Analysis of scoliosis requires thorough radiographic evaluation by spinal curvature estimation to completely assess the spinal deformity. Spinal curvature estimation gives orthopaedic surgeons an idea of severity of spinal deformity for therapeutic purposes. Manual intervention has always been an issue to ensure accuracy and repeatability. Computer assisted systems are semi-automatic and is still influenced by surgeon’s expertise. Spinal curvature estimation completely relies on accurate identification of required end vertebrae like superior end-vertebra, inferior end-vertebra and apical vertebra. In the present work, automatic extraction of spinal information central sacral line and medial axis by computerized image understanding system has been proposed. The inter-observer variability in the anatomical landmark identification is quantified using Kappa statistic. The resultant Kappa value computed between proposed algorithm and observer lies in the range 0.7 and 0.9, which shows good accuracy. Identification of the required end vertebra is automated by the extracted spinal information. Difference in inter and intra-observer variability for the state of the art computer assisted and proposed system are quantified in terms of mean absolute difference for the various types (Type-I, Type-II, Type-III, Type-IV, and Type-V) of scoliosis.

We build an automatic phoneme recognition system based on Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. Models were built by using Stochastic pattern recognition and Acoustic phonetic schemes to recognise phonemes. Since our native language is Kannada, a rich South Indian Language, we have used 15 Kannada phonemes to train and test these models. Since Mel -Frequency Cepstral Coefficients (MFCC) are well known Acoustic features of speech[1,2], we have used the same in speech feature extraction. Finally performance analysis of models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields good results and can be used in developing Automatic Speech Recognition systems.

In this paper, the inverse problems of cardiac sources using analytical and probabilistic methods are solved and discussed. The standard Tikhonov regularization technique is solved initially to estimate the under-determined heart surface potentials from Magnetocardiographic (MCG) signals. The results of the deterministic method subjected to noise in the measurements are discussed and compared with the probabilistic models. Hierarchical Bayesian modeling with fixed Gaussian prior is employed to quantify the uncertainties in source reconstructions. A novel application of Variational Bayesian inference approach has been presented to estimate the heart sources. The reconstruction results of Variational Bayesian model with non-stationary priors are compared with solutions of simplistic Bayesian approach; and the performances are evaluated using Root Mean Square Error (RMSE) and correlation co-efficient metrics. The Bayesian solutions in the study are also extended to localize the MCG sources for two types of Myocardial infarction cases.