Engineering and Materials Research Seminar Series
Power Spectrum and Machine Learning Analysis for Dried Blood Droplets Statistical and Quantitative Discrimination

As part of the School of Science and Technology's Engineering and Materials Research Seminar Series, Dr Lama Hamadeh, NTU presents: Power Spectrum and Machine Learning Analysis for Dried Blood Droplets Statistical and Quantitative Discrimination.
- From: Wednesday 16 January 2019, 1 pm
- To: Wednesday 16 January 2019, 2 pm
- Location: 015, CELS, Nottingham Trent University, Clifton Campus, Clifton Lane, Nottingham, NG11 8NS
- Download this event to your calendar
Event details
As part of the School of Science and Technology's Engineering and Materials Research Seminar Series, Dr Lama Hamadeh, NTU presents: Power Spectrum and Machine Learning Analysis for Dried Blood Droplets Statistical and Quantitative Discrimination.
Abstract
One of the most interesting and everyday natural phenomenon is the formation of different patterns after the evaporation of colloidal droplets deposited on a solid surface. The analysis of dried patterns resulting from biological liquids, such as blood, has recently gained a lot of attention, experimentally and theoretically, due to its potential application in biomedicine and forensic science. This paper presents an entirely novel approach to studying human blood droplet drying patterns. We took blood samples from 30 healthy young men before and after exhaustive exercise which is well known to cause large disturbances in blood chemistry. We objectively and quantitatively analysed 1800 dried blood droplet images by developing sophisticated image processing analysis routines and optimising a multivariate statistical machine learning algorithm. We look for statistically relevant correlations between the blood droplet drying patterns and exercise-induced changes in blood chemistry. An analysis of various measured physiological parameters is also investigated. We have seen that when our machine learning algorithm, which is an optimisation to a statistical model that combines Principal Component Analysis (PCA) as an unsupervised learning method and Linear Discriminant Analysis (LDA) as a supervised learning method, is applied on the logarithmic power spectrum of the images, it can provide a very good prediction accuracy, reaches to 95%, to discriminate the blood conditions, i.e., before or after physical exercise, provided we undertake an average over all the images taken per volunteer per condition.
This seminar is hosted by Dr Dave Fairhurst and Dr Ian Shuttlewoth
All Welcome
For any enquires please contact Dr Ian Shuttleworth
Location details
Room/Building:
Address:
Clifton Campus
Clifton Lane
Nottingham
NG11 8NS
Past event