A commentary on the development and use of smartphone imaging devices

Abstract

Current-age smartphones are known for their wide array of functionality and are now being utilized in the field of healthcare and medicine due to their proven capabilities as smartphone imaging devices (SIDs). Recent technical advancements enabled the integration of special add-on lenses with smartphones to transform them into SIDs. With the rising demand for efficient point-of-care (PoC) devices for better diagnostic applications, SIDs will be a one-stop solution. Additionally, portability, user-friendliness and low-cost make it accessible for all even at remote locations. Furthermore, improvements in resolution, magnification and field-of-view (FOV) have attracted the scientific community to use SIDs in various biomedical applications such as disease diagnosis, food quality control and pathogen detection. SIDs can be arranged in various combinational setups by using different illumination sources and optics to achieve suitable contrast and visibility of the specimen under study. This Commentary illustrates the various illumination sources used in SID and also spotlights their design and applications.

Introduction

Smartphones have been the subject of a great deal of development in recent years. Their inclusion in point-of-care (PoC) devices is becoming increasingly popular. PoC devices are used to perform diagnostic tests on or near the site of patient care (Vashist et al. 2015). These devices are important to public and global health as they not only help to reduce costs but are also more time-efficient and enhance the quality of healthcare in remote areas. Existing PoC devices have helped in the timely and accurate detection of tuberculosis (TB) and human immunodeficiency viruses (HIV) (Ojha et al. 2020; Zimic et al. 2009). For developing countries without significant healthcare budgets to support expensive diagnostic machines, such PoC devices are crucial (Banik et al. 2021a).

PoC devices are structured and designed in a way where the specimen collection is facilitated along with the analytical and post-analytical processing. Early PoC devices required peripheral devices to be attached to perform these functions. Recent developments in smartphones have overcome such previous requirements since they are equipped with powerful chipsets that have great processing power and imaging capabilities (McCracken and Yoon., 2016). Smartphones are becoming increasingly popular, even in areas that lack resources such as proper healthcare facilities. They come equipped with an advanced camera setup, microphones and powerful processors that help in the rapid processing of data, and ample storage can help with storing diagnostic data. Features like Wi-Fi, Bluetooth and Global System for Mobile (GSM) communication can help in the high-speed transfer of collected data across different platforms. These are some of the reasons smartphone-focused PoC is gaining popularity around the world (Xu et al. 2015). Smartphone imaging devices (SIDs) have been developed as substitutes for conventional modes of imaging and related diagnostic tests with SIDs used in the detection of bacteria, viruses and food spoilage (Pawar et al. 2023; Kozel et al. 2017; Buchanan et al. 2022; Xiao et al. 2022; Lu et al. 2019). SID can be improved by substituting parts with better alternatives able to obtain results comparable to those obtained via conventional methods while reducing the overall cost of the device. One such component that has been subject to several variations in different SIDs is the light source. There is a lot of debate about the ideal light source for such devices. An ideal source of illumination should be uniform without any chromatic aberration, as these have been seen to degrade the quality of the images resulting in unwanted glare and shadows with false coloration that leads to unreliable results during disease diagnosis (Skandarajah et al. 2014). Scientists have also tried using illumination from a wide number of sources such as light emitting diode (LED), LASERs and smartphone displays as well as an electroluminescence panel (Kheireddine et al. 2019a; Spigulis 2017). Recently, there have been efforts to explore the possibility of light sources with various configurations. As the number of publications increase, the difficulty in understanding the significance of each light source also increases, thus necessitating a comprehensive description. In this article, we have compared the different light sources used by various SIDs and looked at each of their biomedical applications along with their advantages and disadvantages. In preparing for this searches were performed in the database Pubmed in April 2023. Three search terms were used during the search: “Smartphone Imaging Device” “LED” and “Diagnosis” in the field of title, keywords, and abstract. The options for “Advanced Search” were utilized to exclude reviews, book chapters, letters, reports, conference proceedings and foreign language articles. The search was restricted to studies dated from the year 2013 to 2023. In the end, a total of 40 studies were chosen for data extraction and analysis in this review. The workflow for selecting eligible articles is depicted in Fig. 1.

Fig. 1.

PRISMA diagram showing the selection process of eligible articles using keywords

Different light sources used for SIDs

When a smartphone’s imaging potential is used for purposes other than traditional photography, the smartphone is referred to as an SID. The modern smartphone has an advanced camera setup, an inbuilt sensor, a detector and many more features. Due to this new and improved hardware and software, modern smartphones can now be utilized as bio-medical detection devices, scanning devices, microscopic devices, analytical detectors, pollution monitors and even educational tools. They can be used in areas where the need for rapid and efficient diagnosis is high and the manpower is limited. They are affordable and can be converted into laboratory-grade microscopic devices with some technique incorporation. There are various sources of illumination used in SIDs (refer Table 1).

Table 1.

Different illumination sources used in various SIDs and their respective advantages, disadvantages and applications.

Source	Advantages	Disadvantages	Application	References
LED	Versatile Cheap Long-lasting Low heat dissipation	Broad spectral range	Fluorescence microscopy	Kim et al. (2015)
			Mapping skin chromophores	Spigulis et al. (2017)
			Hg2+ detection in aqueous samples	Shan et al. (2019)
			Petrographic microscope	Febo et al. (2021)
			View and interact with micro-organisms in an educational setting	Kim et al. (2016)
LASER	Offer greater temporal coherence Highly monochromatic High irradiance	Expensive, require high maintenance	PDT assistance	Liu et al. 2019; Hempstead et al. (2015)
			Virus detection	Priye et al. (2017)
			Quantifying the morpho-physiological features of epithelial tissues	Sung et al. (2017)
Smartphone screen	Offers uniform illumination Cost-effective	Data security, software updates	Multi-modal microscopy	Kheireddine et al. (2019a); Kheireddine S. et al. (2019b)
			Ptychographic microscope	Lee et al. (2021)
EL panel	Uniform illumination Portable Easy-to-use	Low intensity	Bright-field microscopy	Banik et al. (2021a)

LED

LEDs are very popular due to their efficiency, cheap cost and great lifespan (Nair and Dhoble 2015; Pode 2020). Also, they use 90% less power than incandescent bulbs. The light emitted by LEDs is directional, i.e. only emitted in a specific direction unlike incandescent bulbs, where light is emitted in all directions. A smartphone’s built-in flashlight utilizes a LED and this can be used for taking high quality images of microscopic structures with or without external lenses for magnification. Single white LED can be used in certain settings, whereas arrangements of multiple LEDs can be used for more specialized requirements. Indeed, combinations of LEDs can be used to make up red-green-blue (RGB) arrays for illumination. LEDs can also be coupled with filters to carry out fluorescence imaging (Toslak et al. 2017). A simple yet powerful “smartphone microscope”, developed with an external lens setup attached to the camera, was developed to address the insufficiency of experimental instruments for students (Grier et al. 2018; Spigulis et al. 2017). The experimental setup used an external add-on lens of the simple glass bead type and an LED ‘push’ light that consists of three white LEDs. Magnification depends on the diameter of the glass beads with two glass beads used to serve the purpose of two lenses, i.e. objective and eyepiece. In total, three magnification clips were available providing a magnification of 100×, 300×, and 780×. However, the resolution and clarity of the image diminished with increasing magnification power of the attachment (Hunt et al. 2021). Spatial resolution was significantly affected by the source of illumination and the spherical aberrations of the glass bead being used. Another downside was that the device image focus was required to be adjusted by hand (Hergemöller and Laumann et al. 2017).

LED SIDs have been used in a variety of medical applications (Hempstead et al. 2015; Hernández-Neuta et al. 2019). The CellScope is a smartphone-based PoC microscope that is used as a supplement in enhancing diagnostic and research capabilities (Saeed and Jabbar. 2018). The original device’s single source of LED illumination is replaced with a programmable domed LED array (Phillips et al. 2015). With all the conventional imaging modes, this upgraded device allows simultaneous multi-contrast imaging (Senarathna et al. 2019). This new apparatus also enables scanning across illumination angles for acquiring light field data points, which in turn are utilized in generating three-dimensional intensity and phase images without requiring any hardware changes (Liu et al. 2020). This virtual refocusing technique can be used for 3D imaging or software-only focus correction, effectively removing the need for precise mechanical focusing during field experiments. It was demonstrated that the device’s capabilities are comparable to those of a commercial microscope using a variety of samples and objective magnifications (Phillips et al. 2015). Another important device development was the LuduScope, a SID made by integrating a smartphone into a device that not only would detect microorganisms but also enable the user to interact with them (Kim et al. 2016). This was possible with the use of a microfluidic chamber that contained a photosensitive organism (Euglena gracilis), a single-celled protozoan. It exhibited negative phototaxis upon being exposed to light. This reaction was quick and visible in seconds which enabled the user to interact with the organism without much lag. The source of illumination was a LED-operated to provide ambient lighting, enough to not cause any phototaxis. Four directional LEDs were placed in such a position that they all faced the stage (as shown in Fig. 2b). The LEDs were controlled with the help of a joystick that was in turn controlled by the user who could use it to operate the LED lights at any intensity.

Fig. 2.

Shows the typical design of LudusScope and how it turns observational microscopy into an interactive experience by accessing open-ended interaction with microorganisms. (Figure reproduced from Kim et al. 2016 with permission from PLoSONE).

A smartphone-based LED array microscope was used to achieve multimodal imaging with various contrast modes and Rheinberg illuminations using a light source composed of 37 LEDs, which led to improved image contrast during live cell imaging (Ogasawara et al. 2018). By changing the colour combinations of the Rheinberg illumination, images of living chromatic structures were able to be produced, building a foundation for high-contrast microscopy (Ogasawara et al. 2018). Decomposition of the image into RGB colour scale, with post-acquisition computational analysis enabled multi-contrast imaging in a single shot using a colour-coded LED array as a light source and a colour image sensor (Lee et al. 2015). Jung et al. have designed a transportable multi-contrast microscope with the ability to produce bright-field, dark-field and differential phase contrast images of tissue samples using a smartphone. This microscopy is based on the “colour-coded light-emitting-diode microscopy (cLEDscope)” imaging concept, in which a sample is illuminated with a cLED array, and the light passing through the specimen is recorded by a colour image sensor (Sirisathitkul et al. 2023). To turn a smartphone into a multi-contrast imaging device, a supplementary apparatus with a patterned colour micro-LED array, specimen stage, and miniature objective is required (Jung et al. 2017). The simple installation of this module on a smartphone facilitates the capability to perform multi-contrast imaging of transparent specimens (Lee et al. 2015). In addition, a smartphone application (an ‘app’) was developed to extract images, perform the necessary computations and display the multi-contrast images in real time. For smartphone-based multi-contrast imaging, this offers a simpler and more economical alternative.

Diagnostics is another biomedical field that has seen an increased SID application (Zhang et al. 2020; Xu et al. 2015). The main goal of applying SIDs in the diagnostic field is to provide an alternative to traditional methods of disease diagnosis that are based on fabrication of PoC devices that are cost-effective and reliable. Viruses like Zika (ZIKV), chikungunya and dengue are dangerous and their numbers are increasing globally (Cao-Lormeau et al. 2016; Sukhralia et al. 2019; Ning et al. 2021; Sanyaolu et al. 2019). A smartphone-based diagnostic device was fabricated to carry out tests for ZIKV, chikungunya and dengue virus using an RGB excitation source controlled via Bluetooth. Reactions are carried out in PCR (polymerase chain reaction) tubes kept on an isothermal heater that provides a constant temperature for 40 min. A smartphone app monitored the reaction in real time and collected data on the illumination of the samples irradiated by the RGB source (Zarei, 2017) with analysis of those signals allowing for virus detection in the sample.

SIDs have also been developed for quantification of morpho-physiological features of epithelial tissues, by use of a blue LED coupled with a filter and a condenser to carry out fluorescence imaging [20]. Used in conjunction with such fluorescence imaging SID apparatus application of deep learning approaches have enabled automatic detection of sickle cell disease (Ilyas et al. 2020; Knowlton et al. 2015; de Haan et al. 2020). In such endeavours image clarity is the key to specific and errorless detection, hence the light source is an important factor (Ilyas 2020). To enhance magnification, an external objective lens can be incorporated (de Haan et al. 2020). Kim et al. contributed to PoC applications by developing a smartphone-guided fluorescence microscope using the lens of a phone camera. The low cost, light weight and mechanically simple device was used to image quantum dot fluorescence on a glass slide (Sung et al. 2017). In a parallel development, a smartphone was also used to make a high-resolution, wide-field ‘multi-modal’ microscopic imaging system, using the LED flash as an optical source and the built-in camera for image acquisition, with this setup used for the imaging of microbeads and other biological samples (Rabha et al. 2021).

Meng et al. developed a hand-held, smartphone-derived quantitative phase microscope (Meng et al. 2017). Phase recovery was carried out using a self-made Android application which computed multiphase intensities by resolving Poisson’s equation that produced precise high-contrast images (Ma, 2017). Using this device, the following were successfully imaged: RBC (red blood cells) smear, pap smear, monocot root and broad bean epidermis sections (Meng et al. 2017). Smartphone confocal microscopes have been shown to be effective in acquiring 2D image slices within a z stack through a combined use of a slit aperture and diffraction grating (Kulkarni et al., 2021). Such approaches have signalled the true potential of smartphone-based confocal microscopes for reconstruction of complex cellular structures like spinous, papillary dermis and basal keratinocytes in great detail thereby potentially facilitating disease diagnosis in resource-limited settings (Freeman et al. 2018).

A LED-based smartphone device able to perform photodynamic therapy (PDT) treatment was developed by Hui et al. (2018). This device could efficiently irradiate target sites intra-orally, resulting in the death of target cells, using a fibre-coupled LED as an irradiation source. The wavelength spectrum of LEDs is broader compared to that of a LASER, hence an increased efficiency of photoactivation per unit of optical power is expected with a LASER. However, this is true under the assumption that the source wavelength matches the peak absorption wavelength of the target cells. In cases where the peak absorption band is very narrow, the broader spectral spread offered by the fibre-LED system is preferred Kim et al. (2016). A SID was developed with the help of a ball lens attached to a smartphone camera, and a LED was used as the illuminating source (as shown in Fig. 3). This device was used to image a variety of micro-fibrous materials. The setup was rather simple and the focus was done free-hand. The FOV of the imaging system was dependent on the ball lens used as it was directly proportional to the ball lens diameter, with image quality decreasing with an increase in the diameter of the ball lens. Due to the spherical shape of the lens, there was a drawback, the blurred outer regions on the image plane leaving only the central part focused. However, for a thin sample, the image quality was found to be comparable to a conventional bright-field microscope (Aryal et al. 2021).

Fig. 3.

Image showing different components of a smartphone microscope. a A ball lens-mounted substrate. b Smartphone with a ball lens module attached. c Optical setup of the smartphone microscope (Figure reproduced from Aryal et al., 2021 with permission from Wiley).

A petrographic microscope is an important imaging tool that is frequently used to observe optically anisotropic minerals present in thin sections of the sample (Cady et al. 1986; Dias et al. 2020). Petrographic microscopy can be an asset in geological sciences, archaeology and material sciences (Vandenabeele et al. 2014; Valentino et al. 2020). These microscopes use costly objective lenses and require additional structures for imaging purposes which typically limit their usage outside of large facilities. A smartphone cast within a suitable physical support can be used as a petrographic microscope (Febo et al. 2021). The source of illumination in the smartphone-based petrographic microscope is multiple LEDs which are connected to a potentiometer to control the intensity to achieve the best results while capturing images. Petrographic microscopes require the presence of a polariser before the specimen, and a second polariser that is placed after the specimen, 3-D printed parts are designed that hold the second polariser. BLIPS ultra-lens (SmartMicroOptics, Italy) is an additional polymeric lens that attaches itself to the smartphone camera and enhances the magnification of the device (Banik et al. 2020). The stage was structured in such a way that the sample could be rotated 360° mimicking the conventional petrographic microscope. The system developed was comparable to the conventional petrographic microscope operated at a low (~4×) magnification setting and could be easily be operated in a resource-limited setting (Di Febo et al. 2021).

Historically, fluorescence microscopes have used mercury gas discharge lamps as excitation light sources (Young et al. 2004). Despite being considered a standard for fluorescence microscopic application, these lamps have some drawbacks such as the requirement for filter sets to prevent UV damage to the sample (i.e. sensitive cells or tissues) and short lamp lifespans of 200–300 h, with lamp intensity gradually decreasing over time (Ladouceur et al. 2021). Furthermore, due to the requirement for stable illumination intensity, lamps must be left on for hours prior to the performance of fluorescent measurements, wasting a significant portion of their lifespan (Flesch, 2006). LED lamps provide illumination at very specific wavelength bands from the UV to infrared, and they show great promise as alternative light sources for fluorescence microscopy. Current ‘white light’ LEDs, on the other hand, are designed to emit a broad excitation spectrum that can be filtered to provide particular wavelength bands, similar to mercury lamps. Additionally, there is no UV emission generated, the LEDs achieve full brightness in microseconds; they glow with constant intensity and are relatively cool (Robertson et al. 2009).

Brightfield (BF), darkfield (DF) and phase contrast (PC) are the most common optical microscopy modes (Liu et al. 2014). Brightfield imaging is best used to study samples with maximum absorption (Garini et al. 2006). Because darkfield imaging only captures high-angle scattered light, it provides good contrast for sub-resolution features (Nellist et al. 2000). For unstained and transparent tissue specimens, phase contrast is used to visualise shape and density variations (Liu et al. 2014). A LED array microscope is capable of enacting these diverse imaging modes (Būtaitė et al. 2022). With regard to generating image contrast, the positional set of LEDs on the array corresponds to a set of illumination angles because each LED can be governed separately to illuminate the sample (Dutta Gupta and Jatothu, 2013). LEDs are cheap and long-lasting, making them a good choice when selecting an illuminating source for a SID (Sung et al. 2017). The broad spectral range of LEDs is one of the few drawbacks, which can be overlooked in some cases considering the cheap costs (Hempstead et al. 2015).

Lasers

Lasers are an expensive alternative to LEDs in fluorescence imaging SIDs, but they do offer greater temporal coherence, narrow band wavelength and can focus light on very tiny spots, achieving high irradiance. Laser diodes are incorporated within the device at certain angles which ensure the best possible results (Haley and Pratt et al., 2017).

Smartphone-based fluorescence microscopes are the most common type of SIDs that contains a LASER as an illumination source. Such a device made by Kim et al. utilized a smartphone along with an external lens (Kim et al. 2016). SIDs that carry out fluorescence microscopy are not as common as SIDs that carry out bright-field microscopy as this ability depends on the illumination source and usage of the filter (Huang et al. 2021). Through fluorescence microscopy, one can track and analyse biological molecules in a non-destructive manner (Kim et al. 2015). Fluorescence microscopy has been performed using a smartphone for the detection of Hg²⁺ ions (as shown in Fig. 4) (Shan et al. 2019). This SID, developed by Shan et al. (2019) was complemented with a mobile application for image processing purposes. The illumination source was a 405-nm LASER diode that was tilted to illuminate the sample at a specific angle in order to reduce the signal-to-noise ratio. The smartphone camera aided by a 20× micro-objective was used for imaging purposes. The add-on objective lens ensured proper magnification to achieve desired results. This was possible due to an eyepiece that was used to connect the smartphone camera to the micro-objective lens. To eliminate any scattered excitation light, an emission filter was positioned in front of the smartphone camera. A 3-D printed design was prepared, to give shape to the smartphone fluorescence microscope. The SID does not require an external power supply since the LASER diode is operated on batteries. This enables the device to be portable, user-friendly and used in remote, resource-limited settings (Banik et al. 2020; Shan et al. 2019). The process of detecting Hg²⁺ in the aqueous solution involved labelling the Hg²⁺ solution with fluorescent probes. The sensitivity of a smartphone-based fluorescence microscope can be determined by the minimum number of molecules detected per diffraction-limited spot using DNA origami nanobeads and pre-specified number of fluorophores. When the performance of colour and monochrome sensors integrated into cutting-edge smartphones was compared, the results revealed that a smartphone’s monochrome sensor can achieve higher sensitivity than a number of alternative optical instruments (Vietz et al. 2019).

Fig. 4.

Smartphone-based fluorescence microscope. A The optical design of the smartphone fluorescence microscope. B System photo and design sketch of the smartphone fluorescence microscope. (Figure reproduced from Shan et al., 2019 with permission from Elsevier).

SID-based fluorescence microscopy has also been used as a detection tool for mycotoxins, whose diagnostics are of greater importance since their infection to mankind through contaminated food and water results in disastrous public health hazards. This detection platform was developed by the amalgamation of multicolour up-conversion nanoparticle barcode technology with a portable smartphone-based image processing unit. After indirect competitive immunoassays using encoded signals in the form of UCNMs (multi-coloured up-conversion nanoparticle-encoded microspheres), images could be captured by the portable device and the camera of a smartphone (Trofymchuk et al. 2021). In less than 1 min, an HSV-based image recognition android application analysed the images and provided a reliable and accurate result (Yang et al. 2018). In general, LASER usage in SIDs is primarily reserved for carrying out fluorescence imaging (Hernandez-Neuta et al. 2019). Apart from the mentioned devices, a few others have also been developed for different imaging purposes. Zhu et al. developed a SID that made use of LASER as the illumination source for illuminating the sample treated with a fluorophore. LASERs are always used to illuminate the sample at an angle (Zhu et al. 2013) with the inclination angle varying for different devices, e.g. 70, 61° (Zhu et al. 2013; Muller et al. 2018). An emission filter is used in order to prevent scattered light from being detected by the sensor. Specificity of the wavelength of the light emitted from the LASER makes it a popular choice as an illuminating source in SIDs that are developed for the sole purpose of fluorescence imaging. Spigulis et al. carried out the mapping of skin chromophores by using triple-wavelength LASER as a light source (as shown in Fig. 5). If skin chromophores are distributed abnormally, it clinically indicates the presence of bruises, burns and various other kinds of skin ailments. The mapping of skin chromophores is insightful information about a skin-related pathology, its progression and/or healing processes. The devices that have been designed, which are commercially available to carry out skin chromophores, consume a lot of time and are expensive, with the need to be coupled with a computer that limits the use of the technique in resource-limited areas and field conditions. On the contrary, the one developed by Spigulis et al. is compact even though it uses an arrangement of six LASER modules as the illumination source. An accessory diffuser was intricated to ensure a consistent three-wavelength illumination of the target. The design of the platform was kept flat and made sticky with the use of double-sided tape. This ensured the use of different smartphone models with different dimensions with the device. The resolution of the device was found to be approximately 0.1 mm (Spigulis et al. 2017).

Fig. 5.

a 3-D model of the three-wavelength illumination unit and b the mobile prototype with a smartphone on it. (1) Laser modules (three pairs, 448–532–659 nm), (2) shielding cylinder, (3) beams collector, (4) diffuser of LASER light, (5) sticky platform, and (6) electronics compartment. (Figure reproduced from Spigulis et al., 2017 with permission from SPIE).

SmartDHM is a smartphone-based digital holographic microscope that is both portable and affordable. It takes an interferogram and uses an Android application to compute phase information using the reference conjugated hologram method without the need for a server or a computer (Goud et al. 2019). SmartDHM is used for real-time acquisition because smartphones have enough frame rates to capture interferograms. This led to miniaturization and low-cost potential mobile health care (Goud et al. 2019). Despite being expensive and requiring high maintenance, LASERs have a variety of benefits when used as a light source for SIDs. They are capable of high powers, and irradiance and are highly monochromatic.

Alternative light sources

Smartphone screen

A SID built by Kheireddine et al. used the illumination from the screen of a second mobile phone device for imaging purposes (Kheireddine et al. 2019b). The phones used were an Apple iPhone 6 for illumination and a Nokia Lumia for imaging. The iPhone’s featured retina display consists of RGB LEDs in high concentration which allows for the projection of different illumination patterns onto the specimen, thus creating different illumination modes (as shown in Fig. 6). The imaging setup additionally uses an external lens which was obtained from an iPhone 5s to provide a 2× magnification providing a theoretical spatial resolution of 1.3 μm.

Fig. 6.

Applications to imaging cells with various illumination patterns. a Diatoms and E. gracilis images zoomed in and cropped under bright-field (BF), dark-field (DF), and Rheinberg (RI) illumination patterns (second and third rows). In the top row, these illumination patterns were created using the Retina display. Used ring illumination patterns with single dark-field (DF-single) and double dark-field (DF-double) (second and third columns). On a dark specimen background, different types of RI were prepared: blue-yellow patterns (fourth column), and half-rings with multiple colors (red-green half-ring pattern) (fifth column). b Images of stained (methyl blue) and unstained human epithelial cheek cells (HECC) under BF and DF illumination patterns, zoomed in and cropped. Images from the second column were subtracted using a single dark-field (DF-single) illumination pattern. (Figure has been reproduced from Kheireddine S. et al., 2019 with permission from RSC publication).

To generate different patterns of illumination, Microsoft PowerPoint was used on the Apple iPhone. The different patterns displayed on the screen (shown in Fig. 6) could be employed for a completely different microscopy configurations. For BF microscopy a range of patterns ranging from a completely white screen to a small full white circle displayed on a black background could be employed. For DF microscopy, a ring of small size was created just outside the imaging lens” numerical aperture (NA). Concentric circles were drawn on a black background to make DF-single, double and triple. A similar technique was used for refractive index (RI) imaging, but instead of a white circle, a ring of contrasting colours against a different colour specimen background was used. The colours chosen are sample-specific, intending to eliminate the need for specimen staining. In this setup, rings were used which had their inside area coloured. The rings were further sectioned into halves for a two sector-pattern or into quarters for a four-sector pattern. It was observed that with the halogen illumination, the setup was able to resolve 406.4 lines pair per mm and 362.0 lines pair per mm with the retina display illumination. This indicates that both illumination sources provide a comparable spatial resolution of below 2 μm (Kheireddine S et al. 2019a). Smartphone illumination was also used to develop a smartphone-based Fourier Ptychographic microscope (Kheireddine et al. 2019b). Since the display screen of a smartphone is being used, the illumination could be programmed according to the requirement. The smartphone’s computational ability was used for the reconstruction of the image. A low-cost external lens and 3-D printed parts were used to fabricate the device. Microscopy imaging took place through the smartphone’s built-in camera, and the computational power of the smartphone processors was used for image reconstruction. An Android application that was developed specifically for this SID was used to store and process images recovered from the smartphone in areas where network coverage was unavailable (Lee et al. 2021). A smartphone-mounted computational microscope can achieve wide-field and high-resolution imaging using the Fourier ptychographic (FP) microscopy technique (Lee et al. 2021). Ambient illumination on a smartphone can also be used as a light source. The method is based on shadow imaging, which involves placing the sample on the image sensor’s surface and capturing direct shadow images under illumination. Pixel super-resolution is achieved by capturing multiple images at various lighting angles (Lee and Ynag et al., 2014). A smartphone screen is used mainly because the screen of the smartphone consists of RGB LEDs that are densely packed. This not only provides uniform illumination but also displays various complex spatial patterns (Kheireddine et al. 2019a). The intensity can be graded across the screen and different colours can be displayed as well. All of these options can be explored without much expertise and more importantly, without the use of complex and costly electrical appliances. The smartphone screen employed for illumination purposes can also serve as a stage where the sample can be kept for observation (Huang et al. 2021; Rezazadeh et al. 2019).

EL panel

Electroluminescent (EL) material emits visible light when a strong electric field or electrical current is applied. It is a well-accepted phenomenon in electronic applications, and EL units that use a high electric field can emit light through the impact of high-energy electrons in a semiconducting material (Moretti et al. 2016). It is important to note that one of the electrodes of the EL should preferably be transparent for a working and efficient EL panel. Indium tin oxide, better known as ITO, fixed on either plastic or glass is one of the most widely used electrodes in EL panels. This is mainly due to its transparency along with its high conductance rate (Fig. 7). The intensity of the light can be increased by increasing the supplied voltage until it reaches saturation, though the required voltage for EL panels is generally between 100 and 1000 V (Nair et al. 2021).

Fig. 7.

The phenomenon of electroluminescence. (Figure has been reproduced from Moretti et al., 2016 with permission from Elsevier).

Development of an EL panel as an illumination source was spurred by the degradation of image quality due to non-uniform illumination. Uniform illumination by an EL panel can serve as an alternate source of illumination and may replace traditional LEDs (Pode et al. 2020). Attempts were made to design a SID with an EL panel as the illumination source (Banik et al. 2021a). The basic structure was fabricated with the help of acrylic sheets with a BLIPS lens (SmartMicroOptics, Italy) attached to the smartphone camera (Banik et al. 2021b). Fig. 8 shows the images of different parts of pumpkin (a and c) and human skin section (d) captured using BLIPS lens-intricated SID. By combining an OLED display with microarray technology, a commercial flat-panel display was created. Despite being a portable and disposable design, this enabled clinically sensitive high-density fluorescent, programmable and multiplexed biorecognition. By sandwiching the fluorescent layer responsible for biorecognition between the emitter and photodiode, OLED display technology eradicates the need for magnifying optics (Katchman et al. 2016).

Fig. 8.

Images acquired with the developed smartphone platform. a Pumpkin leaf (image sizes — 4000 × 3000 pixels). b Flea (image size — 2922 × 3120 pixels). c Pumpkin stem section (image sizes — 2988 × 3222 pixels). d Human skin section (image size — 2688 × 2992 pixels). (Figure reproduced from Banik et al., 2021b with permission from Elsevier).

Conclusion

The cost-effectiveness and time-efficient nature of SIDs as PoC devices have led to significant growth in their demand. Research has proven them to be reliable when used with the correct hardware. The quality of the image taken by SIDs depends not only on the camera and lens quality but also on the source of illumination. There has been a spike in reports on available light sources which has led to difficulty in comparing them individually. In this article, the major light sources have been summarized with medical, technical and even agricultural applications for better understanding. LEDs are the most preferred sources of illumination, due to the fact that they are cheaper than the other sources, more versatile and longer lasting. LEDs arranged in different patterns act as illuminating sources for imaging. Filters, along with certain LEDs, can be employed in fluorescence microscopy. LEDs, though cheap, have a few drawbacks, for example, a wide spectral emission range than laser light. Lasers are also used in a few SIDs but only to carry out fluorescence imaging. They operate at a fixed wavelength illuminating the sample at an angle to reduce the signal-to-noise ratio. They are effective and efficient; however, their price is on the higher side compared to other illumination sources. A smartphone screen is another alternative available illumination source. The screen can be used to display almost any pattern which is a requirement for different types of imaging. It is also a cheap and effective alternative with wide-ranging applications. A fairly new source is known as an EL panel, which emits light through a phenomenon known as electroluminescence. It is a uniform source of light that is cheap and rugged. This light source helps in removing the problem of non-uniform illumination of the sample. Options of exploring other sources that have not been used, e.g. high-density OLED display can be used and tested for its efficacy. SIDs may be upgraded in future by incorporating other emerging technologies such as micro-LED displays.

Author contributions

Gagan Raju and Aashrayi Ranjan planned and designed the concept and wrote the manuscript. Soumyabrata Banik and Ashmini Poddar designed the figures and illustrations. Gagan Raju, Aashrayi Ranjan and Ashmini Poddar compiled the data for the study. Vishwanath Managuli and Nirmal Mazumder supervised the study.

Funding

This work was supported by the Indian Council of Medical Research, grant number ITR/Ad-hoc/43/2020-21, ID No. 2020-3286; the Department of Science and Technology, Government of India, India, grant number GST/DST/TWN/P-95/2021; and the Science and Engineering Research Board (SERB), Government of India, India, project number-SERB/MTR/2020/000058.

Data availability

Datasets used in this work will be available from the corresponding author on reasonable request.

Declarations

Ethics approval

Not required in this review.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The authors declare no competing interests.