Development and Validation of a Low-Cost LEDbased Multispectral Imaging System

Abstract

Hyperspectral (HS) imaging systems are currently widely used in biomedical research for developing new diagnostic tools that could help clinicians to improve accuracy in diagnostic tasks. These HS systems are commonly expensive and for certain medical applications they capture redundant information not needed for the application at hand. Hence, research for developing low-cost HS or Multispectral (MS) acquisitions systems is demanded by the medical instrumentation industry. In this study, a MS imaging system based on Light Emitting Diode (LED) technology has been designed and validated. To achieve this, an Arduino UNO microcontroller, a 4-to-16 demultiplexer, and a set of modular control circuit boards has been utilized for LED control, along with a ring structure to house the set of selected narrow band LEDs. Several tests have been conducted to validate the spectral response of the system by using a calibrated polymer as reference, demonstrating the potential of the proposed approach to develop customized LED-based MS systems.