Research programs/external Projects undertaken by Tapobrata Lahiri as Principal Investigator/ Principal Coordinator:

Broad academic and research interest lies mainly for Knowledge building and dissemination on development of algorithms on Biomedical Techniques, Fractal/Chaos simulation model and fractal measure, digital image processing, digital signal processing, soft computing models (Artificial neural network, fuzzy logic, genetic algorithm and their combinations), pattern recognition and to use them in Biomedical, Bioinformatics and other fields.

Specific research interest (non-exhaustive) has been already given in Major Research Interest Column. Under the scope of it, the summaries of the following external research projects currently being carried out by the applicant as principle investigator are also given.

1) Principal Investigator for the project entitled “Development of a Fractal and Soft Computing simulation based methodology to predict protein surface-roughness-map to help pharmaceutical designing of best possible small binding molecules – A drug design approach” funded by Ministry of Human Resource and Development, Govt. of India (MHRD sanction letter No. F.26-14 /2003. T.S.V. dated 14.01.2004)

Project objective

Ø Our first intention is development of a novel method for extraction of important structural information of protein surface

Ø We feel this new method may be considered as the need of the time because the existing experimental and prediction methodologies has their known limitations.

Ø The idea is to utilize the information of protein assembly (aggregates) to get back their individual property. It stems from the X-Ray Crystallography experiment where an ordered assembly of protein (crystals) is used to get back the individual protein structure. The difference in our approach is: we like to utilize spontaneously formed large aggregates of native protein. The main benefit of it is that these aggregates can easily be formed either spontaneously or artificially by mild centrifugation and subsequent suspension of pellet in very small volume of buffer.

Ø We like to study the above-mentioned apparently irregular aggregates with the help of fractal and other hypothesis meant for extraction of order-parameter from them.

Ø The target in the next step is to formulate a surface parameter of protein that is invariant to the change of orientation of protein. We like to design it in such way that, it can be linked to (i.e., correlated to) the active site of the protein and also to the family of the protein.

Ø The target in the final step is to find the appropriate aggregate parameters (say, mass fractal dimension, perimeter fractal dimension, minimum circle fraction, etc.) which can be mapped to the individual surface parameter (described above) of the protein through a mapping model (say, backpropagation ANN or other suitably chosen models). The validation of this model can be done by comparing the output of this model with the known surface parameter calculated from the PDB coordinates.

2) Principal Coordinator for the project entitled “Application of EMF for evaluation of health of vital organs” funded by ILTP of cooperation between India and Russia, DST, Govt. of India (DST sanction letter No. INT/ILTP/A-9.24 dated, 31.03.05)

Project Summary and Objective

This project is dedicated for the study and succesive adoption of biomedical signal and image information to explore non-invasive spatiotemporal description of heart. The special features in our attempt will be to mine the signal data with an intention to check whether it bears structural and functional information of human cardiovascular system more than that has been exlpored so far. The common intuition and our experience of various signal generating processes provoke the idea that signal is nothing but a cumulative result of a structure/function dualism generated from the dynamic state of human physique. In that sense the biomedical signal may be thought of as an another documented language describing dynamic state of a human physique. Our main effort will be to understand the language of the signal – an alltogether different approach than the previous approach of that kind (Moody and Mark, 1989, Tompkins, 1995; Brooks, 1997; He, 1998, Gulrajanai, 1998, Dixit and Callans, 2002; Coatrieux, 2002; Bondarenko, 2003). The problem involves processing and mining of data with the help of the inverse problem solution algorithms to understand the language of heart. The successful completion of the project will help us to understand the spatiotemporal relationship of the heart in a broad sense. To be specific, in this project we will target to find new spatiotemporal descriptive feature mining non-invasive biomedical signals/images to relate Ischemic Heart Diseases caused by Coronary Artery Blockade.

Keeping these above said views in mind, we shall follow the following broad methodology for our purpose:

1) To collect biomedical signals and images obtained from the normal and the patients of Ischemic Heart Diseases for the use of different users through the creation of web-controlled database.

2) Preprocessing of the signals/images data to eliminate noises/artifacts.

3) To try wavelet and windowed FFT to process the signals/images both in their unimodal as well as multimodal form.

4) To mine the signal data by i) inductive logic programming methods, ii) studying the morphology through shape-parameters, iii) irregularity through nonlinear dynamics like fractals, roughness and chaos theory, etc., to obtain new features describing the cardiovascular system of interest.

5) To apply reverse engineering tools like Hidden Markov Model, Independent Component Analysis, Inverse Solution etc. on the mined features of signals/images to get the spatiotemporal description of heart

3) Principal Investigator for the project entitled “Development of a Computer-aided Microscopic Tool for Structural Derivation of Pathologically Significant Proteins” funded by ICMR which is concluded recently.

Project Summary and Objective

This project will be dedicated to the study of recovery model related to mapping from aggregate to individual protein of interest. The idea stemmed from the fact that ordered assembly of a protein, its crystal, gives protein-structure through computational model of X-Ray crystallographic data obtained from this crystal. Similarly to our view, irregular assembly of protein, its aggregates, should also carry with it the individualistic information, may be, up to its individual structural level. This model will give us the structural information, especially surface-roughness profile of protein. Our initial experiment indicates that this particular profile may lead to identification of protein family and protein active site and moreover, the residue related to active site. The focus area in our study will be pathologically significant proteins and to get details of their deformed structure that triggers their assembly and subsequent fall out of the disease process.

Keeping these above said views in mind, we shall follow the following broad methodology for our purpose:

1) To develop a low cost, simple and fast computer-aided microscopic tool to study the correlation of protein aggregate parameter with the surface profile of a protein, Surface Roughness Index (SRI).

2) Finding active site of a protein of concern from its SRI measure without the help of prior structural knowledge from X-Ray Crystallography or other such structure determination methods so as to help better drug design.

3) To find out the structural discriminatory property of pathologically significant proteins which cause disease by their aggregation.

4) Principal coordinator for the project entitled “Development of new methods and algorithms for pathophysiological characterization of coronary blockade by processing ECG and similar quasiperiodic biomedical signals and biomedical images”, recently approved by ILTP of cooperation between DST,India and RFBR, Russia (DST sanction letter No. INT/RFBR/P-17)

Project Summary and Objective

The idea and goal of the new project is however set to implement the findings of the which will stand on solid technical foundation of the current project work entitled “Development of new methods in search for spatiotemporal descriptive features in the Cardiac signals and images” which was sponsored by DST-India and RAS-Russia under the scope of ILTP (Sanction no. INT/ILTP/A-9.24 dated 31.03.05). The broad achievements of this project was

1) Experimentation on usability of low-cost tools (ECG and Thermogram) in predicting

Coronary blockade.

2) Success in achieving best performance so far in predicting Coronary blockade by the

above-mentioned tools through pilot experiment.

Therefore we will put our joint endeavor to

1) To build a information-fusion platform to incorporate research findings of Russian and

Indian Scientists through jointly managed web-servers.

2) Through this server we intend to give some diagnostic service as it will be developed by us including blue-tooth technology for continuous monitoring of patients having cardiac ailments.

3) develop product that will be ready to by utilized by any diagnostic clinics.