KP Lam

Biography

I undertook my BSc (London) in Electronic Engineering (EE) specialising in computer architectures and digital systems design. After graduation, I have worked there as an RA  (sponsored by CLCC/Plessey Ltd.) before joining the Computing Laboratory at the University of Kent (UKC) under the UK High Performance Distributed Systems Initiative, where I also received my research doctorate in 1994. I was admitted as an IEE corporate member and a Chartered Engineer (CEng) in 1995.

I took up a lectureship (EE) at Keele in autumn 1994 and, after receiving my SEDA HE teacher training award (Distinction), I moved to Computer Science (CS) during 1997/98. My teaching interests have spanned the two subject disciplines from the traditional teaching of signal processing, through telecommunications and computer vision, to the more recent developments in advanced computer architectures, communication networks and digital security, distributed/Cloud systems, digital image processing and multimedia, object-oriented programming, (statistical) data analysis and analytics.

Research and scholarship

For over two decades, my research has developed and applied cutting edge data analysis to a range of visual analytics and static data problems. Given the high computational demands of visual analysis and the scale of data in current problems, I also researched large-scale computation; specifically, the design and development of parallel computing algorithms in distributed/cloud-based applications. The impact of my research has been illustrated previously by my work presented as the basis of both the UoA-11 (Computer Science & Informatics) case studies (3A & 3B), plus one of the two UoA-B15 (General Engineering) submissions by the (then) ISTM/Keele in the 2014 REF. To date, my research track record in collaborative research endeavours include knowledge transfer with a wide range of external/industrial partners summarised as follows:

2013-2017: Spatiotemporal Biometrics for Stem Cell Specific Cellomics - This 4-year BBSRC CASE Award investigated key challenges in developing effective, safe and predictable new cell-based therapies pioneered at the RJAH Orthopaedic Hospital for patients with osteoarthritis and spinal cord injuries. Our results have shown the importance of high-quality dynamic measurements of individual cells and crucially, their spatiotemporal interactions that would allow stem cell characterisation beyond traditional population studies in-vitro¹. The work sought to assist the analysis of data obtained from the MRC-funded clinical trial (ASCOT), which evaluates the therapeutic role of bone marrow-derived mesenchymal stromal cells (BMSCs) in the autologous chondrocyte implantation (ACI) treatment recently approved by NICE.

Following the recent completion of the ASCOT trial (2022/23), further work is currently underway to develop practical semantic analysis tools that will fully exploit the spatiotemporal imaging datasets generated by the longitudinal clinical studies at RJAH. These tools seek to critically assess computationally tractable and biologically meaningful metrics relating to the functional outcomes of cell transplantation; in particular to understand why MSCs+chondrocytes co-cultures were previously shown to display cellular interaction mechanisms indicative of the circadian rhythm and, importantly, the extent to which this correlates with the functional outcomes following transplantation.

2016-2018:  Value Creation with Big Data -  This 30-month KTP project concerned the development of a cloud-based Business Intelligence System (BIS) which sought to enhance the marketing data of the business partner’s clients. This was achieved through integration with big-data sources and a variety of ML-based (big) data analytic tools, including data sanitisation and advanced statistical analytics, contemporary machine learning (ML) and data mining technqiues, to augment client data of the company with market intelligence.

2017-2019: Characterisation of the pyloric rhythm of cancer pagurus stomatogastric ganglion with voltage-sensitive dye imaging (VSDI) -  This 3-year Keele funded PhD study entailed the first-ever development of computational signal processing procedures that sought to extract the pyloric rhythm directly from the VSDI data, enabling an accurate identification of the individual neurons in the pyloric circuit. A multi-resolution procedure based on the sequential Singular Spectrum Analysis (s-SSA) was first constructed as a non-parametric time-frequency analytic tool which separates the pyloric rhythm from the extremely noisy VSDI recordings (at -40dB to -15dB), allowing the accurate detection of potential pyloric neurons via the computed spectra of the respective cells/neurons. To facilitate identifying the different pyloric neurons, computationally tractable and biologically meaningful biometrics were devised to characterise each cell/neuron in terms of its computed spectrum. More importantly, the signal-processing methodology had demonstrated its applicability to a broad spectrum of biological/cellular systems, where the VSDI technology can be used to characterise the electrical activity as measured by cell membrane potentials from which vital/functional information about such cells is obtained. To this end, work is currently underway in collaboration with Yorkshire Farben GmbH (part of the Yorkshire Group Ltd.) to develop a mesoscopic visualisation platform that analyses at single-cell resolution the optical recordings captured in real-time by the VSDs in exactly the same way as fluorescence imaging of cells and tissue.

2019 - 2020: The Alan Turing Institute (ATI) Data Challenge – This EPSRC sponsored pilot study applied data science methods to integrate a large collection of clinical gait analysis datasets, captured using a variety of gait measurement technologies including 3D video motion recordings. The combined dataset comprises data from multiple collaborating research labs/institutions in the UK including Warwick, Imperial College, Keele/Oswestry and Cardiff.  Following the successful conclusion of this study at the ATI last December, I have been working as part of an interdisciplinary team, with external collaborators, on a joint EPSRC and ATI project that seeks to predict progression of the (knee joint) osteoarthritis disease using such an integrated dataset, by developing novel machine learning techniques/models which have the potential to disrupt conventional osteoarthritis management impacting clinical care and disease progression.

2018 - : Automated Analysis of Paediatric Sleep Studies with Big Data techniques – This on-going research/project (funded by the NSMI) accesses the large multichannel cardiorespiratory recording (CRR) archive at Great Ormond Street Hospital (GOSH) in London, with investigations led by a multidisciplinary project team consisting of Keele/CS academics, consultant respiratory paediatricians and physiologists at GOSH.  Analysis of the GOSH data is expected to reveal the complex relationships between the recorded data and the AHI-based historic diagnosis that will permit (a) tentative automated diagnoses to be made and (b) identification of the minimal set of factors requisite to making accurate diagnoses.  By developing large scale data analytics tool which seeks to automate the currently highly complex and laborious assessment of sleep-disordered breathing in children, the project presents unique and exciting challenges/opportunities to work on the most complete archive of paediatric polysomnography with CRR prepared by leading respiratory paediatricians at GOSH.

From 2019 onwards, I have been appointed as an Academic Adviser (CS, IT & Informatics) for the Commonwealth Scholarship Commission to review and assess scholarship applications the Commission offers every year. Appointed by the Secretary of State for Foreign, Commonwealth and Development Affairs (FCDO), the Commission publishes a list of Advisers in its annual report, which is tabled in Parliament and on a dedicated page of its website at https://cscuk.fcdo.gov.uk/academic-advisers/

Teaching

• CSC-40031 Web Technologies and Security
• CSC-40039 Cloud Computing
• CSC-40042 Statistical Techniques for Data Analytics

Research themes