Rajendra Singh Sisodia

Evaluating legacy assets in the context of migration to SOA

A key activity in the introduction of service oriented architecture (SOA) for an organization is to evaluate the suitability of existing assets for service orientation. We identify the core principles of SOA as the guide lines in evaluating the suitability of the existing assets. The existing metrics and guidelines that could be helpful in evaluating these principles are surveyed. This would benefit an organization in understanding the effort needed for migration and also to build proper services from the existing assets.

Segmentation of 2D fetal ultrasound images by exploiting context information using conditional random fields

This paper proposes a novel approach for segmenting fetal ultrasound images. This problem presents a variety of challenges including high noise, low contrast, and other US imaging properties such as similarity between texture and gray levels of two organs/ tissues. In this paper, we have proposed a Conditional Random Field (CRF) based framework to handle challenges in segmenting fetal ultrasound images. Clinically, it is known that fetus is surrounded by specific maternal tissues, amniotic fluid and placenta. We exploit this context information using CRFs for segmenting the fetal images accurately. The proposed CRF framework uses wavelet based texture features for representing the ultrasound image and Support Vector Machines (SVM) for initial label prediction. Initial results on a limited dataset of real world ultrasound images of fetus are promising. Results show that proposed method could handle the noise and similarity between fetus and its surroundings in ultrasound images.

Safe subset-regression test selection for managed code

In this extended abstract, we present our approach to computing a safe subset-regression test-suite, given the current, and the updated versions of the compiled binaries of managed C# code-base; the architecture of our implementation ChiARTS, the algorithm, and argument for the safety of our approach.

Automatic doppler signal analysis to assess utero-placental circulation for identifying high risk pregnancies

High risk pregnancy conditions such as preeclampsia, pregnancy induced hypertension, intra-uterine growth restriction and gestational diabetes are associated with defective utero-placental circulation. These conditions, if undetected early during pregnancy, are associated with poor pregnancy outcomes including high morbidity/mortality for the fetus/mother. The current state of the art for monitoring such conditions is via (color) Doppler ultrasound with key clinical parameters being observed in uterine and umbilical arteries being the resistance index (RI), pulsatility index (PI) and AB index and early diastolic notching. High risk conditions in pregnancy manifest as abnormal flow profiles and indices in select fetal/maternal blood vessels. These parameters are gold standard as far as current clinical practice goes but they still suffer from low sensitivity in detecting and predicting the above mentioned high risk conditions at an earlier stage. In this paper, we propose a method based on Doppler signal analysis that automatically identifies the above conditions with higher sensitivity even when the current RI/PI indices are normal.

Doppler based identification of uterine artery and umbilical artery for monitoring pregnancy

In this paper, we present an algorithm to identify umbilical and uterine arteries from a set of four different maternal and fetal arteries using their Doppler signatures. To distinguish these arteries, we use 132 Doppler signals collected from pregnant women with gestational ages between 24 to 40 weeks. Initially we filter them to remove noise; spectrograms are generated to extract good cycles, which are then analyzed to derive independent features that could uniquely represent an artery. A non-linear classification technique using k-NN (k-nearest neighbor) classifier is further applied to identify umbilical and uterine arteries. The proposed algorithm achieves sensitivity and specificity of above 95% and 97% for identification of uterine artery and above 63% and 80% for umbilical artery.

Combinatorial Approach for Automated Platform Diversity Testing

In recent years, product line engineering has been used effectively in many industrial setups to create a large variety of products. One key aspect of product line engineering is to develop re-usable assets often referred to as a platform. Such software platforms are inherently complex due to requirements of providing diverse functionalities, thereby leading to combinatorial test data explosion problem while validating these platforms. In this paper, we present a combinatorial approach for testing varied features and data diversity present within the platform. The proposed solution effectively takes care of complex interdependencies among diversity features and generates only valid combinations for test scenario. We also developed a prototype tool based on our proposed approaches to automate the platform testing. As part of our case study, we have used our prototype to validate a software platform widely being used across Philips Medical Systems (PMS) products. Initial results confirm that our approach significantly improves the overall platform testing process by reducing testing effort and improve the quality of the platform by detecting all interaction faults.

Automated fetal spine detection in ultrasound images

A novel method is proposed for the automatic detection of fetal spine in ultrasound images along with its orientation in this paper. This problem presents a variety of challenges, including robustness to speckle noise, variations in the visible shape of the spine due to orientation of the ultrasound probe with respect to the fetus and the lack of a proper edge enclosing the entire spine on account of its composition out of distinct vertebra. The proposed method improves robustness and accuracy by making use of two independent techniques to estimate the spine, and then detects the exact location using a cross-correlation approach. Experimental results show that the proposed method is promising for fetal spine detection.

Generating Different Views of Clinical Guidelines Using Ontology Based Semantic Annotation

Clinical guidelines comprise collection of medical information related to clinical goals and are used by people having diverse information needs. Commonly users are only interested in information pertinent to their respective goals. Current clinical guidelines are largely textual in form and lack suitable mechanisms for extracting information relevant to different users. This paper proposes a method to extract information from clinical guidelines using a view based approach. Our view generation approach has four steps namely annotation, view description, view extraction and view presentation. As a case study, we have illustrated the proposed method using the CVD (Cardiovascular diseases) risk management guideline published by WHO by creating views for different clinical care settings.