Major Research interest of Tapobrata Lahiri

1)      Medical Informatics: current research interest is for the study and successive adoption of biomedical signal (ECG primarily) and image (thermograms, MRI-scans) information to explore non-invasive spatiotemporal description of vascular systems (especially, heart). The background hypothesis is that like DNA (which can be thought of as an encoded signal) biomedical signal may be thought of as another documented language (which is rather decoded) describing cumulative result of a structure/function dualism generated from the dynamic state of human physique. Our main effort is to understand the language of the signal through inverse problem solution algorithms, e.g., Hidden Markov Model, Backpropagation Network, Independent Component Analysis, Fractal/Chaos theory based simulations. The preliminary publications and patent application from our work show the benefit of building a low-cost and non-invasive diagnostic substitute of angiogram for the screening of Myocardial Infarction.

2)      Structural Proteomics: Current research interest is to develop a technique to evaluate protein surface property (e.g., functional site) from the simple experimental output of its surface-active processes (means utilization of bulk properties of protein, e.g., microscopic images of heat denatured aggregates and nano-particle mediated aggregates of protein, spectrophotometric and spectrofluoremetric data etc.). So far the objective was to design a simple intermediate parameter, "Surface Roughness Index (SRI)" and also to utilize surface describing Zernike Descriptor which will bridge the experimental output (through listing of orientation independent surface topological properties) with the required target, i.e., functional site. Our preliminary publications emphasizes on the significant benefit of such work for

i) validation of protein structure that has been obtained through various predictive models, and therefore,

ii) fast evaluation of protein active site without bothering about crystallographic or NMR-structures.