PSPranav Shrestha

Pranav Shrestha

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Pranav Shrestha has a total of 51 co-authors. Here are the top co-authors by number of shared publications:

  • Boris Stoeber

    Department of Mechanical Engineering, The University of British Columbia, Vancouver, Canada

    9 shared publications
  • Alaa Alzaki

    Faculty of Medicine, The University of British Columbia, Vancouver, Canada

    4 shared publications
  • Ali Amid

    Faculty of Medicine, The University of British Columbia, Vancouver, Canada

    4 shared publications
  • Heather McCartney

    Division of Hematology & Oncology, BC Children’s Hospital, Vancouver, Canada

    4 shared publications
  • Hendrik Lohse

    Department of Mechanical Engineering, The University of British Columbia, Vancouver, Canada

    4 shared publications
  • Navdeep Sandhu

    Adult Red Cell Disorders Program of BC and Yukon, St. Paul’s Hospital, Vancouver, Canada

    4 shared publications

Latest Publications

  • Open Access

    Author Correction: Morphology-based screening for sickle cell disease using low-cost automated microscopy and machine learning

    by Pranav Shrestha, Hendrik Lohse, Christopher Bhatla, Heather McCartney, Alaa Alzaki, Navdeep Sandhu, Pradip Kumar Oli, Hongquan Li, Manu Prakash, Ali Amid, et al.

  • Open Access

    Morphology-based screening for sickle cell disease using low-cost automated microscopy and machine learning

    by Pranav Shrestha, Hendrik Lohse, Christopher Bhatla, Heather McCartney, Alaa Alzaki, Navdeep Sandhu, Pradip Kumar Oli, Hongquan Li, Manu Prakash, Ali Amid, et al.

    Sickle cell disease (SCD) and β-thalassemia are the most common monogenic diseases, disproportionately affecting low- and middle-income countries, where low-cost and accurate diagnostic tools are needed to reduce the global disease burden. Although the sickling test is commonly used to screen for the sickle mutation, it cannot distinguish between the asymptomatic sickle cell trait (SCT) and SCD, nor identify β-thalassemia. Here, we enhanced the inexpensive sickling test using automated microscopy and morphology-based machine learning classification to detect SCD, trait conditions (SCT and β-thalassemia trait) and normal individuals with an overall area under the receiver operating characteristic (ROC) curve, sensitivity and specificity of 0.940 (95% confidence intervals: 0.938–0.942), 84.6% (84.2–84.9%), and 92.3% (92.1–92.4%), respectively. Notably, the sensitivity and specificity to detect severe disease (SCD) exceeded 97% and 98%, respectively, thus establishing a low-cost automated screening option for disease detection in low-resource settings. Analysis of 6–10 million red blood cells per participant across 40 morphological parameters in 138 individuals revealed that shape features (e.g., circularity and eccentricity) showed the strongest correlations with biochemical indices, including hemoglobin fractions measured via Hb HPLC. Furthermore, leveraging high-throughput microscopy, we generated an open-access dataset comprising over 300,000 images with 1.5 billion segmented cells from these 138 individuals in Canada and Nepal, including individuals with sickle and/or β-thalassemia mutations, to accelerate further research. ClinicalTrials.gov Identifier: NCT05506358; registered on August 17, 2022.

  • Open Access

    Evaluation of low-cost techniques to detect sickle cell disease and β-thalassemia: an open-label, international, multicentre study

    by Pranav Shrestha, Hendrik Lohse, Christopher Bhatla, Heather McCartney, Alaa Alzaki, Navdeep Sandhu, Pardip Kumar Oli, Sanjeev Chaudhary, Ali Amid, Rodrigo Onell, et al.


  • Open Access

    Octopi 2.0: Open and Scalable Microscopy Platform for AI-powered Diagnostic Applications

    by Hongquan Li, Heguang Lin, Pranav Shrestha, Rinni Bhansali, You Yan, Jaspreet Pannu, Kevin Marx, Wei Ouyang, Lucas Fuentes Valenzuela, Ethan Li, et al.

    Access to quantitative, robust, and affordable diagnostic tools is essential to address the global burden of infectious diseases. While manual microscopy remains a cornerstone of diagnostic workflows due to its broad adaptability, it is labor-intensive and prone to human error. Recent advances in artificial intelligence (AI) and robotics offer opportunities to automate and enhance microscopy, enabling high-throughput, multi-disease diagnostics with minimal reliance on complex supply chains. However, current automated microscopy platforms are often costly and inflexible — barriers that are especially limiting in low-resource settings. Here we present Octopi 2.0, an open, highly configurable, general-purpose automated microscopy platform for a broad range of diagnostic applications, including sickle cell anemia and antibiotic resistance that we have reported recently. Applying Octopi to imaging malaria parasites with 4’,6-diamidino-2-phenylindole (DAPI) staining, we discovered a spectral shift in fluorescence emission that allows rapid screening of blood smears at low magnification with throughput on the order of 1 million blood cells per minute. We further developed image processing and deep learning-based segmentation and classification pipelines to enable real-time processing for malaria diagnosis. For real-world performance validation, we collected a data set of 213 clinical samples from Uganda and the United States with a total of 905 million red blood cells and around 1.4 million malaria parasites. Using a ResNet-18 model and only one round of retraining, the model is able to achieve on average less than 5 false positive parasites/µL and a per-parasite level false negative rate of less than 8% in our test dataset. This per-cell performance implies a limit of detection (LoD) around 12 parasites/µL, and we measured patient-level performance of > 97% specificity and sensitivity in our independent test data set of clinical samples from 73 patients/donors. As more data is collected in larger validation studies, we expect the robustness and performance of the model to continue to improve according to what we observe in our proof-of-concept experiments carried out in this study. With significant cost reduction in hardware compared to current automated microscopes and an open and versatile approach for tackling multiple diseases with standard glass slide-based sample preparation, we envision Octopi 2.0 to help enable the “app store” for equitable data-driven, AI-powered diagnostics of many diseases and conditions.


  • Open Access

    Morphology-based classification of sickle cell disease and β-thalassemia using a low-cost automated microscope and machine learning

    by Pranav Shrestha, Hendrik Lohse, Christopher Bhatla, Heather McCartney, Alaa Alzaki, Navdeep Sandhu, Pradip Kumar Oli, Hongquan Li, Manu Prakash, Ali Amid, et al.

    Sickle cell disease (SCD) and β-thalassemia are the most common monogenic diseases, disproportionately affecting low- and middle-income countries, where low-cost and accurate diagnostic tools are needed to reduce the global disease burden. Although the sickling test is commonly used to screen for the sickle mutation, it cannot distinguish between the asymptomatic sickle cell trait (SCT) and SCD, or identify β-thalassemia. Here, we enhanced the inexpensive sickling test using automated microscopy and morphology-based machine learning classification to detect SCD, trait conditions (SCT and β-thalassemia trait) and normal individuals with an overall area under receiver operating curve, sensitivity and specificity of 0.940 (95% confidence intervals: 0.938-0.942), 84.6% (84.2%-84.9%), and 92.3% (92.1%-92.4%), respectively. Notably, the sensitivity and specificity to detect severe disease (SCD) was over 97% and 98%, respectively, thus establishing a low-cost automated screening option for disease detection in low-resource settings. Furthermore, leveraging high-throughput microscopy, we generated an open-access dataset comprising over 300,000 images with 1.5 trillion segmented cells from 138 individuals in Canada and Nepal including individuals with sickle and/or β-thalassemia mutations, to accelerate further research.


  • Force decomposition and toughness estimation from puncture experiments in soft solids

    by Pranav Shrestha, Curtis Geffner, Matthew Jaffey, Zhongnan Wu, Martina Iapichino, Mattia Bacca, Boris Stoeber

    Several medical applications, like drug delivery and biosensing, are critically preceded by the insertion of needles and microneedles into biological tissue. However, the mechanical process of needle insertions, especially at high velocities, is currently not fully understood. Here, we explore the insertion of hollow needles into transparent silicone samples with an insertion velocity v ranging from 0.1 mm s−1 to 2.3 m s−1 (with needle radius R = 101.5 μm, thus strain rates ∼v/R ranging from 1 s−1 to 2.3 × 104 s−1). We use a double-insertion method, where the needle is inserted and re-inserted at the same location, to estimate the fracture properties of the material. The deflection of the specimen's free surface is found to be different between insertion and re-insertion experiments for identical needle positions, which is associated with different force magnitudes between insertion/reinsertion. This aspect was previously neglected in the original double-insertion method, thus here we develop a method based on imaging, image analyses and force measurements to decompose the measured force into individual force components, including deflection force Fd, frictional and spreading force Ff + Fs, and cutting force Ft. We estimate that the toughness Γ of our silicone samples, calculated using the cutting force Ft and the crack dimensions, increases with needle velocity, and ranges within observed values in previous literature for the same material and for some soft biological materials. In addition to toughness Γ, other parameters, such as critical force Fc and mechanical work Wc, also show strain-rate dependence, suggesting tissue stiffening, due to accumulated strain energy, at high speeds.


  • Open Access

    Erratum: “Poroelastic behavior of skin tissue in response to pressure driven flow” [Phys. Fluids 35, 081902 (2023)]

    by Mary-Jo Weir Weiss, Pranav Shrestha, Raunak Basak, Boris Stoeber

    On page 10, Eq. (21) is incorrectly simplified into Eq. (22), making Eqs. (22) and (23) invalid. To correct this, m is instead calculated iteratively from Eq. ( On page 10, Eq. (21) is incorrectly simplified into Eq. (22), making Eqs. (22) and (23) invalid. To correct this, m is instead calculated iteratively from Eq. (

  • Open Access

    Poroelastic behavior of skin tissue in response to pressure driven flow

    by Mary-Jo Weir Weiss, Pranav Shrestha, Raunak Basak, Boris Stoeber

    Better understanding of skin tissue's permeability and poroelasticity could help advance biomedical technologies concerning skin such as intradermal injection and grafting. We applied a pressure driven fluid flow across skin tissue's epidermal and dermal layers in a simple one-dimensional configuration, while measuring the resulting flow rate and capturing cross sectional optical coherence tomography (OCT) images of the tissue's deformation. Digital image correlation was used to analyze the OCT images and via a novel method, image analysis corrected for the change in the tissue's refractive index, which occurred due to flow-induced deformation, thus providing accurate one-dimensional depth-wise deformation profiles. Skin tissue was found to exhibit a nonlinear relationship between pressure and the resulting fluid flow rate, where the increase in flow rate with pressure decreased as pressure increased. The skin tissue was observed to experience compressive strain closest to the supported base, with magnitudes increasing with increasing driving pressure, and the tissue near the free surface experienced relatively little strain. Permeability was found to follow an exponential permeability-volumetric strain relationship with material constants: k0 (initial uniform permeability) of 9.6 × 10−15 m2 and m (extent of nonlinearity for the permeability–strain relationship) of 2.94. Darcy's law and the permeability–strain relationship were used to analyze results with good similarity between observed and calculated flowrates. This work presents a novel and direct method of characterizing soft tissue permeability and provides a fundamental understanding to skin behavior under pressurized driving fluid, which can be generalized to study or model other geometries of induced flow through skin tissue.


  • Open Access

    Imaging fluid injections into soft biological tissue to extract permeability model parameters

    by Pranav Shrestha, Boris Stoeber

    One of the most common health care procedures is injecting fluids, in the form of drugs and vaccines, into our bodies, and hollow microneedles are emerging medical devices that deliver such fluids into the skin. Fluid injection into the skin through microneedles is advantageous because of improved patient compliance and the dose sparing effect for vaccines. Since skin tissue is a deformable porous medium, injecting fluid into the skin involves a coupled interaction between the injected fluid flow and the deformation of the soft porous matrix of skin tissue. Here, we introduce a semiempirical model that describes the fluid transport through skin tissue based on experimental data and constitutive equations of flow through biological tissue. Our model assumes that fluid flows radially outward and tissue deformation varies spherically from the microneedle tip. The permeability of tissue, assumed to be initially homogeneous, varies as a function of volumetric strain in the tissue based on a two-parameter exponential relationship. The model is optimized to extract two macroscopic parameters, k0 and m, for each of the seven experiments on excised porcine skin, using a radial form of Darcy’s law, the two-parameter exponential dependence of permeability on strain, and the experimental data on fluid flow recorded by a flow sensor and tissue deformation captured in real time using optical coherence tomography. The fluid flow estimated by the permeability model with optimized macroscopic parameters matches closely with the recorded flow rate, thus validating our semiempirical model.