Contents lists available at ScienceDirect Infrared Physics & Technology journal homepage: www.elsevier.com/locate/infrared Regular article Temperature profile guided segmentation for detection of early subclinical inflammation in arthritis knee joints from thermal images Mrinal Kanti Bhowmik a, ⁎ , Kakali Das a , Debotosh Bhattacharjee b a Department of Computer Science and Engineering, Tripura University, Suryamaninagar, Tripura, 799022, India b Department of Computer Science and Engineering, Jadavpur University, Kolkata, West Bengal, India ARTICLE INFO Keywords: Hotspot detection Inflammation Region growing Thermal imaging ABSTRACT In arthritis, subclinical inflammation referred to the clinical condition when rheumatologists are in confusion about the presence of inflammation using clinical and pathological observations. Application of Thermal ima- ging in detection of subclinical inflammation is highlighted in this literature. Segmentation of the hotspot area from the thermal image is the initial step for further analysis of the hotspot. Analysis of the hotspot will help in prediction of the subclinical inflammation, impact of inflammation. Methodologies reported in existing literature for segmentation of hotspot or inflamed knee region in medical thermal images suffer from over and under extraction. In the present scope, we try to overcome this limitation by extending the conventional region growing seg- mentation technique with stronger similarity criteria and stopping rule. In this method, hotspot or inflamed region is generated by taking the intersection of two independent regions produced by two different version of Region growing algorithm using a separate set of parameters. An automatic multiseed selection procedure en- sures prevention of missed segmentation. We validate our technique by experimentation on various thermal image datasets like a newly created inflammatory thermal knee-joint-Database of 50 images, DBT-TU-JU Dataset, and DMR-IR Dataset. The effectiveness of the proposed technique is established compared to the per- formance of state-of-the-art competing methodologies. 1. Introduction Diseases cause inflammation of joints known as arthritis. Single or multiple joints may get affected [1] in arthritis. Pain and swelling in joints, stiffness of joints, restriction of movements of joints are common symptoms of arthritis [2]. Arthritis causes joint destruction which produces inflammation in the proximity of that joint. Rheumatologist plan the treatment for arthritis patients based on the inflammation as it can describe the activity of related disease. Reduction in inflammation of joints indicates the decrease in the progression of disease [3–5]. Subclinical inflammation is an important issue to be considered in the treatment of arthritis. Subclinical inflammation is referred to the clinical condition when rheumatologists unable to predict the presence of inflammation [8]. Presence of Inflammation is detected by Rheumatologists using clinical and pathological examination. Swelling, tenderness, pain, restriction of movements and temperature in joint surface are the clinical parameters consider to diagnosing the inflammation. Blood markers such as C-Re- active Protein (CRP) and Erythrocyte Sedimentation Rate (ESR) are the pathological tool for inflammation diagnosis. The condition of sub- clinical inflammation arises when clinical and pathological observa- tions contradict. Imaging technologies are also used for better under- standing of early subclinical inflammation such as ultrasonography (USG), magnetic resonance imaging (MRI) [6,7]. But these modalities are costly and observer dependent. Among them MRI persists radiation during examination. The distant goal of this paper is to provide an al- ternative opinion to the clinicians (radiologists, rheumatologists, etc.), who may feel confused about the possible presence of subclinical in- flammation in knee joint arthritis. In this scope, the paper signifies the importance of thermal imaging towards diagnosis of subclinical inflammation. Every object including human body maintains an inherent temperature profile and change in that profile usually indicate an abnormality [9]. Thermal medical imaging is able to detect the difference in the temperature profile of the concerned area of skin surface. The increase of temperature in the af- fected area reflects as a hotspot or inflamed region in thermal images [13,14]. Therefore, thermal images of arthritis patients with a hotspot confirm the presence of inflammation. As the hotspot contains https://doi.org/10.1016/j.infrared.2019.04.011 Received 12 January 2019; Received in revised form 10 April 2019; Accepted 10 April 2019 ⁎ Corresponding author at: Computer Science and Engineering Department, Tripura University (A Central University), 799022, India. E-mail address: mrinalkantibhowmik@tripurauniv.in (M.K. Bhowmik). Infrared Physics and Technology 99 (2019) 102–112 Available online 11 April 2019 1350-4495/ © 2019 Elsevier B.V. All rights reserved. T