Combining microfluidics with machine learning algorithms for RBC classification in rare hereditary hemolytic anemia

Abstract

Combining microfluidics technology with machine learning represents an innovative approach to conduct massive quantitative cell behavior study and implement smart decision-making systems in support of clinical diagnostics. The spleen plays a key-role in rare hereditary hemolytic anemia (RHHA), being the organ responsible for the premature removal of defective red blood cells (RBCs). The goal is to adapt the physiological spleen filtering strategy for in vitro study and monitoring of blood diseases through RBCs shape analysis. Then, a microfluidic device mimicking the slits of the spleen red pulp area and video data analysis are combined for the characterization of RBCs in RHHA. This microfluidic unit is designed to evaluate RBC deformability by maintaining them fixed in planar orientation, allowing the visual inspection of RBC’s capacity to restore their original shape after crossing microconstrictions. Then, two cooperative learning approaches are used for the analysis: the majority voting scheme, in which the most voted label for all the cell images is the class assigned to the entire video; and the maximum sum of scores to decide the maximally scored class to assign. The proposed platform shows the capability to discriminate healthy controls and patients with an average efficiency of 91%, but also to distinguish between RHHA subtypes, with an efficiency of 82%.

Erysense, a Lab-on-a-Chip-Based Point-of-Care Device to Evaluate Red Blood Cell Flow Properties With Multiple Clinical Applications

“…Our approach simulates capillary flow in rectangular microfluidic channels with channel dimensions slightly larger than the RBC diameter. In contrast to methods that rely on larger geometries ( Piety et al, 2015 ; Alapan et al, 2016 ; Himbert et al, 2021 ), which result in a plethora of RBC shapes, or on constricted funnel-like channels ( Matthews et al, 2015 ; Myrand-Lapierre et al, 2015 ; Rizzuto et al, 2021 ), the technique presented here allows us to assess the flow characteristic of stable RBC shape configurations in a broad velocity range. For deformable RBCs, the geometric asymmetry of the cross-section of the used channels results in the emergence of two dominating morphologies (centered croissants and off-centered slippers) depending on the flow velocity.…”

“…Our approach differs from numerous microfluidic investigations of RBCs by the combination of three main properties: (i) we use in contrast to many other studies ( Iragorri et al, 2018 ; Rizzuto et al, 2021 ) straight channels without any constrictions; (ii) we use a micro-capillary size (8 μm × 11 µm diameter), whereas others use dimensions in the range of 30–50 µm ( D’Apolito et al, 2016 ; Inglebert et al, 2020 ); and (iii) we use flow velocities in the physiological range well below the flow velocity of other commercial devices, such as the AcCellerator by Zellmechanik Dresden (Germany), ( Reichel et al, 2019 ).…”

Microfluidics Approach to the Mechanical Properties of Red Blood Cell Membrane and Their Effect on Blood Rheology

“…A typical diagnosis for hemolytic anemias is made using a technique known as ektacytometry [188,189]. However, in recent years, microfluidic and lab-on-a-chip devices have presented new alternatives to study RBCs deformation using diverse techniques such as deformability cytometry [103,190], magnetic measurement [191], electrical measurement [192], single-cell chamber arrays [193], combination of microfluidics with machine learning [194] and pressure-driven microrheometry [195].…”

Machine Learning-Driven Multiobjective Optimization: An Opportunity of Microfluidic Platforms Applied in Cancer Research

“…When detecting the mechanical properties of a cell, the integration of microfluidics technologies and ML can investigate cell behavior on a large scale through quantitative analysis to provide intelligent decisions and support clinical diagnostics. In order to remove red blood cells in advance in rare hereditary hemolytic anemia, a microfluidic ML platform has been designed to generate the physiological filtering function of a spleen to monitor blood disease based on the analysis of blood cell shapes in vitro [ 42 ]. It evaluates the deformability of red blood cells by squeezing them in planar orientation, while visually observing and calculating the capacity of red blood cells to reinstate their pristine shape after penetrating through constrictions such as microchannels or nano-pores.…”

“…In this way, ML algorithms can distinguish abnormal red blood cells from normal ones. This microfluidic ML platform demonstrates the capability of recognizing and distinguishing between healthy controls and generic anemia patients, or rare hereditary hemolytic anemia subtypes with an average validity of 91% and 82%, respectively [ 42 ].…”