| Title of the Research Problem |
Predictive Forecasting of Alzheimer’s Disease: Use of Optical Coherence Tomography (OCT) as a Diagnostic Tool with Emphasis on Related Molecular Signaling Pathways responsible in Retinal Degeneration |
| Abstract of the Research problem |
The proposal aims using Optical Coherence Tomography (OCT)- non-invasive, diagnostic optometric tool, for predictive forecasting of select neurodegenerative disorders/diseases much early in life, before identifiable symptoms, bio-markers, precipitate. Detecting anomalies/ retinal neve fibre layer (RNFL),thinning, ganglion cell degeneration, loss of axonal projections in the optic nerve, that get altered due to small vessel changes in select brain regions associated with Alzheimer’s that reflects in retina - considered as CNS tissue, developed as an outgrowth of brain followed subsequently by correlating data from IP/OP patients, external databases, will help generate genetic algorithm(s) using artificial intelligence in early precipitation of Alzheimer’s. |
| Target TRL to be achieved |
TRL-3
|
| Previous work done by the faculty |
CNN based deep learning architecture is used to classify the quality of OCT images and differentiate retinopathies. Simulation model was based on dataset from various sources including an open source Kaggle OCT datset, organized into three folders (train, test, val) and contains sub-folders for each image category (Normal, CNV, DME, DRUSEN).
Study involved classification of OCT images using modules, defining the path to dataset, loading the dataset, visualizing the data, defining classes, splitting data to training and test sets followed by construction of CNN model, building a linear stack and sequencing the model, validation labelling of data into various disease sets, prediction and assessment.
The model reproduced detectable retinal damages using OCT images with accuracy of about 96 % for CNV, DME and Drusen.
|
| Related Research Equipment / Lab Space / Facilities Available for solving this problem |
BAsic lab. facilities are available in BITS PIlani and SN-MRF but few equipments needs to be procured as capex besides solvents and testing kits- consumable.
|
| UN SDG Impacted |
UN-SDG-3 (Good Health and Well Being)
|
| Ideal Applicant Background |
Masters in Pharmacy- (pharamacology) with programming background;
Applied mathematics/Computer Science masters student with biomedical application, programming background
|
| Preferred Discipline of the Applicant |
Pharmacy - Pharmacology specialization;
Masters in Computer Science / Applied Mathematics with programming skills
|
| Faculty Name |
Mahesh R
|
| Faculty Web Page |
https://www.bits-pilani.ac.in/pilani/r-mahesh/
|
|
|
|
|
|
|
|
|