Integrating deep learning with microfluidics for biophysical classification of sickle red blood cells

Praljak, Niksa; Iram, Shamreen; Goreke, Utku; Singh, Gundeep; Hill, Ailis; Gürkan, Umut A.; Hinczewski, Michael

doi:10.1101/2020.07.01.181545

Cited by 5 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the integration of machine learning and deep learning algorithms for single RBC microfluidic systems has been becoming popular for improved performances in areas such as structural designs, signal analyses, and operational schemes. Second, further integrations to include more components in the system bears fruits of improvements in terms of different performance evaluation parameters, overall effectiveness, accuracy, and commercial suitability [440][441][442]. Third, RBC-based drug delivery system is another interesting track, where the cargo drug can be stored in the inner space enclosed by the plasma membrane and the outer surface of this membrane can feature unique and favorable pharmacokinetic and biodistribution characteristics [443].…”

Section: Future Considerationsmentioning

confidence: 99%

Advances in Microfluidics for Single Red Blood Cell Analysis

et al. 2023

View full text Add to dashboard Cite

The utilizations of microfluidic chips for single RBC (red blood cell) studies have attracted great interests in recent years to filter, trap, analyze, and release single erythrocytes for various applications. Researchers in this field have highlighted the vast potential in developing micro devices for industrial and academia usages, including lab-on-a-chip and organ-on-a-chip systems. This article critically reviews the current state-of-the-art and recent advances of microfluidics for single RBC analyses, including integrated sensors and microfluidic platforms for microscopic/tomographic/spectroscopic single RBC analyses, trapping arrays (including bifurcating channels), dielectrophoretic and agglutination/aggregation studies, as well as clinical implications covering cancer, sepsis, prenatal, and Sickle Cell diseases. Microfluidics based RBC microarrays, sorting/counting and trapping techniques (including acoustic, dielectrophoretic, hydrodynamic, magnetic, and optical techniques) are also reviewed. Lastly, organs on chips, multi-organ chips, and drug discovery involving single RBC are described. The limitations and drawbacks of each technology are addressed and future prospects are discussed.

show abstract

Section: Future Considerationsmentioning

confidence: 99%

Advances in Microfluidics for Single Red Blood Cell Analysis

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Obtained results have represented similar or superior accuracies with improved overall analysis time by two orders of magnitude compared to human classification. [ 111 ]…”

Section: How Can Microfluidics Benefit From Ai?mentioning

confidence: 99%

Recent Advances of Utilizing Artificial Intelligence in Lab on a Chip for Diagnosis and Treatment

Harofte

Soltani

Siavashy

et al. 2022

Small

View full text Add to dashboard Cite

utilization of biological systems to develop new products at one end of the spectrum to the application of technology for solving the problems in biology at the other end. [1] According to the new experimental technologies like DNA nucleotide sequencing and RNA sequencing genomics, [2] Mass Spectrometry [3] as well as X-ray crystallography and Nuclear Magnetic Resonance spectroscopy [4] in proteomics, large datasets are generated and sorted in the field of biotechnology that make new opportunities for researchers along with companies that present services and products in this area. Nowadays, performing stateof-the-art research in biotechnology is roughly impossible without exploiting database and artificial intelligence (AI) technologies. [5] The integration of AI and biotechno logy is now prevalent. AI elucidates computational systems which extract signals and learn from the input. [1] Historically, AI has a close relationship with biology. Precisely, this relation is between statistics and genetics. Compared to text or image datasets, biological data is more multidimensional, complicated, and noisy. Hence, AI models are more pertinent than simple statistical models for biological datasets. [6] Nowadays, artificial intelligence (AI) creates numerous promising opportunities in the life sciences. AI methods can be significantly advantageous for analyzing the massive datasets provided by biotechnology systems for biological and biomedical applications. Microfluidics, with the developments in controlled reaction chambers, high-throughput arrays, and positioning systems, generate big data that is not necessarily analyzed successfully. Integrating AI and microfluidics can pave the way for both experimental and analytical throughputs in biotechnology research. Microfluidics enhances the experimental methods and reduces the cost and scale, while AI methods significantly improve the analysis of huge datasets obtained from high-throughput and multiplexed microfluidics. This review briefly presents a survey of the role of AI and microfluidics in biotechnology. Also, the incorporation of AI with microfluidics is comprehensively investigated. Specifically, recent studies that perform flow cytometry cell classification, cell isolation, and a combination of them by gaining from both AI methods and microfluidic techniques are covered. Despite all current challenges, various fields of biotechnology can be remarkably affected by the combination of AI and microfluidic technologies. Some of these fields include point-of-care systems, precision, personalized medicine, regenerative medicine, prognostics, diagnostics, and treatment of oncology and non-oncology-related diseases.

show abstract

“…Several research groups have explored image processing or machine learning techniques to study RBC morphologies [11] [12] [13] [14] [15] [16] [17] [18]. These techniques have focused on accurate classification of individual RBC shapes into normal (biconcave), sickle or abnormal (i.e.…”

Section: Mainmentioning

confidence: 99%

Differential sensitivity to hypoxia enables shape-based classification of sickle cell disease and trait blood samples

D’Costa

Sharma

Manna

et al. 2020

Preprint

View full text Add to dashboard Cite

In sickle cell anemia patients, red blood cells (RBCs) are known to become sickle shaped and stiff under hypoxic conditions as a consequence of hemoglobin polymerization. While RBC shape can discriminate sickle blood from healthy, it has not been used until now as a sole biophysical marker to differentiate between homozygous (disease) and heterozygous (trait) sickle blood samples. Here, we establish a technique based on the differential response of disease and trait RBCs to chemically-induced hypoxia to distinguish between these samples for the first time. By comparing the RBC shape distributions in blood treated with high and low concentrations of the hypoxia-inducing agent, we correctly identify 35 unknown blood samples as healthy, sickle cell disease or trait. Finally, we demonstrate our image-based classification technique with a portable smartphone microscope and a disposable microfluidic chip as a prospective point-of-care device enabling fast and confirmed diagnosis of sickle cell anemia.

show abstract

Integrating deep learning with microfluidics for biophysical classification of sickle red blood cells

Cited by 5 publications

References 28 publications

Advances in Microfluidics for Single Red Blood Cell Analysis

Advances in Microfluidics for Single Red Blood Cell Analysis

Recent Advances of Utilizing Artificial Intelligence in Lab on a Chip for Diagnosis and Treatment

Differential sensitivity to hypoxia enables shape-based classification of sickle cell disease and trait blood samples

Contact Info

Product

Resources

About