Dr. Saritha Unnikrishnan

Principal Investigator / Researcher 

Department of Computing and Electronics

 

Research Topic:

“Computer Vision and Machine Learning with a focus on pharmaceutical manufacturing applications.”

PEM would like to introduce you to Dr. Saritha Unnikrishnan who is a lecturer in the Department of Computing and Electronics, IT Sligo, she is working as a Principal Investigator / Researcher in the PEM Strategic Research Centre. She joined the PEM research group in 2015 as a PhD researcher in Computer Vision and Machine Learning with a focus on pharmaceutical manufacturing applications.

 

Dr Saritha’s research work has been carried out in collaboration with GlaxoSmithKline (Ireland & UK) Ltd under the supervision of Dr David Tormey and Dr John Donovan. She also has several years of experience in the software industry working with various multinational companies.

 

Dr Saritha has presented her research at various conferences including the Future Intelligent Manufacturing session in the 4th International Conference on Universal Village, 2018 at MIT, Boston, USA. She won the best oral presentation awards at Microscopy Society of Ireland  Annual Symposium, 2018 and IT Sligo postgrad conference, 2017. She also won the best poster award in the IT Sligo postgrad conference, 2018.

 

She has recently been nominated as a National Delegate and Management Committee member of the EU funded COST Action 18206 (https://glimr.eu/) and is involved in the Working Group which focusses on co-ordinating the identification and quantification of advanced MRI biomarkers for the application in the field of glioma. (Glioma is the most common and aggressive brain tumour that arises throughout the central nervous system.)

Dr Saritha completed her PhD research work in April 2020 and has a patent published from her research work under the title ‘System for controlling an emulsification process’  (International publication number: WO 2020/079283 A1). A license for the technology has also been transferred to GlaxoSmithKline. Her recent publication in the Journal of IEEE Transactions on Industrial Informatics (https://ieeexplore.ieee.org/document/8968624)  provides information on the industrial application of her PhD work.

Figure 1. A comparative analysis of a machine learning approach and manual approach in the in-process classification of a pharmaceutical emulsion. Adopted from a previous paper authored by Saritha Unnikrishnan (https://link.springer.com/article/10.1007/s12247-019-09390-8).

Saritha is currently working as a researcher with PEM in collaboration with GlaxoSmithKline on the inline characterisation of multi-phase systems in manufacturing. This project investigates various real-time image analysis techniques for the characterisation of particles, droplets, and bubbles from multiphase systems such as liquid-liquid, gas-liquid, solid-liquid and solid-gas-liquid. It also explores how these characteristics could be modelled using machine learning to develop a novel automated technique for inline prediction of product quality in industrial processes. This project has the promising potential to avoid over-processing and to reduce the subjectivity associated with the existing conventional techniques. The outputs from this research could benefit a plethora of industries such as food, pharmaceuticals, bio-medical, cosmetics and polymers.

Her current research interests are mainly around AI-driven economy and applications in health innovation. These include digital automation in manufacturing using real-time image analysis and Artificial Intelligence (AI), Image data augmentation using generative models to make AI work with small data, Medical image classification using AI as opposed to manual screening to reduce overloads and to achieve less subjective and faster results.

Her publication list can be viewed at http://scholar.google.com/citations?user=eW64QkcAAAAJ&hl=en

 

Dr. Unnikrishnan has also recently been successful in securing a CUA funding bursary to support a PhD researcher in the area of medical image segmentation using Computer Vision and Artificial Intelligence 

Co-PI: Dr. David Tormey (PEM, ITSligo) and Dr. Ian McLoughlin (GMIT).