Label-free liquid biopsy - through the identification of tumor cells by machine learning-powered tomographic phase imaging flow cytometry

Aug 5, 2026

Label-free liquid biopsy for circulating tumor cells is addressed through a machine learning-powered tomographic phase imaging flow cytometry system that captures high-throughput 3D phase-contrast tomograms of single cells. The method discriminates tumor cells from white blood cells using hierarchical decision-making based on refractive-index features computed from 3D data. Proof-of-concept experiments use neuroblastoma and ovarian cancer cells versus monocytes, achieving over 97% success and accuracy for tumor cell identification and tumor-type recognition, enabling stain-free detection and classification in blood.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Label-free liquid biopsy and circulating tumor cells

Rationale: limitations of tissue biopsy

Liquid biopsy approaches and biomarkers

Focus on circulating tumor cells and label-free detection

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Label‑free liquid biopsy through the identification of tumor cells by machine learning‑powered tomographic phase imaging flow cytometry Daniele Pirone1, Annalaura Montella2,3, Daniele G. Sirico1, Martina Mugnano4, Massimiliano M. Villone4, Vittorio Bianco1, Lisa Miccio1, Anna Maria Porcelli 5,6,7, Ivana Kurelac7,8, Mario Capasso2,3, Achille Iolascon2,3, Pier Luca Maffettone4, Pasquale Memmolo1* & Pietro Ferraro1* Image‑based identification of circulating tumor cells in microfluidic cytometry condition is one of the most challenging perspectives in the Liquid Biopsy scenario. Here we show a machine learning‑ powered tomographic phase imaging flow cytometry system capable to provide high‑throughput 3D phase‑contrast tomograms of each single cell. In fact, we show that discrimination of tumor cells against white blood cells is potentially achievable with the aid of artificial intelligence in a label‑free flow‑cyto‑tomography method. We propose a hierarchical machine learning decision‑maker, working on a set of features calculated from the 3D tomograms of the cells’ refractive index. We prove that 3D morphological features are adequately distinctive to identify tumor cells versus the white blood cell background in the first stage and, moreover, in recognizing the tumor type at the second decision step. Proof‑of‑concept experiments are shown, in which two different tumor cell lines, namely neuroblastoma cancer cells and ovarian cancer cells, are used against monocytes. The reported results allow claiming the identification of tumor cells with a success rate higher than 97% and with an accuracy over 97% in discriminating between the two cancer cell types, thus opening in a near future the route to a new Liquid Biopsy tool for detecting and classifying circulating tumor cells in blood by stain‑free method. The early diagnosis of a tumor condition can be considered nowadays the holy grail in cancer research, as it is crucial for improving the efficacy of therapeutic treatments. To date, the golden standard method for cancer diagnosis is the tissue biopsy, that is an invasive test involving the extraction of the cancerous tissue to be further analyzed1. However, this procedure only reflects the situation in a single site of the tumor at a single time instant, thus limiting the understanding of the complex characterization of a patient’s tumor. Indeed, it has been demonstrated that various areas within the tumor can harbor different genomic profiles2. Moreover, it takes long and cumbersome protocols. In principle, since tumors shed parts of themselves into the circulatory system, it is possible to detect them by analyzing body fluids, such as blood and urine. This approach, known as Liquid Biopsy (LB)3–7, provides a relatively less invasive methodology for detecting disease-derived biomarkers from 1CNR-ISASI, Institute of Applied Sciences and Intelligent Systems “Eduardo Caianiello”, Via Campi Flegrei 34, 80078 Pozzuoli, Naples, Italy. 2CEINGE Advanced Biotechnologies, Naples, Italy. 3DMMBM, Department of Molecular Medicine and Medical Biotechnology, University of Naples “Federico II”, Naples, Italy. 4Department of Chemical, Materials and Production Engineering, DICMaPI, University of Naples “Federico II”, Piazzale Tecchio 80, 80125 Naples, Italy. 5Department of Pharmacy and Biotechnology (FABIT), University of Bologna, Bologna, Italy. 6Interdepartmental Centre for Industrial Research ‘Scienze Della Vita e Tecnologie per La Salute’, University of Bologna, Bologna, Italy. 7Centre for Applied Biomedical Research (CRBA), University of Bologna, Bologna, Italy. 8DIMEC, Department of Medical and Surgical Sciences, Centro di Studio e Ricerca Sulle Neoplasie (CSR) Ginecologiche, Alma Mater Studiorum-University of Bologna, 40138 Bologna, Italy. *email: pasquale.memmolo@ [isasi.cnr.it](isasi.cnr.it); [pietro.ferraro@cnr.it](pietro.ferraro@cnr.it) [www. nat

FAQs

What technology is proposed for label-free liquid biopsy in this work?

A machine learning-powered tomographic phase imaging flow cytometry system that produces high-throughput 3D phase-contrast tomograms for each single cell.

How does the method distinguish tumor cells from white blood cells?

It uses a hierarchical machine learning decision-maker that operates on features calculated from the cells’ 3D tomograms of refractive index.

What evidence is provided to validate tumor-cell identification?

Experiments compare neuroblastoma and ovarian cancer cell lines against monocytes, reporting success rates and accuracies higher than 97% for discriminating tumor cells and recognizing tumor type.

Aug 5, 2026

Upload by Olivia Brown