Life Sciences & Biotechnology
Title : | DTFQPM: Deep Thinking Fluorescence Quantitative Phase Microscopy for the automated classification of cancerous cells |
Area of research : | Life Sciences & Biotechnology |
Principal Investigator : | Dr. Gyanendra Singh, Inter University Center For Teacher Education, Uttar Pradesh |
Timeline Start Year : | 2023 |
Timeline End Year : | 2025 |
Contact info : | gyan2283@gmail.com |
Equipments : | Microscope Objectives
Stabilized White Light Source with driver, extra filter and Collimation unit
Optical Components
Optomechanical Components
sCMOS Color Camera
Stabilized He-Ne Laser with Mount
Excitation and Emission filters for different Fluorescence's (SETs) |
Details
Executive Summary : | Quantitative-phase Imaging (QPI) is one of the most common ways to use quantitative optical path delay (OPD) measurement to get label-free, high visceral contrast images of biological cells. While whole-cell morphology and dynamics can be studied with QPI, it does not have the ability to reach subcellular specificity. Fluorescent imaging is commonly used to obtain the cellular specificity of the biological cell. Off-axis holography captures the complicated wavefront of the sample with a single camera exposure combined with the Fluorescence imaging will be beneficial for the biological cell imaging. Slight angle must be inflicted between reference and sample beam resulting off-axis interference pattern, which is what produces the off-axis hologram. Combining the merits of the QPI and Fluorescence microscopy the imaging system will be developed enhanced quality of images will molecular specificity. Artificial intelligence (AI) integration will help in automate the developed device for the segmentation and classification that will reduce the burden on the pathologist. In this project, we are going to develop an automated Fluorescence based Quantitative-phase microscopy (QPM) for cancer cell classification using deep-learning. This microscope will quantify the molecular specificity of the infected cell along with its Quantitative information such as cell morphology, cell dynamics and its dry mass. The deep learning network will be used for the analysis of the recorded image that will make the system faster and reduced the burden on skilled pathologist. We believe our method can help in the early diagnosis of many types of cancer cells. |
Total Budget (INR): | 31,82,800 |
Organizations involved