Assessing risk for voluminous legal documents such as request for proposal, contracts is tedious and error prone. We have developed "risk-o-meter", a framework, based on machine learning and natural language processing to review and assess risks of any legal document. Our framework uses Paragraph Vector, an unsupervised model to generate vector representation of text. This enables the framework to learn contextual relations of legal terms and generate sensible context aware embedding. The framework then feeds the vector space into a supervised classification algorithm to predict whether a paragraph belongs to a pre-defined risk category or not. The framework thus extracts risk prone paragraphs. This technique efficiently overcomes the limitations of keyword based search. We have achieved an accuracy of 91% for the risk category having the largest training dataset. This framework will help organizations optimize effort to identify risk from large document base with minimal human intervention and thus will help to have risk mitigated sustainable growth. Its machine learning capability makes it scalable to uncover relevant information from any type of document apart from legal documents, provided the library is pre-populated and rich.

Diffuse optical tomography (DOT) is an imaging modality which utilizes an array of near infrared light source (830 nm) for tissue illumination. The multiple-scattered light is detected using a pinhole camera. There are three primary absorbers at this wavelength which includes, water, oxygenated and deoxygenated hemoglobin, all possessing relatively weak absorptions. This provides a spectral window through which we can attempt to localize absorption and scattering in the tissue. Current techniques in imaging involve use of ionizing radiation which cause harm to tissues. For better image quality, the dosage has to be increased which induces the risk of cancer. Hence, we aim to exploit the non-ionizing characteristic of near infrared radiation (NIR), a potentially harmless band to image soft tissues. In our proposed system we are reconstructing 3-D images from 2-D cone beam projections using Feldkamp, Davis and Kress (FDK) algorithm which is most widely used because of its effective spatial resolution and duration time. It can also handle truncated data in longitudinal direction. The object to be imaged is positioned on the turntable and is rotated at 180 degrees. A major requirement of the setup is to position the phantom at equal distance from the source and detector. Both the camera and the stepper motor are controlled using MATLAB and are synchronized to work simultaneously. During the initial trials we propose to develop a phantom using paraffin wax that mimics the soft tissue properties. Eventually, experimentation will be done with different phantom models to test the compatibility and efficiency of the algorithm developed.

No abstract