One of our major contributions to the biomicrofluidics field is a new method for trapping and manipulating micro- and nanoscale particles in solution, called the “hydrodynamic trap”. This is a new class of micro/nanomanipulation tool enabling particle confinement with extensive applications in nanoscience, biophysics, molecular and cellular biomechanics, and polymer physics [1, 2]. While various micro and nanomanipulation tools exist for studying molecular and cellular processes related to health and disease, many rely on force fields—optical, electrical, or magnetic—that are often not physiologically relevant or compatible with biomolecules or cells.

We developed a flow-based trapping and manipulation method which is gentle and much more suitable for studying biomolecules and cells [2, 3]. The hydrodynamic trap allows for capturing and observing single molecules (e.g., DNA) and single cells (e.g., bacterial cells) with ultra-high precision using sole action of fluid flow within a microchannel [4]. In this method, single micro and nanoscale particles are trapped at a stagnation point generated by two opposing fluid streams within a microfluidic junction [2]. Our method compares favorably with competing methods on many performance metrics such as particle size, tightness of confinement, manipulation capability and trapping duration [3].

Two unique features distinguish our method from the existing methods: i) The ability to “tweeze out” (trap and isolate) single particles from a concentrated suspension, ii) Precise control of the liquid medium surrounding the confined particle through fluid flow. Recently, we demonstrated the feasibility of simultaneously trapping and manipulating two particles using coupled planar extensional flows [5]. Our microfluidic confinement method has been widely adopted by numerous research groups. Hydrodynamic trap stands as a transformative technology, enabling researchers to embark on scientific studies that were previously not feasible.

We develop mobile biosensor technologies and novel miniaturized platforms for portable point-of-care diagnostic systems. One example is our microfluidic viscometer that utilizes measuring deflection of flexible micropillars for real-time viscosity measurements of bodily fluids such as blood [1]. This method allows for determination of viscosity within 2-100 cP with a sensitivity down to 0.5 cP. Using this method, we demonstrated shear-thinning behavior of human whole blood samples at various shear rates. Further, employing machine learning algorithms, we also demonstrated a biomedical application where we identified various stages of clot formation by classifying micropillar tip deflection as a measure of changing blood viscosity during blood coagulation [2]. To the best of our knowledge, this was the first study where machine learning algorithms were used to calculate the viscosity of fluids by analyzing fluid–structure interactions within a microfluidic device. Through computational fluid dynamics studies, we also demonstrated that the sensitivity and dynamic range of the device can be tuned by changing the device design including pillar geometry [3]. Our microfluidic viscometer is amenable to high-throughput, multiplexed viscosity measurements [4].

We are also developing technologies for isolation, detection and quantification of microorganisms. To this end, we developed a droplet-based microfluidic platform for high-throughput screening of gut microbes [5]. Traditional microbiological cultivation methods are limited by labor-intensive processes and biased sampling, hindering the representation of rare and slow-growing microbes. Our microfluidic platform addresses these issues by anaerobically isolating and cultivating microbial cells in picoliter droplets, enhancing the representation of diverse taxa and revealing antibiotic resistance profiles not detected by conventional plate-based methods. Additionally, we are developing a novel enumeration method for microorganisms in liquid samples. Traditional bacterial quantification methods rely on plating and counting colonies grown on agar medium, or counting bacteria in Petroff chambers, which is time-consuming, labor-intensive, and prone to error. Our technology is based on splitting the sample into tiny liquid compartments, and detecting and quantifying bacterial colonies within these compartments. This approach achieves better accuracy, faster turnaround time and higher throughput.

We are currently working on three projects involving the development of microfluidic platforms for studying mechanobiology and mechanotransduction at the molecular and cellular level:

1) We are developing a novel multifunctional mechanobiology platform to analyze how cells sense, respond to and integrate mechanical forces through cell-cell interactions. This technology combines high–throughput microfluidics for cell manipulation, 3D light sheet microscopy for imaging cellular and molecular dynamics, cell retrieval for omics analysis, and deep learning-based image analysis and modelling. To achieve these goals, we recently received funding through NSF Engineering – UKRI Engineering and Physical Sciences Research Council Lead Agency Opportunity (ENG-EPSRC) program, where we are collaborating with Dr. Michael Smutny and Dr. Till Bretschneider at the University of Warwick.

2) In collaboration with Dr. Anita Saraf at the University of Pittsburgh, we are developing a tissue-engineered model of congenital heart disease (CHD). The model is based on a bioreactor system that emulates in vivo intraventricular pressures and cyclic mechanical strains observed by the native myocardium. Using tissue-engineered model of CHD, we aim to elucidate the impact of cyclic mechanical strain on contractility and calcium dynamics in cardiac tissues in both physiological and pathological conditions and determine molecular and secretory changes in iPSC-CM constructs undergoing cyclic mechanical strain using proteomics and transcriptomics, towards identifying potential therapeutic targets.

3) In collaboration with Dr. Jerome Charmet and Dr. Fabrizio Albertetti at the University of Applied Sciences and Arts of Western Switzerland (HES-SO), and Dr. John Viator at Duquesne, we are developing a novel platform for photoacoustic mechanotyping of soft colloidal particles. The platform will enable capture, detection and classification of soft microscale colloidal particles based on their unique photoacoustic signatures. Specifically, our photoacoustic mechanotyping platform combines photoacoustics, advanced microfluidics, a novel ultra-sensitive acoustic transducer, and machine learning algorithms to relate photoacoustic response of soft colloidal particles – such as cells –  to their mechanical (and optical) properties.

We demonstrated, for the first time, lasing action in droplets doped with fluorescent dyes within a microfluidic channel [1,4]. We generated a stream of microdroplets in oil using a T-junction geometry and pumped the droplets with a coherent light source via an embedded optical fiber. We observed lasing action within the microdroplets based on bulk fluorescence emission from the dyes. The laser light frequencies can be fine-tuned by modifying morphological properties such as droplet size, relative refractive indices of droplet and oil and the choice of fluorophore. We demonstrated a biomedical application of this novel platform through label-free detection of single bacterial cells in liquid samples [2], and patented this technology [3]. Our microfluidic platform laid the groundwork for the development of droplet-based microlasers as a light source for biomedical lab-on-a-chip systems [4]. Our work represents a milestone in the development of miniaturized, tunable light sources for developing portable biomedical diagnostic devices.