JMIR Publications

ABSTRACT

Background:

Cells concentration in body fluid is an important factor for clinical diagnosis. Traditional methods of analyzing cell concentration in the laboratory typically involve skilled clinicians manually preparing blood smears and counting cells under microscopes. However, this approach presents challenges in terms of expertise and labor intensity. While automated cell concentration estimation can be achieved using flow cytometers, their high cost limits accessibility, particularly in rural or lower-socioeconomic areas. Additionally, microfluidic systems, although cheaper than flow cytometers, still require expensive equipment such as high-speed cameras and syringe pumps to drive the flow and ensure video quality, further hindering accessibility. In this paper, we present SmartFlow, a low-cost solution for cell concentration estimation using smartphone-based computer vision on 3D-printed pump-free microfluidic platforms.

Objective:

The objective was to design and fabricate a microfluidics chip, coupled with clinical utilities, including smartphones and optical microscopes, for cells counting and concentration analysis. We also validated the following hypotheses: gravity can drive the flow in the microfluidics chips (hypothesis 1), bottleneck design on microfluidics chips had better performance on decreasing the flow speed to ensure video quality compared to straight microfluidics channel design (hypothesis 2), cells count and concentration can be estimated from smartphone-captured videos (hypothesis 3).

Methods:

Two experiments were conducted to validate three hypotheses. In cells flow velocity experiment, diluted sheep blood was used to flow through the microfluidics chips with and without bottleneck design to validate hypotheses 1 and 2. In cells concentration analysis experiment, sheep blood diluted into 13 concentrations flowed through the microfluidics chips respectively, while videos were recorded by smartphones for the concentration measurement.

Results:

In cells flow velocity experiment, we designed and fabricated two versions of microfluidics chips. The cells velocity was estimated by dense optical flow, and ANOVA test (Straight: F6, 99=6144.45, P<.001; Bottleneck: F6, 99=3475.78, P<.001) showed the height difference had significant impact on the cell’s velocity, which implied gravity could drive the flow. The video sharpness analysis demonstrated that video quality followed an exponential decay with increasing height differences (Video Quality =100e^{-k\ \times\ Height}) and a bottleneck design could preserve video quality effectively (Straight: R2=0.95, k=4.33; Bottleneck: R2=0.91, k=0.59). Samples among 13 cells concentrations were used for cell counting and cell concentration estimation analysis. The accuracy of cells counting (n=35, 60-second samples, R2=0.96, mean absolute error=1.10, mean squared error=2.24, root mean squared error=1.50) and cells concentration regression (n=39, 150-second samples, R2=0.99, mean absolute error=0.24, mean squared error=0.11, root mean squared error=0.33 on logarithmic scale and mean average percentage error=0.25) were evaluated by 5-fold cross-validation by comparing the algorithmic estimation to ground truth.

Conclusions:

In conclusion, we demonstrated the importance of the flow velocity in microfluidics system, and we proposed SmartFlow, a low-cost system for computer vision based cellular analysis. The proposed system could count the cells and estimate cells concentrations in the samples.