Saikat Mondal

Assistant Professor

Dr. Arun More


Sayed Asaduzzaman

Assistant Professor



Android based application to predict heart attack risk.

Nowadays, Ischemic Heart Disease (IHD) (Heart Attack) is ubiquitous and one of the major reasons of death worldwide. Early screening of people at risk of having IHD may lead to minimize morbidity and mortality. A simple approach is proposed in this paper to predict risk of developing heart attack using smartphone and data mining. Clinical data from 835 patients was collected, analyzed and also correlated with their risk existing clinical symptoms which may suggest underlying non detected IHD. A user friendly Android application was developed by incorporating clinical data obtained from patients who admitted with chest pain in a cardiac hospital. Upon user input of risk factors, the application categorizes the level of IHD risks of the user as high, low or medium. It was found by analyzing and correlating the data that there was a significant correlation of having an IHD and the application results in high & low, medium & low and medium & high categories; where the p values were 0.0001, 0.0001 and 0.0001 respectively. The experimental results showed that the sensitivity and accuracy of the proposed technique were 89.25 % and 76.05 % respectively, whereas, using C4.5 decision tree, accuracy was found 86% and sensitivity was obtained 91.6%. Existing tools need mandatory input of lipid values which makes them underutilized by general people; though these risk calculators bear significant academic importance. Our research is motivated to reduce that limitation and promote a risk evaluation on time.

Heart Sound Abnormalities

Wireless Electronic Stethoscope to Classify Children Heart Sound Abnormalities

In this research project, a wireless stethoscope has introduced that can communicate with a smartphone to receive children's heart sound. Along with an automated method that recognizes children heart sound abnormalities. That isolation of heart sound is based on time-frequency characteristics. Where it is preceded using Mel-frequency Cepstral Coefficients (MFCCs) signal processing method. The processed sounds are extracted using five feature extraction algorithms. Then it is classified using four support vector machines (SVM) kernel. Total 60 heart sounds were collected, where 30 sounds having abnormalities and rest 30 sounds containing normal heart sound. Though massive measures of action have already been done in this area, still necessity of more bearable cost device and accurate method is present. Here, the submitted apparatus cost is approximately 18 USD, which is the cheapest than most other device used in previous work. Simultaneously it is lightweight and bearable to use in rural and underprivileged area. With RBF kernel of SVM, the proposed method shows 94.12% accuracy which is the highest.


An Application Programming Interface to Save the Disaster Affected People

The automated chatbot system has introduced a new era of modern technology. Recently the chatbot system plays an important role as a virtual agent in different respects. A chatbot system cannot replace a human agent but it can provide initial support at any time instantly. This type of instant support can help a victim at the time of the natural disaster period efficiently. It can also play a role to reduce the amount of damage. In this research paper, we have proposed a chatbot application programming interface (API) system named Safeguard that can be integrated into different social media as well as in any application. This system will be able to support a victim and give guidelines on disaster period. For this purpose, natural language understanding was used by Dialogflow tool. Dialogflow has helped to create the application programming interface system by using intents, entities and text responses by the implementation of the natural language processing system, cloud storage and JSON.

Acute Myocardial Infarction (AMI) Prediction

Smartphone-Based Heart Attack Prediction Using Artificial Neural Network

Heart attack is among a few of the deadly diseases that cause the death of thousands of people each year globally. It is possible to minimize morbidity and mortality by early screening for those who are at high risk of getting acute myocardial infarction (AMI), known as a heart attack. Android software was implemented to anticipate the risk of getting a heart attack to walk of sudden death. We conducted a survey and collected clinical data from 835 patients that have been analyzed and correlated with 14 risk factors. To predict the heart attack, we used the neural network technology to learn from the clinical data and make predictions. We chose Nesterov-accelerated adaptive moment estimation (Nadam) as an optimizer and categorical cross-entropy as loss function as it fit the best for our neural network model for the best prediction performance. We were able to train our model to predict AMI with 91% accurately. Then, we evaluated our model performance by computing sensitivity (i.e., 81%), specificity (i.e., 98%), precision (i.e., 96%), and F1-score is (i.e., 88%). This trained model was used to implement android software. A user has to answer 14 questions, and based on these answers, the software will predict if the user has a chance to get AMI. This software is free to use, and anyone can use it. The main goal of our research is to implement a simple system to track AMI on a daily basis to lead a healthy life and to avoid sudden deaths.