 2mm Further studies should be undertaken.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2002
- Accession Number
- ADA410297
Entities
People
- Stephen G. Bown
Organizations
- University College London