The Biostatistics & Medical Informatics (BMI) Laboratory in the School of Medical Science & Technology (SMST), IIT Kharagpur offers a very inspiring & encouraging environment to the student by providing opportunities for research in very diverse field of specialization. The major research areas of the lab include Microscopic Imaging, Biomedical Signal & Image Analytics, Medical Expert System, Statistical Modeling & Mechine Learning, Medical Informatics, Automated Screening Tool Development.
Facilities
The use of information technology in healthcare has started more than 25 years ago. An automated, robust, user-friendly patient record system is very much needed which will improve the present status of Indian healthcare system. A web based distributed patient health (clinical and pathological information both) record system is being developed in BMI lab. This system records patient's epidemiological, clinical and pathological information and it maintains doctor's reports. The developed system has been enabled to work with or without internet connection. In belief, after a patient's arrival in a hospital for health checkup, his / her information is recorded into hospital's database. The patient is given a unique ID which acts as identification in the system. The stored information from hospital is uploaded in central server. Clinical experts can then view patient's record and give their diagnosis. Reports of the clinical experts can be saved into database. Hence, the patient can get his report from hospital or by visiting the website. The developed system will also help in developing a huge database for epidemiological study.
Pathological image processing is increasingly gaining importance for better characterization, visualization as well as disease recognition. Amongst various pathological tests of human being, microscopic evaluation of peripheral blood smears is very common and indispensable towards the characterization of pathological parameters. In the conventional practice, pathologists observe the abnormalities viz., parasitemia, anaemia, thalasaemia, leukemia, psoriasis etc based on the microscopic changes in terms of cellular morphology, texture and intensity of blood images. Usually this visual process of characterization and disease identification lead to complexities because of ambiguity. In order to improve the reliability of the decision by minimizing this uncertainty, the present work primarily focuses on the development of fuzzy set and intuitionistic fuzzy set theoretic approaches for segmentation, feature extraction and classification of light microscopic images of peripheral blood smears towards the detection of malaria, anaemia and leukemia. The total work is being depicted through the following schematic diagram.
Figure: Schematic Representation of Malaria/Anaemia/Leukemia detection process
Coronary artery disease (CAD) has become one of the mostly occurring diseases in the world. According to WHO (World Health Organisation) report, CAD has become a modern epidemic and it is estimated that by 2010 India is expected to have 60% of world’s heart disease with the younger generation being at maximum risk. In India mortality attributed to cardio vascular disease (CVD) is expected to rise by 103% in men and 90% in women from 1985 to 2015. CAD is the result of accumulation of plaques within the coronary arteries, condition known as Atherosclerosis. These plaques are made up of fat, cholesterol, calcium and other substances. When plaques grow, it narrows the lumen of the coronary arteries; thereby decreasing the blood flow to the heart muscle causing pain. The pain is felt in chest, neck, jaws, abdomen, arms and shoulder commonly called Angina. The amount of blood supplied to the tissue becomes inadequate resulting in myocardial ischemia. One of the solutions to this massive problem is to make the people aware of their respective CAD risks in advance and take preventive measures accordingly. It is only possible when an early detection of CAD occurs. In view of this a CAD screening expert system is being developed as follows:
Figure: Schematic Representation of CAD screening process
Polycystic ovarian syndrome (PCOS) is the most common reproductive endocrinopathy of women during their childbearing years, with a reported prevalence of 5–10% (Franks, 1995). The overall prevalence among women in this age group is between 4% and 8 % (Chang R, 2004) although the prevalence may be as high as 30% in women with secondary amenorrhea, 75% in women with oligomenorrhea, and 90% in women with hirsutism. Its clinical manifestations may include menstrual irregularities, signs of androgen excess and obesity. The syndrome seems to be associated with an increased risk for DM type II, infertility, ovarian cancer. One of the solutions to this problem is to screen the people at an early stage and take preventive measures accordingly. In view of this a PCOS screening system is being developed as follows: The following flowchart shows the overall screening system for the PCOS patients at an early stage. With the help of system, doctors will be assisted in the early screening of the PCOS patients.
Figure: Flow chart of Screening System of PCO
