10 March 2020 in Blood Cells, Molecules, and Diseases
29 November 2021 in PLOS Computational Biology
Sickle cell disease, a genetic disorder affecting a sizeable global demographic, manifests in sickle red blood cells (sRBCs) with altered shape and biomechanics. sRBCs show heightened adhesive interactions with inflamed endothelium, triggering painful vascular occlusion events. Numerous studies employ microfluidic-assay-based monitoring tools to quantify characteristics of adhered sRBCs from high resolution channel images. The current image analysis workflow relies on detailed morphological characterization and cell counting by a specially trained worker. This is time and labor intensive, and prone to user bias artifacts. Here we establish a morphology based classification scheme to identify two naturally arising sRBC subpopulations—deformable and non-deformable sRBCs—utilizing novel visual markers that link to underlying cell biomechanical properties and hold promise for clinically relevant insights. We then set up a standardized, reproducible, and fully automated image analysis workflow designed to carry out this classification. This relies on a two part deep neural network architecture that works in tandem for segmentation of channel images and classification of adhered cells into subtypes. Network training utilized an extensive data set of images generated by the SCD BioChip, a microfluidic assay which injects clinical whole blood samples into protein-functionalized microchannels, mimicking physiological conditions in the microvasculature. Here we carried out the assay with the sub-endothelial protein laminin. The machine learning approach segmented the resulting channel images with 99.1±0.3% mean IoU on the validation set across 5k-folds, classified detected sRBCs with 96.0±0.3% mean accuracy on the validation set across 5k-folds, and matched trained personnel in overall characterization of whole channel images withR2= 0.992, 0.987 and 0.834 for total, deformable and non-deformable sRBC counts respectively. Average analysis time per channel image was also improved by two orders of magnitude (∼ 2 minutes vs ∼ 2-3 hours) over manual characterization. Finally, the network results show an order of magnitude less variance in counts on repeat trials than humans. This kind of standardization is a prerequisite for the viability of any diagnostic technology, making our system suitable for affordable and high throughput disease monitoring.
15 February 2022 in Lab on a Chip
Single cells have unique biophysical signatures that can rapidly change during various disease states. For instance, cellular density is an inherent property differing between cell types. Characterizing changes in fundamental density properties down to the single-cell level can reveal sub-populations in pathological states. Here, we have developed a microfluidic, magnetic levitation-based assay (MagDense) that detects minute density differences of individual red blood cells (RBCs) down to 0.0001 g mL−1 resolution. This assay fractionates RBCs based on their density profiles in a non-ionic paramagnetic medium flowing in a capillary microchannel placed between magnets with same poles facing each other. Based on precisely measured levitation height and density of individual RBCs at their specific equilibrium state, we demonstrated that MagDense can accurately analyze the density of sickle hemoglobin (HbS)-containing RBCs and normal hemoglobin (HbA)-containing RBCs. In addition, the precise density and cell size measurements at the single cell level showed three different sub-populations of RBCs in blood samples from individuals with homozygous sickle cell disease receiving blood transfusions; where less dense, HbA-containing RBCs levitated higher, while the denser, HbS-containing RBCs levitated lower. We compared the mean RBC densities of sickle cell disease subjects with healthy controls and found distinctly separated bands of RBC density for each group denoting the likely range of cell densities seen in the blood samples. The high resolution of our method enabled measurement of deviation from the mean RBC density. Moreover, we introduced a new term as a measure of density dispersion, “RBC levitational density width, RLDW”. Mean RBC density in sickle cell disease associated with hemoglobin from complete blood count (p = 0.032, linear regression) and RLDW associated with absolute reticulocyte count (ARC) and RBC distribution width (RDW) from complete blood count (p = 0.002 for ARC and p = 003 for RDW, linear regression). Our magnetic levitation-based assay enables rapid, accurate, density-based imaging, profiling and label-free monitoring of single RBCs. Our approach can be broadly applicable to investigate blood cell disorders and the effects of emerging pharmacological and curative therapies in patient outcomes.
29 May 2019 in IEEE Sensors Journal
This paper presents a turbine-based MEMS sensor for spirometry application. The compact sensor dimensions allow integration into smart wearable devices for mobile spirometry without a dedicated hand-held component. The sensor is composed of a turbine with embedded magnets, which is surrounded by two stators with micro coils in a PMMA package. Tangential gas flow actuates the turbine on ball bearings, and results in voltage generation due to electromagnetic induction. The sensor was initially characterized with pressurized nitrogen in the range of breathing flow rates. Sensitivity values up to 0.35 V/lpm were demonstrated. The device was then tested with actual breathing by a healthy subject. Sensor output voltages sampled under varying breath strengths ranged from 0.68 V to 2.94 V, corresponding to breathing flow rates of 3.5 lpm and 22.4 lpm, respectively. The device was also tested for power generation for a potential selfpowered operation. A maximum output power of 67 lW was achieved on a resistive load. Energy storage tests using a 1 mF capacitor showed that normal breathing can raise the capacitor voltage to 0.2 V, while a deep breath resulted in 0.6 V. The difference in capacitor voltage poses a potential alternative method for lung volume measurements. The results presented here demonstrate the potential of this sensor to perform spirometry tests in wearable electronic devices such as smart watches and smart bands.
7 March 2024 in The Analyst
Oxygen (O2) binds to hemoglobin (Hb) in the lungs and is then released (dissociated) in the tissues. The Bohr effect is a physiological mechanism that governs the affinity of Hb for O2 based on pH, where a lower pH results in a lower Hb-O2 affinity and higher Hb-O2 dissociation. Hb-O2 affinity and dissociation are crucial for maintaining aerobic metabolism in cells and tissues. Despite its vital role in human physiology, Hb-O2 dissociation measurement is underutilized in basic research and in clinical laboratories, primarily due to the technical complexity and limited throughput of existing methods. We present a rapid Hb-O2 dissociation measurement approach by leveraging the Bohr effect and detecting the optical shift in the Soret band that corresponds to the light absorption by the heme group in Hb. This new method reduces Hb-O2 dissociation measurement time from hours to minutes. We show that Hb deoxygenation can be accelerated chemically at the optimal pH of 6.9. We show that time and pH-controlled deoxygenation of Hb results in rapid and distinct conformational changes in its tertiary structure. These molecular conformational changes are manifested as significant, detectable shifts in Hb's optical absorption spectrum, particularly in the characteristic Soret band (414 nm). We extensively validated the method by testing human blood samples containing normal Hb and Hb variants. We show that rapid Hb-O2 dissociation can be used to screen for and detect Hb-O2 affinity disorders and to evaluate the function and efficacy of Hb-modifying therapies. The ubiquity of optical absorption spectrophotometers positions this approach as an accessible, rapid, and accurate Hb-O2 dissociation measurement method for basic research and clinical use. We anticipate this method's broad adoption will democratize the diagnosis and prognosis of Hb disorders, such as sickle cell disease. Further, this method has the potential to transform the research and development of new targeted and genome-editing-based therapies that aim to modify or improve Hb-O2 affinity.
8 October 2015 in IEEE Sensors Journal
The design, optimization, and performance characterization of turbine prototypes for the development of a Microelectromechanical Systems spirometer is reported. Four different turbines were designed based on large-scale Savonius turbine architecture. The number of turbine blades was optimized through finite-element simulations to maximize the induced moment and rotational speed at normal breathing flow rates. The turbines were manufactured and tested for their speed performance with respect to input flow rate, pressure difference, and actuation power, showing good agreement with the simulation results. The highest rotational speed was achieved with involute-bladed turbine design with eight blades, and was measured to be 10.56 kr/min at the peak 25-lpm flow rate. Real-life testing with this design was carried out on healthy subjects to demonstrate its capability of performing respiration flow rate measurements. The turbine prototypes presented in this paper allow for the development of low-cost and portable MEMS spirometers for remote- and self-monitoring of lung malfunctions in chronic obstructive pulmonary disease and asthma diseases.