Application of crisp and fuzzy clustering algorithms for identification of hidden patterns from plethysmographic observations on the radial pulse.

Radial Pulse forms the most basic and essential physical sign in clinical medicine. The paper proposes the application of crisp and fuzzy clustering algorithms under supervised and unsupervised learning scenarios for identifying non-trivial regularities and relationships of the radial pulse patterns obtained by using the Impedance Plethysmographic technique. The objective of our paper is to unearth the hidden patterns to capture the physiological variabilities from the arterial pulse for clinical analysis, thus providing a very useful tool for disease characterization. A variety of fuzzy algorithms including Gustafson-Kessel (GK) and Gath-Geva (GG)have been intensively tested over a diverse group of subjects and over 4855 data sets. Exhaustive testing over the data set show that about 80 % of the patterns are successfully classified thus providing promising results. A Rank Index of 0.7739 is obtained under supervised learning, which provides an excellent conformity of our process with the results of plethysmographic experts. A correlation of the patterns with the diseases of heart, liver and lungs is judiciously performed.

Application of neural networks in the interpretation of impedance cardiovasograms for the diagnoses of peripheral vascular diseases.

An Impedance Cardio-vasograph (ICVG) system has been developed at the Electronics Division, Bhabha Atomic Research Centre (B.A.R.C) for the assessment of Peripheral Blood Flow and has been installed at the Department of Medicine, J.J. Hospital, Mumbai, India. Impedance cardio-vasography (ICVG) gives an indirect assessment of blood volume changes by measurement of normalized rate of change of electrical impedance (N dZ/dt) of the body segment. Parameters like Blood Flow Index (BFI) and Differential Pulse Arrival Time (DPAT) at different locations in both lower extremities (upper thigh, knee, calf and ankle) can be computed from these measurements. This work deals with the analysis of these parameters by a neural network system for obtaining proper diagnosis of subjects with peripheral vascular diseases. The designed network identified the presence of anatomical block or narrowing for most of cases presented to it during testing and also the status of collateral circulation in the lower limbs. The neural network was trained again, with the few cases, which were not predicted correctly. The collaterals after the site of occlusion were classified as good, moderate or poor as an aid to the physician. The network identified cases with athero-sclerotic narrowing satisfactorily and was also able to categorize cases where changes are observed only in one extremity, other remaining normal as in the cases of hemi-Leriche's syndrome. An additional parameter CVS (Coefficient of venous Statis) was calculated which is useful for the diagnosis of primary and secondary varicosity of the veins.