SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2019 | Register Today

SPIE Defense + Commercial Sensing 2019 | Register Today

2019 SPIE Optics + Photonics | Call for Papers



Print PageEmail PageView PDF

Biomedical Optics & Medical Imaging

Noninvasive confocal microscopy targets malignant cellular morphology

Since cancer will eat away flesh and may kill, its efficient and effective detection at an early stage is important.
22 December 2009, SPIE Newsroom. DOI: 10.1117/2.1200912.002526

Our skin represents an exquisite feat of bioengineering. The surface provides an easily accessible imaging environment containing the vascular, gland, and nervous systems, hair bulbs, and rapidly multiplying stem cells that proliferate to replenish the epidermis. We must protect ourselves from the outside world with this organ and remain free of cancer. Observation of its microscopic physiology must be achieved through imaging to understand the system or assess and treat any skin disease. Noninvasive inspection of morphological structure and development in the skin's dynamic setting requires synthesis of optical engineering fundamentals with clinical research. Ongoing research focuses on building tools to image the skin system. The ultimate goal is to treat disease armed with biological understanding gained by technology-enabled vision, improving on the current histopathological standard, and enabling detection of cancer earlier and more completely than previously possible. Two of our main research projects involve confocal microscopy to detect cancer: noninvasive detection of melanoma (differentiation from benign nevus or blemish) and rapid surveillance of biopsy margins to detect residual tumor sites.

Noninvasive melanoma detection

Detection of melanoma causes substantial angst, both clinically and histopathologically. Noninvasive confocal microscopy may offer reliable quantitative diagnosis and rapid surveillance of suspicious nevi. The evolution of clinical melanoma treatment has reached three major milestones, including recommendation of 5cm marginal excision by Lancet in 1907, Clark's level staging in 1969, and the more reproducible Breslow thickness measurement. To achieve the next major milestone, we must quantify standard, malignant microscopic features that can be measured reliably with noninvasive clinical techniques and validate against the histopathology to determine sensitivity and specificity in the clinic.

Confocal microscopy is ideal for targeting early lesions because it is the only noninvasive technique offering quasi-histological resolution and penetration into the dermis, which encompasses all in situ diseases. Noninvasive screening of pigmented lesions is currently achieved with a dermatoscope, a digital camera that uses polarized light. Using the standard ABCD method of classification, which is 97.9% sensitive and 90% specific, clinicians judge the macroscopic features asymmetry, border, color, and diameter. In contrast, our method uses confocal images at diffraction-limited resolution to include cellular and subcellular detail and implements computer vision to quantitatively identify the presence or absence of malignant morphology. A recent, unpublished study based on our earlier work1,2 shows 100% sensitivity and 100% specificity, and our method generates a numerical malignancy factor that is higher for malignant melanoma than for benign nevus. This technique may change the paradigm of melanoma detection, enabling investigations of many more suspicious lesions with sensitivity and specificity equivalent to or better than those achieved by histopathology.

Confocal multimodal Mohs mosaic microscopy

A primary objective for cancer treatment in cosmetically sensitive areas, such as the head and neck, is to preserve the healthy tissue around a lesion. The aim of Mohs surgery is to remove as little healthy tissue as possible while containing the entire tumor. Staged micrographic excision is the standard of care for minimizing collateral damage while effectively removing a tumor. Histopathology of the previous excision is used to guide each subsequent excision, but it requires highly trained technicians, is expensive, and takes a long time, leaving the patient to anxiously wait in pain between stages while the diagnosis is completed.

Confocal multimodal Mohs mosaic microscopy (cMMMMM) can be used to guide tissue-sparing staged cancer excisions by screening margins for the presence of tumor, providing a large field of view with sufficient resolution of cellular and nuclear detail. cMMMMM can expedite surgery, thus improving clinical treatment of basal- and squamous-cell carcinoma (BCC and SCC) based on Mohs surgery because the polygon scan engine works rapidly.3 Early work on tumors4,5 showed reflectance mosaics of superficial and nodular BCC using acetic acid as a contrast agent to brighten nuclei. However, the strongly scattering dermis appeared bright as well. Large tumors with dense nucleation were easily detected in mosaics but small tumors, especially tiny strands of infiltrative and sclerosing BCC containing as few as 5 to 10 nuclei, were not. Although easily seen in individual confocal images that display small fields of view with high magnification, they become hidden in large fields with low magnification.

Figure 1. Confocal multimodal mosaic microscope. (left) To excite fluorescence, 488 and 532nm lasers are used. Added to reflectance (center), both observations are used to generate the composite image (right). Infiltrative basal-cell carcinoma (bottom) shows malignant morphology.

We demonstrated feasibility6 for fluorescence-mode imaging with acridine orange to label nuclei, which dramatically improves nuclear contrast. Sensitive and specific detection of all subtypes of BCC, including the tiny strands, has also been shown.7 Figure 1 shows the bench-top microscope and a mosaic with infiltrative BCC. Our ongoing efforts aim at maximizing clinical impact by creating data presentation that can be interpreted easily8 and fabricating simpler scanning devices,9 as well as developing cMMMMM techniques to distinguish SCC from normal skin.

Dan Gareau
Oregon Health and Science University
Portland, OR

Daniel Gareau develops biomedical technologies that use optics. Having studied under Steve Jacques and Milind Rajadhyaksha, he enjoys exploiting the power of light for diagnostic gain. With a focus on imaging with lasers, spatial and temporal resolutions can be achieved that solve medical problems and provide biological insight.