An image-to-answer algorithm for fully automated digital PCR image processing

Developing centrifugal force real-time digital PCR for detecting extremely low DNA concentration

Digital PCR (dPCR) is a technique for absolute quantification of nucleic acid molecules. To develop a dPCR technique that enables more accurate nucleic acid detection and quantification, we established a novel dPCR apparatus known as centrifugal force real-time dPCR (crdPCR). This system is efficient than other systems with only 2.14% liquid loss by dispensing samples using centrifugal force. Moreover, we applied a technique for analyzing the real-time graph of the each micro-wells and distinguishing true/false positives using artificial intelligence to mitigate the rain, a persistent issue with dPCR. The limits of detection and quantification were 1.38 and 4.19 copies/μL, respectively, showing a two-fold higher sensitivity than that of other comparable devices. With the integration of this new technology, crdPCR will significantly contribute to research on next-generation PCR targeting absolute micro-analysis.

SPEED: an integrated, smartphone-operated, handheld digital PCR Device for point-of-care testing

This study elaborates on the design, fabrication, and data analysis details of SPEED, a recently proposed smartphone-based digital polymerase chain reaction (dPCR) device. The dPCR chips incorporate partition diameters ranging from 50 μm to 5 μm, and these partitions are organized into six distinct blocks to facilitate image processing. Due to the superior thermal conductivity of Si and its potential for mass production, the dPCR chips were fabricated on a Si substrate. A temperature control system based on a high-power density Peltier element and a preheating/cooling PCR protocol user interface shortening the thermal cycle time. The optical design employs four 470 nm light-emitting diodes as light sources, with filters and mirrors effectively managing the light emitted during PCR. An algorithm is utilized for image processing and illumination nonuniformity correction including conversion to a monochromatic format, partition identification, skew correction, and the generation of an image correction mask. We validated the device using a range of deoxyribonucleic acid targets, demonstrating its potential applicability across multiple fields. Therefore, we provide guidance and verification of the design and testing of the recently proposed SPEED device.

An approach for integrating droplet generation and detection in digital polymerase chain reaction applications based on a bifunctional microfluidic cross-structure

Digital polymerase chain reaction (dPCR) is an approach for absolute nucleic acid quantification with high sensitivity. Although several successful commercial dPCR devices have been developed to date, further miniaturizing device dimensions, decreasing cross-contamination, and improving automation level are still research highlights. In this study, we developed a fully contamination-free dPCR detection chip with fluorescence flow cytometry and micro droplet approach. A bifunctional cross-structure (BFCS) was designed to realize monodisperse sample droplet generation in forward flow and droplet detection in backward flow with simple pneumatic control and fixed chip position. In order to improve droplet detection efficiency and accuracy, droplets morphology and sequence pattern during microfluidic droplet generation and backward flow droplet detection at the same cross-structure were observed and analyzed under different pneumatic pressures. In addition, during backward flow droplet detection, an optimized declination angle of the chip was applied to increase droplet reflux rates. For the validation of PCR performance, temperature changing processes during PCR cycles were achieved by heating the monodispersed droplet array with a customized PCR amplification device. The fluorescence signal of each droplet right after passing the cross-structure was excitated and detected. The absolute quantification ability of our integrated dPCR microfluidic chip utilizing flow fluorescence cytometry was tested and verified with Influenza A virus gene (from 7.5 copies/μL to 30000 copies/μL). Thus, our platform provides a novel and integrated approach for ddPCR analysis.

Facile prepared microfluidic chip for multiplexed digital RT-qPCR test

The chip-based digital polymerase chain reaction (PCR) is an indispensable technique for amplifying and quantifying nucleic acids, which has been widely employed in molecular diagnostics at both fundamental and clinical levels. However, the previous designs have yet to achieve widespread application due to limitations in complex chip fabrication, pretreatment procedures, special surface properties, and low throughput. This study presents a facile digital microfluidic chip driven by centrifugal force for digital PCR analysis. Interestingly, regardless of the hydrophilicity or hydrophobicity of the inner chip surface, an efficient digitization process can be achieved. PCR reagents introduced into the inlet can be allocated to 9600 microchambers and subsequently isolated by the immiscible phase (silicone oil). The centrifugal priming approach offers a facile means to achieve high-throughput analysis. The design was further employed for the quantification of nucleic acids using digital PCR. The calculated result exhibited a strong correlation with the measured value at the concentrations from 1 copy/μL to 1000 copies/μL (R2  = 0.99). Additionally, the chip also allowed digital multiplexed analysis, thereby indicating its potential for multi-target detection applications.

Smartphone integrated handheld (SPEED) digital polymerase chain reaction device

We demonstrate a smartphone integrated handheld (SPEED) digital polymerase chain reaction (dPCR) device for point-of-care application. The device has dimensions of ≈100 × 200 × 35 mm³ and a weight of ≈400 g. It can perform 45 PCR cycles in ≈49 min. The device also features integrated, miniaturized modules for thermal cycling, image taking, and wireless data communication. These functions are controlled by self-developed Android-based applications. The only consumable is the developed silicon-based dPCR chip, which has the potential to be recycled. The device's precision and accuracy are comparable with commercial dPCR machines. We have verified the SPEED dPCR prototype's utility in the testing of severe acute respiratory syndrome coronavirus 2, the detection of cancer-associated gene sequences, and the confirmations of Down syndrome diagnoses. Due to its low upfront capital investment, as well as its nominal running cost, we envision that the SPEED dPCR device will help to perform cancer screenings and non-invasive prenatal tests for the general population. It will also aid in the timely identification and monitoring of infectious disease testing, thereby expediting alerts with respect to potential emerging pandemics.

Compact wide-field femtoliter-chamber imaging system for high-speed and accurate digital bioanalysis

The femtoliter-chamber array is a bioanalytical platform that enables highly sensitive and quantitative analysis of biological reactions at the single-molecule level. This feature has been considered a key technology for "digital bioanalysis" in the biomedical field; however, its versatility is limited by the need for a large and expensive setup such as a fluorescence microscope, which requires a long time to acquire the entire image of a femtoliter-chamber array. To address these issues, we developed a compact and inexpensive wide-field imaging system (COWFISH) that can acquire fluorescence images with a large field of view (11.8 mm × 7.9 mm) and a high spatial resolution of ∼ 3 μm, enabling high-speed analysis of sub-million femtoliter chambers in 20 s. Using COWFISH, we demonstrated a CRISPR-Cas13a-based digital detection of viral RNA of SARS-CoV-2 with an equivalent detection sensitivity (limit of detection: 480 aM) and a 10-fold reduction in total imaging time, as compared to confocal fluorescence microscopy. In addition, we demonstrated the feasibility of COWFISH to discriminate between SARS-CoV-2-positive and -negative clinical specimens with 95% accuracy, showing its application in COVID-19 diagnosis. Therefore, COWFISH can serve as a compact and inexpensive imaging system for high-speed and accurate digital bioanalysis, paving a way for various biomedical applications, such as diagnosis of viral infections.

Deep Learning-Assisted Droplet Digital PCR for Quantitative Detection of Human Coronavirus

Since coronavirus disease 2019 (COVID-19) pandemic rapidly spread worldwide, there is an urgent demand for accurate and suitable nucleic acid detection technology. Although the conventional threshold-based algorithms have been used for processing images of droplet digital polymerase chain reaction (ddPCR), there are still challenges from noise and irregular size of droplets. Here, we present a combined method of the mask region convolutional neural network (Mask R-CNN)-based image detection algorithm and Gaussian mixture model (GMM)-based thresholding algorithm. This novel approach significantly reduces false detection rate and achieves highly accurate prediction model in a ddPCR image processing. We demonstrated that how deep learning improved the overall performance in a ddPCR image processing. Therefore, our study could be a promising method in nucleic acid detection technology.