Machine Learning and Computational Intelligence

We work on the development and application of machine learning and computational intelligence methods to address biomedical and engineering problems characterised by large volumes of complex data.

Research Overview

Large volumes of complex data is often a key feature of biomedical and engineering problems, including for example the understanding of cell behaviour and fate using high resolution microscopy data, the analysis of structural and functional integrity of concrete structures using non-destructive imaging data, the assessment of the medication state of Parkinson’s disease patients using high resolution multi-accelerometer data, or prediction of deformation of materials used in high-pressure industrial processes.

We use computational intelligence and machine learning techniques to address such problems. We are interested in applications of neural networks, support vector machines, reservoir computing, evolutionary optimisation, pattern recognition methods and other techniques to develop practical and validated solutions to hard biomedical and engineering problems. We also work on the theoretical development of machine learning techniques, for example on novel methods for syntactic pattern recognition, Bayesian interpretation of swarm optimisation, and new approaches to approximation of functions defined on high-dimensional spaces.

This figure is about automated recognition of cells using confocal imaging data.

We expect that our work will have significant impact in many areas. For example, our automated analysis of ageing of construction materials is expected to lead to much improved infrastructural building maintenance; our work on automated analysis of biomedical imaging data can speed up significantly the progress of medical research requiring the understanding of large volumes of images; and our work on syntactic image analysis may lead to improved novel ways of representing and storing images.


Research Lead :-

Members :-


  • Machine learning analysis of industrial data, KTP project in collaboration with XACT PCB Ltd, 2013-2015, GBP 153k – Prof Peter Andras (PI; Newcastle University).
  • Development of e-Science platform technology, EPSRC, 2006-2009, GBP 208k – Prof Peter Andras (CI; Newcastle University).
  • Data mining of eBay data, NStar proof-of-concept funding, 2006, GBP 90k – Prof Peter Andras (PI; Newcastle University).
  • Computational analysis of communications about chemical safety, DEFRA, 2004-2005, GBP 70k – Prof Peter Andras (CI; Newcastle University).
  • Syntactic pattern recognition (Dr Fletcher). The aim of this project is to develop a method for robust recognition of complex recursively structured geometric patterns, in the presence of noise, vagueness, occlusion, distortion and overlapping of patterns. For examples see

PhD Students


  • Mr Adam Wootton (supervisor: Dr Charles Day)
  • Mr Mohammed Al-Janabi (supervisor: Prof Peter Andras)


  • Ms Kanida Sinmai (superviosr: Prof Peter Andras; Newcastle University, 2015)
  • Dr Nils Hammerla (supervisor: Prof Peter Andras; Newcastle University, PhD, 2014)
  • Ms Charlotte Blackburn (supervisor: Prof Peter Andras, Newcastle University, MPhil, 2014)
  • Dr A. Ryad Soobhany (supervisor: Dr K.P. Lam, Keele University, PhD, 2013)
  • Dr John Butcher (supervisor: Dr Charles Day, Keele Univeristy, PhD, 2011)
  • Dr Kieren Lythgow (supervisor: Prof Peter Andras, Newcastle University, PhD, 2010)
  • Dr Shaun Fitch (supervisor: Prof Peter Andras, Newcastle University, PhD, 2008)
  • Ms Fatma Eldresi (supervisor: Prof Peter Andras, Newcastle University, MPhil, 2003)
  • Mr Ying Zhang (supervisor: Prof Peter Andras, Newcastle University,MPhil, 2004)

Past Research Associates

  • Dr Pablo Suau (supervisor: Prof Peter Andras; 2013-2014; Newcastle University; currently research associate at Newcastle University)
  • Dr Denis Besnard (supervisor: Prof Peter Andras; 2006; Newcastle University; currently research associate at Mines – Paris Tech, Toulouse)
  • Mr Jacques Chang (supervisor: Prof Peter Andras; 2006)
  • Dr Dominic Searson (supervisor: Prof Peter Andras; 2003-2004; Newcastle University; currently senior research associate at Newcastle University)


  • Professor Anand Pandyan
  • Professor Peter Haycock
  • Dr Ata Kaban (Birmingham University)
  • Dr Lehel Csató (Babes-Bolyai University – Cluj)
  • Mr Andrew Kelley (XACT PCB Ltd)
  • Professor Paul Watson (Newcastle University)
  • Dr Thomas Plötz (Newcastle University)


Andras, PE (2018) Cooperation in Repeated Rock-Paper-Scissors Games in Uncertain Environments. In: The 2018 Conference on Artifical Life (ALIFE 2018), 23-27 Jul 2018, Tokyo. (In Press)

Lam, K and Collins, D and Rigby, C and Bailey, J (2017) Towards Accurate Predictions of Customer Purchasing Patterns. In: IEEE Computer and Information Technology 2017, 21/08/2017-23/08/2017, Helsinki. (In Press)

Day, CR and Jabbar, SI and Heinz, N and Chadwick, EK  (2016) Using Convolutional Neural Network for Edge Detection in Musculoskeletal Ultrasound Images.  , 25-29 Jul 2016, Vancouver.  

Fisher, JM and Hammerla, NY and Ploetz, T and Andras, PE and Rochester, L and Walker, RW  (2016) Unsupervised home monitoring of Parkinson's disease motor symptoms using body-worn accelerometers.  Parkinsonism & Related Disorders.   ISSN 1873-5126 2015, 12-17 Jul 2015, Killarney

Andras, PE  (2015) High-dimensional function approximation using local linear embedding.  In: 2015 International Joint Conference On Neural Networks (LJCNN), 12-17 Jul 2015, Killarney.  

Hammerla, N, Fisher, J, Andras, P, Rochester, L, Walker, R, Ploetz, T (2015). PD disease state assessment in naturalistic environments using deep learning. Accepted for publication in the Proceedings of the AAAI – 2015.

Butcher, J.B., Verstraeten, D., Schrauwen, B., Day, C.R. and Haycock, P.W., 2014, "Defect detection in reinforced concrete using Reservoir Computing and Extreme Learning Machines", Computer-Aided Civil and Infrastructure Engineering, 29(3): 191-207 

Andras, P (2014). Function approximation using combined unsupervised and supervised learning. IEEE Transactions on Neural Networks and Learning Systems, 25: 495-505.

Sinmai, K, Andras, P (2014). Mapping on surfaces: supporting collaborative work using interactive tabletops. In Collaboration and Technology (Proceedings of the 20th International Conference on Collaboration and Technology – CRIWG 2014), Springer International Publishing, LNCS 8658, pp.319-334.

Smith WA, Lam K-P, Dempsey KP, Mazzocchi-Jones D, Richardson JB, Yang Y. 2014. Label free cell tracking in 3-D tissue engineering constructs with high resolution imaging. DYNAMICS AND FLUCTUATIONS IN BIOMEDICAL PHOTONICS XI (vol. 8942).

Dempsey KP, Richardson JB, Lam KP. 2014. On measuring cell confluence in phase contrast microscopy.IMAGING, MANIPULATION, AND ANALYSIS OF BIOMOLECULES, CELLS, AND TISSUES XII (vol. 8947)

Lam K, Collins DJ, Sule-Suso J, Bhana R. 2013. On evaluation of a multiscale-based CT image analysis and visualisation algorithm. In JX. Gao, DG. Xu & X. Sun (Eds.). IEEE Computer Society

Soobhany AR, Lam K,  Fletcher P. 2013. Source Identification of Camera Phones using SVD. IEEE Signal Processing Society

Lam K, Howden C, Liu L, Ding ZJ. 2013. Moments in Time: A Forensic View of Twitter. Beijing: IEEE Computer Society

Lam K, Smith WA, Collins DJ. 2013. Estimation of Depth Map Using Image Focus: A Scale-Space Approach for Shape Recovery. In T. Sobh (Ed.). USA: Springer New York

Butcher, J.B., Moore, H.E., Day, C.R., Adam, C.D., and Drijfhout, F.P., 2013, "Artificial neural network analysis of hydrocarbon profiles for the ageing of Lucilia sericata for post mortem interval estimation", Forensic Science International, 232(1-3): 25-31.

Butcher, JB, Day CR, Haycock, PW, Verstraeten, D, Schrauwen, B. 2013. DEFECT DETECTION IN REINFORCED CONCRETE USING RANDOM NEURAL ARCHITECTURES. Computer-Aided Civil and Infrastructure Engineering.

Butcher JB, Moore HE, Day CR, Adam CD, Drijfhout FP. 2013. Artificial Neural Network analysis of hydrocarbon profiles for the ageing of Lucilia sericata for Post Mortem Interval estimation. Forensic Science International, vol. 232, 25-31.

Butcher, J.B., Verstraeten, D., Schrauwen, B., Day, C.R. and Haycock, P.W., 2013, "Reservoir computing and extreme learning machines for non-linear time series data analysis", Neural Networks, 38: 76-89. 

Hammerla N, Kirkham R, Andras P, Ploetz T (2013). On preserving statistical characteristics of accelerometry data using their empirical cumulative distribution. Proceedings of the International Symposium on Wearable Computers (ISWC), pp.65-68.

Andras, P (2012). A Bayesian interpretation of the particle swarm optimization and its kernel extension. PLoS ONE, 7(11): e48710. doi:10.1371/journal.pone.0048710.

Yu, S-Y, Hammerla, N, Yan, J, Andras, P (2012). Aimbot detection in online FPS games using a heuristic method based on distribution comparison matrix.  In the Proceedings of the ICONIP’2012, pp.654-661.

Yu, S-Y, Hammerla, N, Yan, J, Andras, P (2012). A statistical aimbot detection method for online FPS. Proceedings of the IJCNN’2012, DOI: 10.1109/IJCNN.2012.6252489.

Lam K, Smith, WA, Collins, DJ. 2012. FORTHCOMING: Scalable 2-1/2D Reconstruction of Cell Objects.International Journal on Industrial Electronics, Technology and Automation

Lythgow, KT, Hudson, G, Andras, P, Chinnery, PF (2011). A critical analysis of the combined usage of protein localization prediction methods: increasing the number of independent data sets can reduce the accuracy of predicted mitochondrial localization. Mitochondrion, 11: 444-449.

Hammerla, N, Plötz, T, Andras, P, Olivier, P (2011). Assessing motor performance with PCA. In Proceedings of the International Workshop on Frontiers in Activity Recognition using Pervasive Sensing.

Soobhany AR, Lam K, Leary, R. 2011. On the Performance of Li’s Unsupervised Image Classifier and the Optimal Cropping Position of Images for Forensic Investigations. International Journal of Digital Crime and Forensics, vol. 1(3)

Austin JC, Day CR, Kearon AT, Evans DL, Haycock PW. 2010. Comparison method to differentiate between painted objects using polychromatic X-rays. INSIGHT, vol. 52(3), 140-143.

Butcher JB, Verstaeten D, Schrauwen B, Day CR, Haycock PW. Pruning reservoirs with random static projections.Machine Learning for Signal Processing (MLSP), 2010 IEEE International Workshop on (pp. 250-255). 

Austin JC, Day CR, Kearon AT, Haycock PW. 2009. Single element mapping in radiography. X-RAY SPECTROMETRY, vol. 38(6), 492-504.

Day CR, Austin JC, Butcher JB, Haycock PW, Kearon AT. 2009. Element-specific determination of X-ray transmission signatures using neural networks. NDT & E INTERNATIONAL, vol. 42(5), 446-451.

Malhotra S, Pandyan AD, Day CR, Jones PW, Hermens H. 2009. Spasticity, an impairment that is poorly defined and poorly measured. Clin Rehabil, vol. 23(7), 651-658.

Andras, P (2009). Molecular neuroimaging – A proposal for a novel approach to high resolution recording of neural activity in nervous systems. Medical Hypotheses, 73: 876-882.

LAM KP and FLETCHER P. 2009. Concurrent grammar inference machines for 2-D pattern recognition: a comparison with the level set approach. In JT. Astola, KO. Egiazarian, NM. Nasrabadi & SA. Rizvi (Eds.). Image Processing: Algorithms and Systems VII (vol. SPIE vol. 7245). SPIE-IS&T. 

Butcher JB, Lion M, Day CR, Haycock PW, Hocking MJ. 2009. A Low Frequency Electromagnetic Probe for Detection of Corrosion in Steel-Reinforced Concrete. In M. Grantham & C. Majorana (Eds.). Concrete Solutions(pp. 446-451). CRC Pres

LAM KP and FLETCHER P. 2009. Concurrent grammar inference machines for 2-D pattern recognition: a comparison with the level set approach. In JT. Astola, KO. Egiazarian, NM. Nasrabadi & SA. Rizvi (Eds.). Image Processing: Algorithms and Systems VII (vol. SPIE vol. 7245). SPIE-IS&T

Lam KP and Emery R. 2009. Image Pixel Guided Tours: A Software Platform for Non-Destructive X-ray Imaging.IMAGE PROCESSING: ALGORITHMS AND SYSTEMS VII (vol. 7245).

Hinks, J, Bush, J, Andras, P, Garratt, J, Pigott, G, Kennedy, A, Pless-Mulloli, T (2009). Views on chemical safety information and influences on chemical disposal behaviour in the UK. Science of the Total Environment, 407, 1299-1306.

Austin JC, Day CR, Kearon AT, Valussi S, Haycock PW. 2008. Using polychromatic X-radiography to examine realistic imitation firearms. Forensic Sci Int, vol. 181(1-3), 26-31.

Austin JC, Day CR, Kearon AT, Valussi S, Haycock PW. 2008. Characterisation of metallic powder impregnated pastes using polychromatic X-radiography. INSIGHT, vol. 50(10), 550-553.

Fitch, S, Jackson, TR, Andras, P (2008). Unsupervised segmentation of cell nuclei using geometric models. In Proceedings of 5th IEEE International Symposium on Biomedical Imaging – From Nano to Macro, pp.728-731.

Lam KP, Austin JC, Day CR. 2007. A coarse-grained spectral signature generator - art. no. 63560S. Eight International Conference on Quality Control by Artificial Vision (vol. 6356, p. S3560)

Lam KP, Austin JC, Day CR. 2007. A coarse-grained spectral signature generator - art. no. 63560S. Eight International Conference on Quality Control by Artificial Vision (vol. 6356, p. S3560)

Lam KP. 2007. Towards a practical differential image processing approach of change detection. INNOVATIVE ALGORITHMS AND TECHNIQUES IN AUTOMATION, INDUSTRIAL ELECTRONICS AND TELECOMMUNICATIONS(pp. 229-234)

Nehmzow U, Iglesias R, Kyriacou T, Billings S. 2006. Robot Learning Through Task Identification. Robotics and Autonomous Systems, vol. 54, 66-778.

Kyriacou T, Akanyeti O, Nehmzow U, Iglesias R, Billings S. 2006. Visual task identification using polynomial models. Proc. Towards Autonomous Robotic Systems, Taros

Andras, P and Idowu, O (2005). Kohonen networks with graph-based augmented metrics. In Proceedings of WSOM’2005, pp.179-186.

Sandor, Z. and Andras, P. (2004). Alternative Sampling Methods for Estimating Multivariate Normal Probabilities. Journal of Econometrics, vol.120, no.2.,  207-234.

Andras, P. (2002). The Equivalence of Support Vector Machine and Regularization Neural Networks. Neural Processing Letters, vol.15., no.2., pp.97-104.

Andras, P. (2002) Spectrum-based Design of Sinusoidal RBF Neural Networks. In Proceedings of the International Joint Conference on Neural Networks 2002, pp. 1421-1426.

Fletcher P. 2001. Connectionist learning of regular graph grammars. CONNECTION SCIENCE, vol. 13(2), 127-188.

Andras P. (2001) RBF Neural Networks with Orthogonal Basis Functions. In: Howlett, R.J. & Jain, L.C. (Eds.) Radial Basis Function Networks 1. Recent Developments in Theory and Algorithms, Physica-Verlag, Heidelberg, pp.67-94.

Ndlovu P and Preater J. 2001. Calibration using a piecewise simple linear regression model. Communications in Statistics: Theory and Methods, vol. 30(2), 229-242

Fletcher P. 2000. The foundations of connectionist computation. CONNECTION SCIENCE, vol. 12(2), 163-196.

Gibbens RJ, Whittle P, Ansell P, Bather J, Collins EJ, Gittins J, Zervos M, Bauerle N, Birge JR, Frostig E, Weiss G, McDiarmid C, Mohring RH, Uetz M, Schulz AS, Owen RW, Gregorio-Dominguez MM, Preater J, Rieder U, Stidham S, Thomas L, Yao DD. 1999. The achievable region approach to the optimal control of stochastic systems - Discussion. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, vol. 61, 776-791

Andras, P. (1999). Orthogonal RBF Neural Network Approximation. Neural Processing Letters, vol.9., no.2.,  pp.141-151.

Andras, P. (1999) Approximation of Chaotic Shapes with Tree-structured Neural Networks. In Proceedings of IJCNN '99, Washington, D.C., IEEE Press, paper #187.

Lam KP. 1999. A component-based design for parallel moment generators. PARALLEL AND DISTRIBUTED METHODS FOR IMAGE PROCESSING III (vol. 3817, pp. 137-145).

Lam KP. 1998. High-performance thresholding with adaptive equalisation. PARALLEL AND DISTRIBUTED METHODS FOR IMAGE PROCESSING II (vol. 3452, pp. 148-157).

Lam KP and Furness A. 1997. K-AVE+GNN+Sobel equals an effective, highly parallel edge detector approach.PARALLEL AND DISTRIBUTED METHODS FOR IMAGE PROCESSING (vol. 3166, pp. 190-198).

Lam KP. 1997. UHC - A massively parallel and distributed realisation of hierarchical classifier networks. THIRD INTERNATIONAL SYMPOSIUM ON PARALLEL ARCHITECTURES, ALGORITHMS, AND NETWORKS, PROCEEDINGS (I-SPAN '97)

Lam KP and Furness A. 1996. On parallelization of neural classification algorithms. SECOND INTERNATIONAL SYMPOSIUM ON PARALLEL ARCHITECTURES, ALGORITHMS, AND NETWORKS (I-SPAN '96), PROCEEDINGS(pp. 337-340).