The past ten years have seen two major socio-economic factors affecting the research field of handheld sensors, notably for health-related applications. The first factor is related to the continuous increase of the world population, which is reaching 7.5 billion this year.
If we look at the population structure across the developed countries, we find that around 12~20% of the population is over 65 years. The ageing and the growth of such a large population has fundamental consequences on the entire society, especially on medical care delivery system. According to statistical data from Organization for Economic Co-Operation and Development (OECD)5, in France, the maintenance of socio-health care system takes up nearly 11% of the annual Growth Domestic Product (GDP), which corresponds to about 230 billion of euros per yeara. In Taiwan, also the health sector takes up around 6.6 % of the GDP annually. In other words, a tremendous burden is now placed on our health care delivery systems due to the ever-growing population as well as longer life expectancy.
The second major and obvious trend is the booming of consumer electronic device market such as smart phones, feature phones and tablets. Around 70% of the world population owns at least one mobile devices. These devices are undoubtedly the most
a Considering that France has a GDP of 2133 billion in 2016.
doi:10.6342/NTU201700804
2
ubiquitous and powerful devices ever for integrating Point-Of-Care (POC) sensors.
Smartphones are already equipped with a number of integrated sensors and now offer a large processing power, but they also possess sufficient inputs and outputs for a wired or
wireless control of external sensors and data acquisition. For example, the smartphone audio jack (cf. Fig. 1), with proper electronic integration, serves as a 16 bits analog/digital converter (1 input, 2 outputs) - which is comparable to a number of performant acquisition cards – to extract and process data from sensors. In addition, the audio jack has synchronized data capture time resolution down to 26 µs per point, which is sufficient, in principle, to perform higher end fluorescence time measurements (such as FLINb), or to
b FLIN : Fluorescence-lifetime imaging microscopy Audio phone jack
Add-on Camera Wifi/internet:
Universal Serial Bus
Embedded system(s) Screen and Back LED
power Data
Fig. 1 Overview of smartphone I/O capacity
doi:10.6342/NTU201700804
3
extract amplitude and phase of signals modulated at relatively high frequency. To date, the audio jack channel has notably been used in Electrocardiogram (ECG) and pulse oximetry type of measurement6. For optical detection systems, we should also mention that smart devices often come with optical sources in the back of the phone (white LED for photo), as well as the front LED screen. Interestingly, the front LED may serve as light source having three emission wavelengths as shown in Fig. 2.
In some works, the light emitted from smart devices was used to carry our Surface Plasmon Resonance (SPR) detection7, which is one of the preferred method for bio-sensing. As for the optical detection side, the built-in smartphone CMOSc camera can also
c CMOS : Complementary metal–oxide–semiconductor. The CMOS sensor is now more common than the CCD camera in consumer electronic devices.
Fig. 2 Samsung Note 2 front LED panel emission spectrum
The spectrum is measured via a commercial UV-Vis spectrometer in National Taiwan University. The Samsung Galaxy note 2 is arbitrarily choose for demonstration.
doi:10.6342/NTU201700804
4
be used as a point detector for optical powermeter measurement4, 8, 9.
Finally, regarding connectivity, we should note that smartphones provide seamless connection to other embedded system through USB-OTGd or wireless access. Except for high power devices, the connection can also provide a stable power supply to small diagnostic devices.
Therefore, with a more and more equipped population facing health issues, the concept of m-healthe is emerging as a disrupting health care approach. The generalized use of smartphone as a smart platform for Point-Of-Care (POC) technology could certainly alleviate part of the healthcare costs and increase global disease prevention.
With POC diagnostic tools integrated in smartphone, a large portion of the population across the world and developing countries could benefit of a new, decentralized, medical access, in contrast with conventional health care delivery model. As a result, the cost of transporting medical resource to hospital and number of medical personnel can be reduced. This scenario is supported by reports from public sector. For example, the US Food and Drug administration recently mentioned in Mobile Medical Applications (2015):
~“The Food and Drug Administration (FDA) recognizes the extensive variety of actual and potential functions of mobile apps, the rapid pace of innovation in mobile apps,
d USB-OTG : USB On-The-Go. This common term specifies that the USB device performs both master or slave roles in establishing a communication link with another USB device via specific protocols.
e m-health stands for “mobile health” and typically refers to diagnostics or intervention done through smartphones or equivalent devices. The term e-health (electronic-health) is more generic.
doi:10.6342/NTU201700804
5
and the potential benefits and risks to public health represented by these apps.”
We have also seen World Health Organization stating that 10:
~“Mobile technologies have the potential to bridge systemic gaps needed to improve access to and use of health service, particularly among underserved populations.”
Beyond public sector, we have also been observing a fast growth in the m-health market11 with the emergence of new products.
To sum up, diagnostic tools traditionally centralized in laboratories and hospitals have the opportunity to be integrated onto mobile Consumer Electronic Devices (CEDs) to provide timely diagnostic assistance without geographic and temporal hindrance.
However, such goal cannot be simply realized without gathering cross-disciplinary skills and efforts to produce the required integrated technologic platforms. The sensor optical designs, the electronic readout, the software integration have to be considered carefully, and most importantly, relevant diagnostic scenario with identified end-users have to be proposed to promote the m-health approach.
In this thesis, we want to tackle part of these objectives by proposing smart integrated optical devices for mobile-based POC diagnostic. To achieve this goal, we combine the know-hows of the Micro-Sensor and System Laboratory of the biomedical department of National Taiwan University (NTU) and the Laboratory of Nano-optics and Optical Instrumentation (LNIO) of the Université de Technologie de Troyes (UTT). In the context of m-health, the constraints we set on ourselves it that the system should sensitive, very affordable, and versatile in term biologic targets. To achieve this goal, we focus on
doi:10.6342/NTU201700804
6
biosensors based on well-established Surface Plasmon Resonances and work on the system integration and performance improvement strategies beyond the state-of the-art for handheld devices.
In the first part of the thesis, we focus on building a DNA biosensor combining single wavelength colorimetry and digital Lock-In Amplifier (LIA) within a smartphone. In this section, we explore the possibility of utilizing audio jack in building a LIA-based optical biosensor. First, we give an introduction on the background of conventional method for integrating gold nano-particle colorimetry (AuNP) colorimetry with smartphone. A detailed literature review on smartphone based diagnostic tool is then given, which should serve to provide a general background for the thesis. The working principle of classical Lock-In Amplifier (LIA), the synthesis of AuNP, and the hardware /software integration are provided. We discuss the relevance of the smartphone audio-jack channel for conducting such measurement by evaluating its performances. Finally, at the end of this first part of the thesis, we demonstrate a 15-mer DNA detection within 15 minutes of diagnostic time down to 0.77 nM, using a software-based LIA.
In the second part, we take advantage of the developed LIA approach to explore phase-sensitive measurement of Surface Plasmon Resonance (SPR) biosensor for future portable application. As will be discussed, compared to traditional SPR, phase sensitive based SPR usually excels in terms of sensitivity. Such superior sensitivity can be highly
doi:10.6342/NTU201700804
7
desirable when the sensor is challenged with complex bio-sample (i.e., urine, saliva or serum). In this part, we present a novel phase sensitive detection design, named “Shearing Interferometry based Surface Plasmon Resonance” (SiSPR). SiSPR, together with a novel
phase extraction method, will be demonstrated. To show the feasibility of SiSPR in diagnostic application, we have targeted am atherosclerosis relevant protein—cardiac Troponin I (cTnI)—as potential biomarker. Performance of SiSPR, surface modification of the chip as well as a scheme for constructing laboratory prototype will be detailed.
doi:10.6342/NTU201700804
8