This PDF file contains the front matter associated with SPIE Proceedings Volume 12832, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

A scalable method of fabricating microscale devices in a low resource environment is shown using 3D printing. Microdevices are typically fabricated using silicon microfabrication techniques that require high resources, such as a cleanroom, that inhibits device fabrication in low resource environments. In this work, the use of 3D printing to make microfluidic devices for particle sorting, PCR detection and low-loss integrated waveguides is reviewed. The demonstrations are all performed in a low resource environment, without the use of a cleanroom, with an inexpensive custom 3D printer and off-the-shelf resin. The microdevices are made within a few minutes with training at the sophomore undergraduate level. This demonstrated a scalable fabrication method that is inexpensive, quick and facile.

Measuring scattering samples’ absolute optical properties is valuable in bio-medicine, agriculture, material characterization, and beyond. These measurements may be achieved by analyzing the sample’s frequency-domain diffuse reflectance or transmittance. However, successfully achieving these absolute measurements is complicated by the need for calibration. We present a calibration-free method to conduct these measurements. This method, dubbed dual-ratio, creates a measured data type that cancels most coupling and calibration factors that confound traditional reflectance or transmittance measurements, specifically multiplicative factors associated with source emission, detector efficiency, or optical coupling with the sample. Furthermore, we have applied the dual-ratio method to measure absolute optical properties of a small-volume sample (i.e., the size of a standard cuvette). Applications include tissue hemodynamics and oxygenation assessment (brain, muscle, etc.), water turbidity and chemical analysis, food quality determination, and more. This work builds on our previous work developing the dual-ratio method for a cuvette-sized sample volume. We also expand on other previous work, combining frequency-domain and continuous-wave measurements to achieve absolute broadband absorption spectra. Optical properties recovered by small-volume dual-ratio agree well with semi-infinite medium multi-distance scanning, which we consider the gold standard. Such calibration-free methods may make sample quantitative analysis more accessible and allow for easy quantitative measurements outside the traditional laboratory setting.

Digital radiographic imaging systems are becoming more widely deployed in low-resource settings, potentially reducing the inequitable access to medical imaging that persists today. Even when resources are made available to install digital x-ray equipment, challenges remain with respect to ongoing maintenance and recommended quality assurance programs. Recent studies have indicated that a significant fraction of radiographic installations in Africa are not assessed at recommended intervals due to the lack of high-cost x-ray exposure meters. As a result, errors in x-ray exposure parameters (e.g. current, exposure time, or kilovoltage) can lead to suboptimal image quality, repeated exams, and unnecessary radiation exposure to patients and staff. We have developed a low-cost solution for routine x-ray quality assurance measurements, which takes advantage of commercial-off-the-shelf (COTS) electronic components integrated with a low-power microprocessor controller. The device employs four sensitive phototransistors connected to a multiplexed 16-bit analog-to-digital converter. Light input to the optical sensors is provided by a rare-earth phosphor screen (Lanex regular), which emits green light under exposure to x-rays. Estimation of spectral properties is enabled by the use of aluminum filters on two of the four photosensors. Acquisition is controlled by an open-source microprocessor (Arduino Nano 33 BLE) and the total cost of all components is less than $100 USD. Comparisons against a commercial sensor indicate that the optical-based measurements of x-ray exposure are linear and accurate to within ±3%, over the range from 60 to 140 kVp. This project demonstrates the feasibility of providing accurate, robust solid-state x-ray exposure measurements in low-resource settings.

Introduction: Low-cost optical microscopes are under development by several groups to bring point-of-care malaria diagnosis services to under-served communities in low-resource settings. We describe a Fourier ptychography microscope (FPM) system based on an open-source design and a method of measuring spatial resolution in terms of the optical modulation transfer function (MTF) using images of the 1951 USAF resolution test device.

Methods: The FPM system uses a Raspberry Pi computer and 196-LED matrix light source. FPM images were reconstructed using in-house Python code. The MTF was determined from bar-patterns using a least-squares analysis to fit a square wave, including odd-harmonic terms, to image profiles through the bars. These were normalized to larger uniform regions of the pattern and combined to generate the MTF.

Results: Total component cost of the FPM was less than $200. The theoretical diffraction limit imposed by the pupil function of this system was 280 cycles/mm, slightly less than the measured MTF 10% frequency of 300 cycles/mm. The 10% frequency in FPM images was 550 cycles/mm.

Conclusions: The USAF test pattern provides a practical method for assessing FPM performance in terms of the achieved MTF of the $200 Raspberry Pi based Fourier ptychography microscope. The limiting frequency in FPM images, 550 cycles/mm, was slightly less than the Nyquist sampling cut-off frequency of 670 cycles/mm imposed by pixel spacing.

Environmental enteric dysfunction (EED) is a subclinical disorder of intestinal function common in tropical countries and settings of poverty and economic disadvantage. EED manifests during infancy and is associated with undernutrition, poor sanitation, and gut infections. EED is characterised by inflammation, reduced absorptive capacity, and reduced barrier function (i.e., increased permeability) in the small intestine. The precise mechanisms underlying changes in gut barrier function (and other aspects of intestinal function) in EED remain elusive. Furthermore, current diagnostic methods to assess gut permeability (e.g., endoscopic biopsies or permeability assays such as the Lactulose:Mannitol test) are invasive, unreliable and/or challenging to perform in infants and patients with other coexisting urological conditions. Consequently, there is an urgent need to develop diagnostic technologies that can non-invasively and affordably monitor intestinal permeability in low-resource settings where EED is prevalent.

To address this need, we present a prototype semi-wearable, wireless sensor for non-invasive assessment of intestinal permeability via transcutaneous fluorescence spectroscopy. The approach relies on the ingestion of a fluorescent contrast agent (fluorescein) and the subsequent detection of its permeation from the gut into the bloodstream using a wearable probe. We outline the development of the semi-wearable sensor and report preliminary in vivo deployment. This showcases the potential of transcutaneous fluorescence spectroscopy as a wearable and non-invasive diagnostic tool for assessing gut function in low-resource settings.

Many biologics such as vaccines are temperature sensitive and must be stored and transported at refrigerated/freezing temperatures. This need for temperature-controlled storage and distribution is especially challenging in low-resource settings. The goal of this project is to develop a new processing method, light-assisted drying (LAD), to dehydrate biologics in preparation for long-term dry-state storage at ambient temperatures to reduce or eliminate the need for cold storage. This study explores strategies for further expediting LAD processing times and for the simultaneous processing of multiple samples.

Early identification of conditions that can lead to blindness is critical for saving vision. The optical red-reflex test (RRT), which assesses the light reflections from the back of the eye, is a key exam for identifying adverse eye conditions in very young children. However, healthcare workers generally learn the RRT using peer practice and do not have the opportunity to observe abnormal reflexes, especially for rarer conditions, during training. The light reflections also differ in appearance between populations due to different pigmentation levels, so effective training requires practice with a diverse population. We have developed a set of 3D model eyes that aim to accurately mimic the response of eyes with varying pigmentation levels in the RRT, both for healthy eyes and pathologies that can be identified using the RRT. We characterized the optical properties of a set of full-color 3D printing materials (a white scattering material and four transparent colors - cyan, magenta, yellow and black). These properties were used to determine the number of layers, layer thicknesses, and color and scattering material combinations needed to match the reflectance of different fundi, given the constraints of the 3Dprinter. The model eyes can be used as an inexpensive tool for training a wide variety of health professionals to recognize abnormal reflections from the eye and as a reference standard for developing or calibrating eye screening instruments and tools.

To address an increasing demand for accessible and affordable tools for at-home oral health assessment, this paper presents the development of a low-cost intraoral camera integrated with a deep learning approach for image analysis. The camera captures and analyzes images of soft and hard oral tissues, enabling real-time feedback on potential tooth staining and empowering users to proactively manage their oral health. The system utilizes an Azdent intraoral USB camera with the Raspberry Pi 400 computer and Intel® Neural Computing Stick for real-time image acquisition and processing. A neural network was trained on a dataset comprising 102,062 CIELAB and RGB values from the VITA classical shade guide. Ground truth annotations were generated through manual labeling, encompassing tooth number and stain levels. The deep learning approach demonstrated high accuracy in tooth stain identification with a testing accuracy exceeding 0.6. This study demonstrates the capacity of low-cost camera hardware and deep learning algorithms to effectively categorize tooth stain levels with high accuracy. By bridging the gap between professional care and homebased oral health monitoring, the development of this low-cost platform holds promise in facilitating early detection and monitoring of oral health issues.

The rapid and accurate detection of infectious diseases like HIV, COVID-19, and TB is crucial for effective public health initiatives. We demonstrate a transformative approach in TB detection through a simulated Loop-Mediated Isothermal Amplification (LAMP) assay. This computational model was trained using established LAMP protocols and evaluated under varying conditions, including template and primer concentrations and cycling conditions, showing robust performance. Simulated results were used to optimise the LAMP protocol, reducing the time and resources required. The study opens avenues for cost-effective, rapid, and precise diagnosis of infectious diseases. Future work will focus on expanding the model to other infectious diseases and integrating it into real-world diagnostic workflows.

Timely and effective detection of COVID-19 continues to be a critical aspect in managing the pandemic and setting the groundwork for future pathogen detection techniques. Present diagnostic technologies, especially those designed for POC (Point of Care) use in resource-constrained settings, must meet the evolving needs for rapidity, accuracy, and affordability. Within this context, preparing for pandemics necessitates the development of swift POC diagnostics. This research investigates the potential of the Arduino-driven LAMP (Loop-Mediated Isothermal Amplification) method for SARS-CoV-2 detection. An inexpensive, accessible, and portable "LAMP box" is proposed in this study, integrating the LAMP assay with an Arduino to create a potentially effective detection system for COVID-19. The device features a wireless technology network connecting all components. The reaction is performed at 65°C on an isothermal heating pad located beneath a solid aluminum base, providing a stable platform for reaction tubes and assays. Control of the entire system, powered via a 5-volt USB source, is provided through a built-in Bluetooth connection linked to an Arduino, facilitating control through a computer. The study concludes that such simple systems can effectively determine the presence or absence of artificial SARS-CoV-2 genetic material in samples. The presented research underscores the promise of Arduino-based LAMP technology, showcasing its specificity and sensitivity compared to the conventional RT-PCR (Reverse Transcription Polymerase Chain Reaction).