Nanocytometer for smart analysis of peripheral blood and acute myeloid leukemia: a pilot study

Abstract

We realize an ultracompact nanocytometer for real-time impedimetric detection and classification of subpopulations of living cells. Nanoscopic nanowires in a microfluidic channel act as nanocapacitors and measure in real time the change of the amplitude and phase of the output voltage and, thus, the electrical properties of living cells. We perform the cell classification in the human peripheral blood (PBMC) and demonstrate for the first time the possibility to discriminate monocytes and subpopulations of lymphocytes in a label-free format. Further, we demonstrate that the PBMC of acute myeloid leukemia and healthy samples grant the label free identification of the disease. Using the algorithm based on machine learning, we generated specific data patterns to discriminate healthy donors and leukemia patients. Such a solution has the potential to improve the traditional diagnostics approaches with respect to the overall cost and time effort, in a label-free format, and restrictions of the complex data analysis.

StarPEG–heparin biosensors for rapid and portable diagnostics in complex biofluids

“…Our approach demonstrates an excellent detection capability without the need of nanomaterials or signal amplification techniques. Rather than compete with other nanoscopic transducers such as metallic 33 or semiconductor nanowires, 34 their synergy with the hydrogel technology could boost the results to subfemtomolar detection limits.…”

Deciphering impedance cytometry signals with neural networks

“…In contrast to other microfluidic fields (especially imaging flow cytometry), the application of machine learning to impedance cytometry is relatively unexplored. Recent studies have considered the use of machine learning tools to classify biological cells based on electrical features [14][15][16][17][18][19][20] (i.e., scalar parameters extracted from the raw impedance signals). For instance, Schütt et al 14 used a k-means algorithm for subpopulation clustering of peripheral blood mononuclear cells, based on peak voltage and phase.…”

“…Recent studies have considered the use of machine learning tools to classify biological cells based on electrical features [14][15][16][17][18][19][20] (i.e., scalar parameters extracted from the raw impedance signals). For instance, Schütt et al 14 used a k-means algorithm for subpopulation clustering of peripheral blood mononuclear cells, based on peak voltage and phase. Ahuja et al 16 used a support vector machine (SVM) classifier to discriminate between live and dead T47D breast cancer cells, using peak impedance magnitude and phase at four frequencies.…”

Nanochips assisted peptide screening for clinical development of CAR-T cell immunotherapy

“…Nanobioelectronics is an emerging field considered for future clinical applications, offering highly sensitive, label-free, rapid and reagent-saving analytical tools. As a part of this fieldnanobiosensors, have shown their efficiency in ultrasensitive detection of low concentrations of biomolecules [16][17][18][19][20][21][22][23][24][25][26] and single cells [27][28][29][30]. While the major focuses in the community are dedicated to the delivery and characterization of new efficient transducer materials for pointof-care biosensors [31][32][33][34], there is a limited number of reports with the clinical applications of such nanodevices [27,[35][36][37][38][39][40].…”

“…As a part of this fieldnanobiosensors, have shown their efficiency in ultrasensitive detection of low concentrations of biomolecules [16][17][18][19][20][21][22][23][24][25][26] and single cells [27][28][29][30]. While the major focuses in the community are dedicated to the delivery and characterization of new efficient transducer materials for pointof-care biosensors [31][32][33][34], there is a limited number of reports with the clinical applications of such nanodevices [27,[35][36][37][38][39][40]. We attribute it to the necessary complexity of the system, need for a long-term process monitoring, and thus, a stable performance of the sensor.…”