UCL DEPARTMENT OF MEDICAL PHYSICS AND BIOENGINEERING
FACULTY OF ENGINEERING SCIENCES
UCL logo

This website is maintained by Jem Hebden.

Last update of this page: December 31, 2003.

BORL Website

MONSTIR: Technical Description

The purpose of the MONSTIR imaging system is to acquire measurements of the flight times of photons travelling through the tissue between multiple pairs of points on the surface. MONSTIR has 32 independent time-resolved detectors are operated in parallel. The imaging device is illustrated schematically below.

MONSTIR Schematic
Figure 1: A schematic of the 32-channel time-resolved imaging system.
Click here to view an animated version of a similar diagram.

Pulses of light from a dual-wavelength fibre laser (custom built for the system by IMRA America, Inc) are coupled successively into a series of optical fibres so that the point of illumination on the tissue surface is varied sequentially. Each pulse has a duration of around 1 picosecond. The output end of each source fibre is permanently integrated along the central axis of a corresponding detector bundle, 3.2 mm in diameter. Meanwhile the transmitted light is collected by 32 detector fibre bundles simultaneously. The source/detector fibre bundles are arranged over the surface of the breast or neonatal head. The output of each detector fibre bundle is coupled to the cathode of a microchannel-plate photomultiplier tube (MCP-PMT) via a variable optical attenuator. The attenuators ensure that the intensity of detected light does not saturate or damage the MCP-PMT, and the flux of photons is sufficiently small to prevent detection of multiple photons during each electronic cycle. The system employs four 8-anode MCP-PMTs. Electronic pulses generated each time a photon is detected are sampled by a sophisticated NIM-based electronic system manufactured by EG&G ORTEC (see Wells et al [1]). By measuring the delay between these pulses and a reference signal received directly via the laser, histograms of photon flight times (temporal point spread functions, or TPSFs) are gradually built up within the storage memory of the device. Each represents the TPSF for a distinct line-of-sight across the head. For 32 source/detector positions, the instrument can acquire up to a total of 1024 TPSFs.

MONSTIR photo Figure 2: MONSTIR adjacent to an infant cot in the foreground.

The instrument, known as MONSTIR (Multi-channel Opto-electronic Near-infrared System for Time-resolved Image Reconstruction), is contained in a rack 1.8 m high and 0.9 m deep. A separate, smaller rack houses the fibre laser, a cooling unit for the MCP-PMTs, and the power supplies. Despite its size, MONSTIR is designed to be fully portable and is able to operate at the cotside in the neonatal intensive care unit. Development of MONSTIR is being conducted by a research team led by Jem Hebden. Data processing involves the use of the TOAST image reconstruction algorithm developed at UCL by Simon Arridge and Martin Schweiger. Construction of the system was completed in May 2001, and it is now being evaluated in the clinic and the laboratory. MONSTIR and its performance is described in a recent publication by Schmidt et al [2]. For a more comprehensive description, please see the PhD thesis of Florian Schmidt [3], a copy of which is available from this website (click here).

This co-axial arrangement of the source fibre within the detector bundle has two major benefits. First, it decreases the number of connectors required to couple 32 sources and 32 detectors to the infant head. Second, it enables back-reflected light at the surface to be used to characterise the temporal characteristics of the time-resolved imaging system, and thus calibrate the instrument in situ [4]. This replaces an earlier calibration method [5] which involved a time-consuming series of measurements on an appropriate phantom. For clinical measurements, the ends of the combined source/detector bundles are held approximately 10 mm from the skin surface, providing a common circular illumination and detection area with a diameter of 6 mm. This is sufficiently broad to ensure that a moderate amount of hair (e.g. on the infant head) is not an overwhelming influence on the coupling of light into and out of the skin. The large illumination area also facilitates use of a higher input power without exceeding maximum permissible exposures.

  1. Wells, K, Hebden, JC, Schmidt, FEW, and Delpy, DT (1997): The UCL multichannel time-resolved system for optical tomography, Proc. SPIE 2979, 599-607.
  2. Schmidt, FEW, Fry, ME, Hillman, EMC, Hebden, JC, and Delpy, DT (2000): A 32-channel time-resolved instrument for medical optical tomography, Review of Scientific Instruments 71(1), 256-265. Download PDF file.
  3. Schmidt, FEW (1999): Ph.D. Thesis, Development of a time-resolved optical tomography system for neonatal brain imaging, University of London.
  4. Hebden, JC, Gonzalez, FM, Gibson, A, Hillman, EMC, Yusof, R, Everdell, N, Delpy, DT, Zaccanti, G, and Martelli, F (2002): Assessment of an in situ temporal calibration method for time-resolved optical tomography, Journal of Biomedical Optics 8, 87-92. Download PDF file.
  5. Hillman, EMC, Hebden, JC, Schmidt, FEW. Arridge, SR, Schweiger, M, Dehghani, H, and Delpy, DT (2000): Calibration techniques and datatype extraction for time-resolved optical tomography, Review of Scientific Instruments 71(9), 3415-3427. Download PDF file.

 


University College London - Malet Place Engineering Building - Gower Street - London - WC1E 6BT - Telephone: +44 (0)20 7679 0200 - Copyright © 1999-2005 UCL


Search by Google