Convex Hull Based Detection of Overlapping Red Blood Cells in Peripheral Blood Smear Images

• Conference : WACBE 2015
• Date : 06/07/2015
• Used Technologies : Java & ImageJ libraries

The Segmentation of Red Blood Cells (RBCs) in blood smear images to obtain their count is often the first step in the diagnosis of various pathological conditions. Although several procedures have been devised for this task, a majority of them suffer from performance degradation due to the overlapping of cells. Various researches have been carried out to split these overlapping cells. This project aims to find the concavity points in the overlapping RBCs’ contours. This approach's dip point identification is based only on the convex hull of the overlapping cell. The contours of the concave regions are analyzed from the perspective of the centroid.

Retinopathy of Pre-maturity

• Duration : Aug 2015 - Dec 2015
• Used Technologies : C# & .NET, EmguCV (C# wrapper for OpenCV)

Retinopathy of Pre-maturity (RoP) is the project for early detection of Retinopathy from the set of fundus images. Aim is to stitch all fundus images into single tile image and applying various Image processing algorithms to detect the disease. The developed algorithm of this project from our colleagues, I just implemented those algorithms in .NET platform using EmguCV (C# Wrapper for OpenCV) library.

ARTSens - Android version

• Duration : Jun 2016 - Nov 2016
• Used Technologies : Java, Android Studio IDE

ARTerial Stiffness Evaluation for Non-invasive Screening (ARTSENS) is the project for measuring arterial stiffness of the patients, using ultrasound signals and it's reflections. Android version of this project is on development stage and we have worked on Data acquisition block and Plotting graph block. Ultrasound probe, which sends 5000 data-sets per cycle with the average of 25 cycles/sec using Android USB device/host serial communication method. From the device (ultrasound probe) to host (android device), we need to receive the data continuously and processing it with the Digital Signal Processing algorithms, and continuing blocks of artery wall detection and wall tracking. Still this project is on under development . I just worked on this project up-to data acquisition and plotting those values into graph, without noise filter.

Automated Diet Suggesting System

• Duration : Jan 2016 - Dec 2016
• Used Technologies : Drupal 7.56 CMS with PHP & MySQL

Diet suggesting for the patients, dieticians will prepare the diet-chart based on their allergies, diseases, requirements of calories & others and patient's food habit. For automate this whole system we developed this project. This project is developed as web based application using Drupal CMS (PHP, MySQL), because this application is requires centralized list of food items, diseases and allergies and other details. So admin of this web application, easily can able to add/modify those details from any ware. Initially it'll calculate the requirements of calories, carbohydrates, fat and etc., using the patien's bio-details then suggesting them 2/3 combinations of diet chart.

iQuant Analyser - Windows Embedded OS

• Date : Jan 2017 - Apr 2017
• Used Technologies : Windows Embedded 8.1 Industry Pro, Powershell & Batch file scripting

HTIC developed a lateral flow immunofluorescence assay test kit reader, which is iQuant Analyser. For deployment of this project, device should work on customized OS for make it as Kiosk machine. For this, Windows enables us to customize their OS (Windows 8.1 Embedded Industry Pro) with their support. The device is implemented with the filters like start-up application, file-write protection, keyboard filter, USB filter and etc., Using these features we customized the full Kiosk ready OS for the large scale deployment. The customized OS packaged with the drivers & dependent packages, and Powershell & Windows batch scripts. Finally full captured OS as WIM file distributes for deployment in the production .

ImageQuant (Android version)

• Duration : May 2017 - July 2017
• Used Technologies : Java, Android Studio IDE

ImageQuant - An image-based quantitative Immunoassay Analyzer. Blood sample cartridges will be captured using android device's integrated camera (Camera with the capability of capturing RAW images & manual 3A controls), and the captured image will be processed using Imaging techniques. Android version of this application started to develop in Andriod Studio, and the app has challenging requirements as follows:

1. Always initiate with the primary camera
2. Camera should have capabilities of capture RAW (.DNG extension) image &
3. Manually able to control the Shutter speed, ISO & Focus
4. GUI handling with the real-time change
5. integrate OpenCV native library into the project, process the captured images with it

ImageQuant - Windows 10 with AWS integrations

• Duration : Aug 2018 - till the date
• Used Technologies : Java, C# .NET, AWS DynamoDB & AWS S3, Windows Batch scripts

ImageQuant Windows based device: will push all log files to the AWS S3 cloud and device's hardware and software details will be maintained in the AWS DynamoDB, and device will be updated through our own applications from the cloud.

SONIO

• Date : Current project
• Used Technologies : Native and CLR based C++ & C#

Integration of hardwares Universal Robot, Clarius HD Ultrasound probes with 3D-Reconstruction library.