Researchers at the Indian Institute of Technology in Kharagpur (IIT- Kharagpur) have developed a new highly accurate, affordable, and non-invasive device that detects oral cancer in resource-constrained clinical settings. The technology includes a portable handheld unit that combines various sensors and controllers that feed the measured data to a computer simulation engine to classify normal, pre-cancer, and cancer cases in the oral cavity. It does not need referrals to specialised medical centres for resource-intensive diagnostic procedures.
The device is a portable and user-friendly blood perfusion imager (BPI) that combines a miniature far-infrared (FIR) camera and a humidity sensor, which are electronically controlled and interfaced with a combined physics-based and data-driven software engine. The process has proven to be technologically superior to thermal imaging-based screening technologies currently in use. This is because the temperature in the tissue itself varies with the surrounding conditions. With combined variabilities in the local blood flow and metabolism, there is not always a specific indicator of the diseased state under investigation. The new device offers an automated, touch-free approach to estimating blood flow variations in different regions of the potentially diseased tissue, specifically relating to the diseased condition.
Experts from the Guru Nanak Institute of Dental Sciences and Research supervised the clinical trials and have established the efficacy of the method in differentiating cancerous and precancerous stages of suspected oral abnormalities, as verified by high-standard biopsy reports. The research was recently published in the Proceedings of the National Academy of Sciences, USA.
According to a report, decisive exclusivities of the technology have also been achieved by infusing a machine learning (ML)-based classification approach with physics-based analytics, based on thermal images obtained from the portable device. Since oral cancer, at its early stage, is known to manifest an increase in blood flow whereas full-grown cancer reveals a decrease in blood flow similar to pre-cancer or normal cases; such data-science augmented interpretation algorithm has minimised inevitable instances of misclassification due to similar common deceptive features among different medical conditions. This may be compounded by obvious inter-patient variability that, in borderline cases, may lead to wrong clinical decisions.
The device consists of a probing unit for screening, and a processing unit for obtaining blood perfusion data and disease recognition. The probing unit is composed of a holder and sensor housing. The holder is used for guiding the sensor housing to the measurement site and the sensor housing maintains a stable environment, minimising the effect of breathing. The sensor housing consists of an on-chip long-wave infrared camera to measure tissue temperature and a fully-calibrated digital humidity sensor to measure ambient temperature and relative humidity inside the mouth. The camera sensor array captures the spectral radiance and uses additional signal-processing electronics to convert the radiometric values into temperature values.
The patent for this new technology has already been filed. Cancer of the oral cavity remains one of the major causes of morbidity and mortality in socially challenged communities, which reveals an on-an-average 80% chance of five-year survival rate if diagnosed early. The survival rate drops to 65% or less in more advanced stages. Most oral cancers remain undetected until reaching an advanced stage. In resource-constrained settings, there is a serious dearth of accurate yet affordable diagnostic tools to arrive at a decisive recommendation during the first possible clinical examination of the patient.
