Novel optical technology detects severe neutropenia

Naveed Saleh, MD, MS, for MDLinx | August 09, 2018

Inventors unveiled a new optical-imaging technology that screens for chemotherapy-induced neutropenia in nailfold capillaries non-invasively, according to a recent study published in Scientific Reports. The novel technology is portable and fast.


MIT researchers have developed a way to measure white blood cell levels by imaging the cells as they flow through capillaries at the base of the fingernail. In these images, the white blood cells are marked by crosses. Image: MIT/The Leuko Project

“Our results demonstrate the relevance of using non-invasive videos of the microcirculation in nailfold capillaries,” wrote the authors, led by Aurélien Bourquard, PhD, Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA. “They also validate our overall classification strategy as a proof of principle towards the long-term objective of non-invasive WBC counting using a portable optical prototype and data analysis.”

Chemotherapy reduces white blood cell counts, which can lead to serious infection or even death. Currently, white blood cell counts are determined by blood sampling, an approach that limits accessibility and diagnostic frequency by requiring trained personnel. The researchers estimate that their take-at-home technology could decrease the number of chemotherapy-related infections by half.

In the current study, researchers tested whether their proof-of-concept, optical-imaging prototype could effectively screen for severe neutropenia (

The tabletop optical-imaging prototype records blood moving through miniscule capillaries at the level of the fingernail. White blood cells are about the width of the capillaries, whereas red blood cells are much smaller. By basking the capillaries in blue light, the team turned red blood cells dark and white blood cells transparent. They could then manually count optical absorption gaps, or “white gaps,” that punctuate the dark flow of red blood cells during a one-minute recording to assess severe neutropenia. These “count events” served as a substitute for the progression of white blood cells through these tiny fingernail capillaries.

“Moving forward, the event rating could be performed automatically on the segmented capillary videos,” the authors wrote. “Such an algorithm could follow approaches proposed to detect objects moving through capillaries or more advanced strategies such as machine learning.”

During this pilot diagnostic validation study, the team employed their prototype to obtain 22 microcirculatory-video datasets representing 11 patients. These patients were being administered high-dose chemotherapy and autologous stem-cell transplantation (ASCT). The team obtained capillary videos before chemotherapy at baseline (>1,500 neutrophils/μL) and after chemotherapy at severe neutropenia. The videos were produced within 8 hours of blood testing to supply reference data.

Of interest, the inventors tracked the same capillaries during both baseline and follow-up measures to circumvent bias and limit confounding, including discrepancies in capillary volume. Although possible, it remains to be elucidated whether comparable results could be secured from the use of unpaired capillaries.

On the basis of their video datasets, the authors found that counting white gaps in capillaries reliably distinguished severe neutropenia from non-neutropenia. Furthermore, the event counts tabulated during this study were consistent with WBC reference values assessed during subsequent blood sampling.

“Our system proved to be 95% accurate in determining whether an individual’s white cell levels were reduced to dangerous levels,” said Carlos Castro-Gonzalez, PhD, one of the authors of the study. “This was achieved using the counts of our human observers; preliminary results indicate that the automated machine-vision counting system we are developing can improve this level of accuracy.”

For now, this technology is only in its early stages. “More patients, time points, and ranges will be needed to define the range of detection of differences in cell concentration, before such a device can be deployed,” the authors wrote.

Looking forward, the investigators suggested that the capabilities of this optical-image prototype could be expanded to screen for WBC ranges other than severe neutropenia. Extending this technology’s diagnostic range could broaden its clinical applicability to other conditions such as infection or leukostasis.

“The automation, replication, and refinement of these results may lead to a new paradigm in the monitoring of cancer patients at risk of severe neutropenia,” concluded the authors. “Furthermore, from a more general standpoint, the proposed imaging technique and conceptual approach could constitute one first step towards non-invasive, in-vivo WBC counting.”

This study was funded in part by the National Institute of Biomedical Imaging and Bioengineering (NIBIB).