The fever chip

Lead Research Organisation: University of Glasgow
Department Name: College of Medical, Veterinary, Life Sci

Abstract

Studentship strategic priority area:Healthcare device innovation
Keyword:Diagnosis, infection, silicon chip, immunoassay fever

Fever is an extremely common symptom of most infectious diseases as well as a range of non-communicable diseases too. In countries such as Uganda where malaria is widespread and the most frequent cause of fever, there is an unfortunate trend to assume anyone presenting with fever as having malaria without making a formal diagnosis. This practise is dangerous for several reasons. Firstly, there is a risk that a patient with fever diagnosed and treated for malaria is actually infected with another infectious agent. Not only will that patient be at risk due to a failure to take appropriate action against their specific infection, they will also act as a reservoir to transmit those diseases. Given the potentially explosive danger associated with some infections that first present as fever, e.g. Ebola virus, having easy and routine test for the virus could also enhance the speed at which such diseases are discovered, allowing the fastest possible implementation of public health programmes to stop their spread. Treatment of the patient with anti-malarials also introduces a risk of selection of anti-malarial resistance, since that person may actually become infected with malaria retrospectively when drug levels in their blood have diminished to concentrations not lethal to the malaria parasite (Plasmodium falciparum, or other Plasmodium species) and yet offering some selective advantage to parasites that carry mutations rendering them less sensitive to drug. Although good tests exist for malaria, these are often not employed due to cost and specialisation required in their use. For other infectious diseases, diagnostic approaches are often not well advanced. Even where they are, their use in resource-poor settings, like Uganda, is constrained.
We propose to develop a simple, cheap "Fever chip" capable of making differential diagnosis of malaria and an (expandable) range of other infectious diseases. The device will be integrated with smart phone reader technology enabling rapid interpretation of results at reference centres thus opening use to many health workers with different qualifications.
We have already developed prototypic devices that could successfully identify HIV and Hepatatis C infection.
Our platform is remarkably simple. Antigens that are specific for individual pathogens, and known to promote seroconversion in infected patients, are fixed to clearly demarcated domains on a CMOS chip surface. Serum, containing antibodies from patients, is then applied to the surface and if antibodies indicative of infection are present they bind to their cognate antigen. Once bound, secondary antibodies, with enzyme's linked to them, can bind to the primary antibody. The enzyme then converts a colorimetric substrate to a product that absorbs light emitted from narrow bandwidth LEDs. Photodiode sensors at the CMOS chip surface detect the change in photon penetrance and transmit this an electronic signal to our reading app built into a smart phone.
Below we outline a number of antigens that are either already in clinical use for serological diagnosis of malaria and a number of other diseases, or else are being investigated for potential future use. The list we give below is not comprehensive and the technology can be readily adapted to the inclusion of any new antigen that may be of use in identifying an infectious disease. We plan to initiate our work within a Ugandan context where we have close collaborators, but also emphasise that the basic platform can be adapted to include agents endemic in other localities.

Publications

10 25 50
 
Description The aim of this project was to develop a device which detects fever-related infectious diseases. The complementary metal oxide semiconductor (CMOS) based detection platform had already been developed before the beginning of this project. The method of detection employed by this device is optical -the chip is made of an array of 16x16 pixels, each having a photodiode sensor to measure any change in light transmission through the surface. Therefore, a reaction with a change in colour is required for the detection process.

A protocol for singleplex detection using silver precipitation was developed. Human African Trypanosomiasis (HAT) was the infectious disease chosen for the development of the detection protocol. The biomarkers, i.e. 4 recombinant antigens -two variant and two invariant surface glycoproteins- for HAT specific-antibody detection were chosen and developed. The expression of the recombinant antigens along with that of native antigens were tested through immunoassays carried on human serum obtained from HAT infected patients and controls from endemic regions. Both the native and the recombinant proteins separated well the sera from infected patients from that from controls. However, in comparison to these, the number of false positives was slightly higher in the recombinant antigens which could be explained by the remaining E.Coli contaminants in the samples.

The next step of the project was the surface functionalisation of the device in order to prepare it for antigen immobilisation. Because of the lack of sufficient chips for protocol development and optimisation, poly-methyl methacrylate (PMMA) slides which were placed on top of the chip surface were used as an alternative to directly using the chip surface. A surface functionalisation protocol was developed using (3-aminopropyl)triethoxysilane and glutaraldehyde which ensures the immobilisation of antigens on the PMMA surface.

Titrations with one of the recombinant antigens used for HAT detection, rISG65, were performed on the CMOS chip in order to develop and optimise the immunoassay protocol. The immunoassays were performed on the PMMA slides and, at the end, when silver had precipitated, the slide was placed on top of the chip and the light transmittance reaching the PDs was read. The results were not satisfactory, however, at the same time, results indicate some potential for diagnosis in this manner, because for every titration of the antigens/antibodies there was no reaction for the negative control and thus, no silver precipitate.

In terms of the signal processing, an algorithm for signal processing was developed. This was based on a machine learning algorithm, i.e. Sequential Monte Carlo particle filtering. In this case, the SMC particle filtering is used for identifying the position and intensity of a circle on the surface of the chip. This algorithm has proved to be effective even when background noise is high.
The chip has been previously proven to work effectively using enzymatic assays, thus enzymes could be used instead of antibody detection. In order to further explore the biomarker discovery for fever-related diseases, already available metabolomics datasets were analysed. The datasets consisted from HPLC-MS analysis of human serum from healthy controls and patients infected with fever related diseases (HAT, Zika, Malaria and Visceral Leishmaniasis). In order to be able to align all the datasets, the retention time drifts between the datasets were corrected using a machine learning algorithm based on Gaussian processing. This has also opened up the possibility of discovering biomarkers that different infectious disease have in common or are specific to a certain disease.
Exploitation Route In terms of the biosensor development, further work would be required in order for the device to be tested on field. However, this device has the potential for rapid diagnosis of infectious diseases. Areas which could benefit from this would be related to Healthcare and Medical Biotechnology.
The metabolomics alignment algorithm could be of use in the Bioinformatics field and biomarkers discovery, as this has proved to be working effectively across metabolomics datasets analysed on the same mass spectrometry platform. However, further evaluation of the algorithm is required.
Sectors Healthcare

Manufacturing

including Industrial Biotechology

Pharmaceuticals and Medical Biotechnology

 
Title Development of recombinant antigens related to Human African Trypanosomiasis 
Description Four recombinant antigens were developed (rISG65, rISG75, rLiTat1.3, rLiTat1.5) using E.Coli to be used for detecting HAT. 
Type Of Material Antibody 
Year Produced 2018 
Provided To Others? No  
Impact Expression of the above mentioned proteins was not very clear, so this method requires further optimisation. 
 
Title Image Data Analysis - Pattern Detection 
Description Python was used for developing an algorithm for the quantitative detection of the antigen-antibody reaction on the biosensor. The algorithm uses the principles of sequential Monte Carlo (SMC) particle filtering for identifying the position and intensity of a certain pattern on the image generated by the PD-CMOS chip. 
Type Of Material Computer model/algorithm 
Year Produced 2018 
Provided To Others? No  
Impact Improvements need to be made in order to be able to use it on a real image obtained after running an immunoassay. Currently, the SMC particle filtering is used for identifying the position and intensity of a circle. 
 
Title Metabolomics Retention Time Alignment Algorithm 
Description In order to identify other possible fever-related biomarkers, several metabolomics datasets were analysed. These datasets were combined in a meta-dataset using a machine learning algorithm based on Gaussian processing. 
Type Of Material Computer model/algorithm 
Year Produced 2020 
Provided To Others? No  
Impact Combining several metabolomics datasets in this way, may lead to the discovery of new potential biomarkers of a certain disease. 
 
Description C3Bio Training Workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact The workshop provided both theoretical and practical knowledge related to the field of biosensors. Around 30 people participated from several institutions around the world. Useful knowledge regarding micro-fluidics was obtained as a consequence of this workshop.
Year(s) Of Engagement Activity 2019
URL https://www.bath.ac.uk/events/c3bio-training-workshop-on-lab-on-chip/
 
Description Engineering lectures related to biosensors 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Undergraduate students
Results and Impact I have attended lectures related to biosensors offered by the School of Engineering from my university. This has helped me gain more knowledge in this field.
Year(s) Of Engagement Activity 2018
 
Description Institute of Infection, Immunity & Inflammation - Away Day 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact I have participated at the Institute's Away Day for which I've prepared a poster related to my PhD work. As a consequence of this event, I managed to engage with other postgraduate researchers and interchange ideas.
Year(s) Of Engagement Activity 2019