Till date the quantification of the pain is done using the scale of the pain given by the patient, like ranking the pain from 1 to 10. Patients who are in pain are known to over exaggerate or underrate the amount of pain and on the other hand, patients who are bedridden with coma or some sort of brain injury cannot feel or explain the pain to doctors. There are no proper standards or method to objectively quantify the pain. This has become an obstacle for the doctors in taking the clinical decisions. The present quantifying techniques like patient questionnaire filling is not perfect and other latest techniques like “Quantifying cerebral contributions to pain beyond nociception” are complex.
Quantifying the pain can also help the doctors to recognize the arthritis at the initial stages. As per the survey done by health Canada in 2008, 15.3% of the Canadians who age more than 12 years are diagnosed with arthritis. Which amounts to 4.3 million of the total Canadian population. It has a chance to become more chronic condition with increase of age. Quantifying pain can also help doctors in tracking the injury recovery. Hence, there is a need for easy and automated technique to sense and quantify pain any region.
The nervous system in the skin controls the flow of blood and heat emissions from the skin. They can be captured using the thermal camera. At normal temperature, thermal imaging cameras can capture the human body who acts as a perfect emitter of infrared radiation, this radiation is emanating from the tissue. Infrared thermography (IRT) in the field of medicine is an intrusive, radiation-free, contactless technology for monitoring the physiological functions of skin temperature control. The effectiveness, safety and low cost of the IRT make it a useful auxiliary tool for detecting and locating thermal abnormalities characterized by increases or decreases in the temperature of the skin surface.
Human body has the tendency to radiate heat (infrared energy) from the pain region and the amount of radiation increases with the amount of pain. This temperature variations can be caught by sensitive thermal cameras and by extracting the features from the image we can classify the type of pain and also quantify the pain.
The main goal of my PhD is to create a trained system-using machine learning to recognize the possible pain regions and quantify the pain in patient and help the doctors to diagnose efficiently and also to explore the possibility of implementing the same on the infants & toddlers. I intend to do this by training the system with standard healthy body temperature of certain people and use it to recognize the region of interest in patients, which is susceptible to stress and pain.
1. Naresh Pal* and Aravind Kilaru, "Brain Tumour Segmentation using Weighted K-Means based on Particle Swarm Optimization", accepted for the special issue Green computing techniques for the smart world, Green Engineering, River Publisher. (Nov. 2017)
2. Aravind Kilaru and Naresh Pal*, “Local Chain Descriptor with deep learning for plant leaf disease detection”, Journal of Advanced Research in Dynamical and Control Systems, (Oct.2017).
3. Aziz Oukaira, Naresh Pal*, Ouafaa Ettahri, Emmanuel Kengne and Ahmed Lakhssassi, ''Simulation and Implementation of Thermal Convection Equation for Complex System Design'', International Journal on Engineering Applications, (Feb. 2017).
4. Naresh Pal* and Vivek Verma, “Visual Text Localization and Extraction for Devanagari Script”, IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278- 0661, p-ISSN: 2278-8727, Volume 18, Issue 5, Ver. 6 (Oct. 2016).
5. Chauhan Vaishali, Naresh Pal*, R. Vikram Raju, Sandeep Joshi, and Roheet Bhatnagar. "A New Method for Minimizing Unnecessary Handoff in 802.11." In Advance Computing Conference (IACC), 2017 IEEE 7th International, pp. 349-354. IEEE, 2017.